A Survey on Object Detection in Optical Remote Sensing Images

Gong ChengJunwei Han

article2016arXiv1,360 citations

Systematizes approximately 270 generic object detection studies across aerial and satellite imagery into four core methodological paradigms while reviewing standard benchmarks, evaluation metrics, and future opportunities in deep and weakly supervised learning.

Listen

Automatic object detection in optical remote sensing images is critical for practical applications such as urban planning, precision agriculture, environmental monitoring, disaster response, and updating geographic databases. With the rapid growth of high-resolution satellite and aerial imagery, manual identification has become impractical. At the same time, automated detection faces major hurdles due to complex backgrounds, clutter, shadows, varying viewpoints, and wide variations in object appearances.

The main objective of the article is to provide a comprehensive survey of generic object detection methods across diverse categories, rather than focusing on a single target type like roads or buildings. To accomplish this, the article evaluates approximately 270 research publications, systematically organizing the technical landscape into four primary methodological frameworks, reviewing five public benchmark datasets, and outlining standard evaluation metrics.

The review identifies key trade-offs across current approaches. Early template matching methods are simple but struggle with appearance variations, though deformable templates offer more flexibility at higher computational cost. Knowledge-based methods rely on geometric and contextual rules, which often lack robustness if rules are defined too strictly or loosely. Object-based image analysis successfully groups homogeneous pixels to classify land cover, but setting automated segmentation scales remains challenging. Machine learning approaches achieve higher detection accuracy by extracting features and training statistical classifiers; however, most deployed systems still depend heavily on handcrafted visual descriptors and extensive manual annotations.

These findings indicate that existing operational workflows face significant cost and scaling bottlenecks due to the labor-intensive requirement for detailed bounding-box labeling. Shifting toward modern feature representation and reduced human intervention is essential to handle large data streams efficiently and reduce the risk of detector failure across complex real-world environments.

The article highlights two actionable research directions to build more robust systems: adopting deep learning architectures to extract high-level feature representations directly from imagery, and developing weakly supervised learning frameworks that only require image-level presence labels rather than full bounding annotations. Organizations developing remote sensing pipelines should prioritize deep learning for improved detection power while investing in weakly supervised algorithms capable of detecting multiple object classes simultaneously.

Decision-makers must note key current limitations: deep neural networks require massive training datasets to prevent overfitting and carry high computational costs during real-time feature extraction. Additionally, weakly supervised methods in remote sensing are still in early development, with performance currently trailing supervised alternatives. Operational deployments should therefore balance deep learning accuracy against available computing resources and labeling budgets.

Northwestern Polytechnical University deep learning detection algorithm geo-spatial optical remote sensing images object detection learning-based object detection methods

arXiv: 1603.06201

Paper: Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks, Shaoqing Ren et al. (2015). Introduces Faster R-CNN and region proposal networks, establishing the foundational modern machine learning detection pipeline reviewed and adapted to remote sensing imagery in the survey.
Paper: Rich feature hierarchies for accurate object detection and semantic segmentation, Ross Girshick et al. (2014). Pioneers the application of deep convolutional neural networks to region-based object detection, setting the baseline for deep learning methods analyzed in the survey.
Paper: Object Detection with Discriminatively Trained Part-Based Models, Pedro F. Felzenszwalb et al. (2010). Establishes the deformable part-based model (DPM) framework using HOG and latent SVMs, providing essential context for the traditional machine learning approaches cataloged in the survey.
Paper: The Pascal Visual Object Classes Challenge: A Retrospective, Mark Everingham et al. (2014). Details the standard evaluation protocols and benchmarks like PASCAL VOC that define the evaluation metrics adopted and discussed in the survey.
Paper: OverFeat: Integrated Recognition, Localization and Detection using Convolutional Networks, Pierre Sermanet et al. (2014). Presents an early integrated framework for joint localization and detection via sliding-window convolutional networks, bridging traditional scanning approaches with deep learning.
Paper: Beyond Bags of Features: Spatial Pyramid Matching for Recognizing Natural Scene Categories, Svetlana Lazebnik et al. (2006). Develops spatial pyramid matching to capture spatial structure from bag-of-visual-words, a key classical feature representation reviewed in the survey.
Paper: Visual categorization with bags of keypoints, Gabriella Csurka et al. (2004). Introduces the bag-of-keypoints visual representation for categorization, serving as a core foundation for early machine learning and template matching detectors.
Paper: A Trainable System for Object Detection, CONSTANTINE PAPAGEORGIOU et al. (2000). Demonstrates early trainable sliding-window detection using multiscale wavelets and support vector machines, underpinning the machine learning category of object detection methods.

Paper: Object Detection in Optical Remote Sensing Images: A Survey and A New Benchmark, Ke Li et al. (2019). Directly succeeds the 2016 survey by introducing the large-scale DIOR benchmark and evaluating twelve modern deep learning detectors across 20 remote sensing categories.
Paper: DOTA: A Large-Scale Dataset for Object Detection in Aerial Images, Gui-Song Xia et al. (2017). Addresses the survey's call for large-scale benchmarks by introducing DOTA and standardizing oriented bounding box detection in aerial imagery.
Paper: Deep learning in remote sensing: a review, Xiao Xiang Zhu et al. (2017). Broadens the survey's discussion of deep learning in optical imagery to multi-sensor Earth observation modalities including SAR and hyperspectral analysis.
Paper: Deep Learning for Generic Object Detection: A Survey, Li Liu et al. (2018). Provides a comprehensive follow-up survey tracing the subsequent rapid evolution of deep learning detector architectures and multi-scale feature hierarchies.
Paper: TPH-YOLOv5: Improved YOLOv5 Based on Transformer Prediction Head for Object Detection on Drone-captured Scenarios, Xingkui Zhu et al. (2021). Applies transformer-based prediction heads within YOLOv5 to address specific remote sensing challenges highlighted in the survey, such as tiny and occluded targets.
Paper: Remote Sensing Image Scene Classification: Benchmark and State of the Art, Gong Cheng et al. (2017). Complements optical object detection by establishing NWPU-RESISC45, a large-scale benchmark for aerial scene classification evaluating deep feature learning.
Paper: AID: A Benchmark Data Set for Performance Evaluation of Aerial Scene Classification, Gui-Song Xia et al. (2016). Supplies a large-scale aerial dataset (AID) to systematically evaluate deep learning baselines on overhead imagery.
Paper: Object Detection in 20 Years: A Survey, Zhengxia Zou et al. (2019). Provides a broad retrospective survey tracking the progression from early handcrafted methods to modern deep learning and vision transformer architectures.

Cover for A Survey on Object Detection in Optical Remote Sensing Images

Abstract

Object detection in optical remote sensing images, being a fundamental but challenging problem in the field of aerial and satellite image analysis, plays an important role for a wide range of applications and is receiving significant attention in recent years. While enormous methods exist, a deep review of the literature concerning generic object detection is still lacking. This paper aims to provide a review of the recent progress in this field. Different from several previously published surveys that focus on a specific object class such as building and road, we concentrate on more generic object categories including, but are not limited to, road, building, tree, vehicle, ship, airport, urban-area. Covering about 270 publications we survey 1) template matching-based object detection methods, 2) knowledge-based object detection methods, 3) object-based image analysis (OBIA)-based object detection methods, 4) machine learning-based object detection methods, and 5) five publicly available datasets and three standard evaluation metrics. We also discuss the challenges of current studies and propose two promising research directions, namely deep learning-based feature representation and weakly supervised learning-based geospatial object detection. It is our hope that this survey will be beneficial for the researchers to have better understanding of this research field.

Topics

deep learning detection algorithm geo-spatial object detection method remote sensing images template matching object detection algorithms object detection systems high resolution image

Keywords

optical remote sensing images object detection learning-based object detection methods template matching-based object detection hyperspectral target detection spatial pyramid matching aircraft recognition airport detection vehicle detection

1. Introduction
2. Taxonomy of methods for object detection
3. Template matching-based object detection
3.1 Rigid template matching
3.2 Deformable template matching
3.2.1 Free-form deformable templates
3.2.2 Parametric deformable templates
4. Knowledge-based object detection
4.1 Geometric knowledge
4.2 Context knowledge
5. OBIA-based object detection
6. Machine learning-based object detection
6.1 Feature extraction
6.1.1 Histogram of oriented gradients (HOG) feature and its extensions
6.1.2 Bag-of-words (BoW) feature
6.1.3 Texture features
6.1.4 Sparse representation (SR)-based features
6.1.5 Haar-like features
6.2 Feature fusion and dimension reduction
6.2.1 Feature fusion
6.2.2 Dimension reduction
6.3 Classifier training
6.3.1 Support vector machine (SVM)
6.3.2 AdaBoost
6.3.3 k-nearest-neighbor (kNN)
6.3.4 Conditional random field (CRF)
6.3.5 Sparse representation-based classification (SRC)
6.3.6 Artificial neural network (ANN)
6.3.7 Other classifiers
7. Datasets and evaluation metrics
7.1 Datasets
7.2 Evaluation metrics
8. Promising research directions
8.1 Deep learning-based feature representation
8.2 Weakly supervised learning-based geospatial object detection
9. Conclusion
Acknowledgements
References

Knowls

Knowl 1 — Taxonomy of Object Detection Methods in Optical Remote Sensing Images
model/method
Object detection in optical remote sensing images (RSIs) encompasses both discrete man-made objects with well-defined boundaries (e.g., vehicles, aircraft, ships, buildings) and continuous landscape/land-cover parcels with diffuse boundaries (e.g., urban areas, forest tracts). Methods in the literature are structured into four primary methodological paradigms:
1. Template Matching-Based Methods: Identify objects by scoring the similarity between target image regions and a reference template. Subdivided into:
  - Rigid Template Matching: Employs fixed geometric/radiometric profiles or morphological structuring elements.
  - Deformable Template Matching: Incorporates flexible geometric transformations, categorized into free-form deformable templates (e.g., active contours/snakes) and parametric deformable templates (parameterized shape prototypes and principal component deformation modes).
2. Knowledge-Based Methods: Formulate object detection as a hypothesis generation and verification problem guided by explicit domain rules:
  - Geometric Knowledge: Shape constraints such as rectilinear outlines, aspect ratios, or height profiles from digital elevation models.
  - Context Knowledge: Spatial relationships and interactions with surrounding scenery, notably shadow-casting geometry for elevation and boundary verification.
3. Object-Based Image Analysis (OBIA / GEOBIA) Methods: A two-stage paradigm tailored to very high resolution (VHR) imagery:
  - Image Segmentation: Aggregating spatially contiguous, homogeneous pixels into image segments (objects) using scale, shape, and compactness criteria.
  - Object Classification: Classifying segmented regions using spatial, spectral, geometric, contextual, and GIS-derived features.
4. Machine Learning-Based Methods: Frame detection as a region-level classification task comprising three principal modules:
  - Feature Extraction: Mapping image regions to discriminative representation spaces (e.g., HOG, BoW, texture filters, sparse representations, Haar-like features).
  - Feature Fusion and Dimension Reduction (Optional): Linearly or non-linearly combining complementary descriptors and reducing dimensionality (e.g., PCA, LDA, PLS).
  - Classifier Training: Supervised, semi-supervised, or weakly supervised learning algorithms (e.g., SVM, AdaBoost, $k$ NN, CRF, SRC, ANN/CNN) to separate objects from background.
Knowl 2 — Template Matching Frameworks in Remote Sensing
model/method
Template matching operates through two successive stages: (1) Template Generation, in which an object prototype $T$ is handcrafted or learned from training data, and (2) Similarity Measurement, where $T$ is slid across candidate image coordinates across variations in translation, rotation, and scale to optimize a similarity or distance criterion.

Standard similarity metrics include Sum of Absolute Differences (SAD), Sum of Squared Differences (SSD), Normalized Cross-Correlation (NCC), and Euclidean Distance (ED).

Rigid Template Matching

Rigid template matching applies fixed spatial and radiometric profiles or morphological operators (such as the Morphological Hit-or-Miss Transform, HMT, across binary, grayscale, or multispectral bands) to identify structures with low intra-class variation (e.g., standard road tracks, oil storage tanks). Its primary limitation is extreme sensitivity to viewpoint variations, scale changes, occlusions, and irregular geometric deformations.

Deformable Template Matching

Deformable template matching introduces deformation degrees of freedom:
- Free-Form Deformable Templates (Active Contours / Snakes): An energy-minimizing contour is iteratively deformed by internal forces (enforcing smoothness, continuity, and elastic restitution) and external forces (attracted to image gradients, edges, or potential fields) to segment target boundaries.
- Parametric Deformable Templates: Objects are modeled via parametric mathematical equations or prototype models with deformable parameters learned via techniques such as Principal Component Analysis (PCA) combined with kernel density estimation, balancing global shape prior constraints with local image evidence.
Knowl 3 — Knowledge-Based Hypothesis Generation and Context Verification
model/method
Knowledge-based object detection formulates extraction as a rule-driven hypothesis generation and testing pipeline across a coarse-to-fine hierarchical structure:
1. Hypothesis Generation via Geometric Rules: Domain-specific geometric models are enforced to hypothesize candidate target locations. For instance, buildings are modeled as rectilinear combinations of rectangular primitives ("box", "L", "T", or "E" shapes), while roads are modeled by parallel boundary margins and spectral continuity.
2. Hypothesis Verification via Context and Shadow Knowledge: Hypothesized regions are validated or rejected by checking spatial context relations. In monocular optical RSIs, solar illumination geometry dictates cast shadow positions. Directional spatial relationships between elevated structures (e.g., rooftops) and their adjacent cast shadows serve as unambiguous cues to eliminate false alarms caused by spectrally similar ground-level clutter.
Operational Trade-off: The accuracy of knowledge-based systems depends critically on rule thresholding: overly restrictive rules yield high false negative rates (missed targets), whereas overly permissive rules yield high false positive rates.
Knowl 4 — Object-Based Image Analysis (OBIA / GEOBIA) Paradigm
model/method
Object-Based Image Analysis (OBIA) replaces single pixels with homogeneous pixel clusters (segments or image objects) as the fundamental processing unit for classification in very high resolution (VHR) remote sensing data.

Segmentation and Multi-Resolution Segmentation (MRS)

The primary segmentation engine in OBIA is Multi-Resolution Segmentation (MRS), a bottom-up region-merging technique controlled by three parameters:
- Scale Parameter: Determines the maximum permissible internal heterogeneity of the resulting segments, directly governing the average spatial size of the extracted objects.
- Shape Parameter: Specifies the relative weight assigned to geometric shape homogeneity versus spectral homogeneity during adjacent segment merging.
- Compactness Parameter: A sub-parameter of shape that balances segment compactness against perimeter smoothness.
Objective Scale Optimization

To mitigate subjective trial-and-error selection of scale parameters, objective metrics analyze Local Variance (LV) graphs of image layers as a function of object resolution. Tools such as Estimation of Scale Parameters (ESP) identify thresholds where the rate of local variance change indicates optimal semantic object boundaries.

Segment Classification and Assessment

Once segmented, objects are characterized by multi-source feature vectors—combining spectral statistics, morphology (area, elongation, rectangularity), texture, topology, and GIS spatial contextual layers—and classified using rule-based membership functions, decision trees, nearest neighbors, or support vector machines.
Knowl 5 — Machine Learning-Based Object Representation Features for RSIs
model/method
Machine learning detection pipelines rely on discriminative visual feature extraction from candidate windows or spatial proposals. Five feature representations predominate in optical RSI analysis:
1. Histogram of Oriented Gradients (HOG) & Extensions: Accumulates local 1D gradient orientation histograms across spatial cells, normalized over overlapping blocks via $\ell_2$ -norm ( $\mathbf{v} \leftarrow \mathbf{v} / \sqrt{\|\mathbf{v}\|_2^2 + \xi^2}$ ) or $\ell_1$ -norm ( $\mathbf{v} \leftarrow \mathbf{v} / (\|\mathbf{v}\|_1 + \xi)$ ). In RSIs, rotation invariance is achieved by orienting blocks along dominant SIFT-like gradients, circle-frequency transforms, or polar angle normalization.
2. Bag-of-Words (BoW) & Spatial Pyramid Matching (SPM): Extracts local interest points (e.g., Harris-Laplacian, Difference-of-Gaussians), computes local descriptors (SIFT), clusters descriptors via $k$ -means to form a visual vocabulary, quantizes keypoints into visual words, and pools them into histograms over multi-level spatial sub-regions to preserve layout.
3. Texture Descriptors:
  - Gabor Filter Banks: Convolves images with pairs of symmetric even and anti-symmetric odd 2D Gabor wavelets across selective scales $\sigma$ and orientations $\theta = \arctan(k_y / k_x)$ : $G_{\text{even}}(x,y) = \cos(k_x x + k_y y) \exp\left(-\frac{x^2 + y^2}{2\sigma^2}\right)$ $G_{\text{odd}}(x,y) = \sin(k_x x + k_y y) \exp\left(-\frac{x^2 + y^2}{2\sigma^2}\right)$
  - Local Binary Patterns (LBP): Quantizes local neighborhood pixel intensities against a center pixel $p_c$ over $N$ circularly symmetric neighbors at radius $r$ : $\text{LBP}_{N,r}(p_c) = \sum_{i=1}^N 2^{i-1} \phi(I(p_i) - I(p_c))$ where $\phi(\Delta) = 1$ if $\Delta \ge 0$ and $0$ otherwise.
4. Sparse Representation (SR) Features: Encodes high-dimensional visual signals into sparse linear combinations of atoms from an over-complete dictionary.
5. Haar-like Features: Evaluates differences between summed pixel intensities of adjacent rectangular regions normalized by local standard deviation, computed rapidly via integral images.
Knowl 6 — Sparse Representation and Sparse Representation-Based Classification (SRC)
equation
In sparse representation (SR) feature extraction, a signal $\mathbf{x} \in \mathbb{R}^d$ is approximated over an over-complete dictionary $\mathbf{D} = [\mathbf{d}_1, \mathbf{d}_2, \dots, \mathbf{d}_M] \in \mathbb{R}^{d \times M}$ ( $M \gg d$ ) by solving the $\ell_1$ -regularized optimization problem:

$\boldsymbol{\alpha}^* = \arg\min_{\boldsymbol{\alpha}} \left\{ \|\mathbf{x} - \mathbf{D}\boldsymbol{\alpha}\|_2^2 + \lambda \|\boldsymbol{\alpha}\|_1 \right\}$

where $\boldsymbol{\alpha} \in \mathbb{R}^M$ is the sparse coefficient vector, $\|\boldsymbol{\alpha}\|_1 = \sum_{i=1}^M |\alpha_i|$ , and $\lambda > 0$ is a regularization parameter balancing reconstruction fidelity and sparsity.

In Sparse Representation-Based Classification (SRC), the dictionary is constructed directly by concatenating labeled training samples from $C$ distinct object classes: $\mathbf{A} = [\mathbf{A}_1, \mathbf{A}_2, \dots, \mathbf{A}_C] \in \mathbb{R}^{d \times \sum m_i}$ , where $\mathbf{A}_i \in \mathbb{R}^{d \times m_i}$ contains the training samples of class $i$ .

Given a test sample $\mathbf{x} \in \mathbb{R}^d$ , it is sparsely coded over the global dictionary $\mathbf{A}$ :

$\hat{\boldsymbol{\alpha}} = \arg\min_{\boldsymbol{\alpha}} \left\{ \|\mathbf{x} - \mathbf{A}\boldsymbol{\alpha}\|_2^2 + \lambda \|\boldsymbol{\alpha}\|_1 \right\}$

The class-specific residual $r_i(\mathbf{x})$ is computed using only the subset of sparse coefficients $\hat{\boldsymbol{\alpha}}_i$ corresponding to class $i$ :

$r_i(\mathbf{x}) = \|\mathbf{x} - \mathbf{A}_i \hat{\boldsymbol{\alpha}}_i\|_2, \quad i = 1, 2, \dots, C$

The test sample $\mathbf{x}$ is assigned to the class with the minimal reconstruction residual:

$\text{class}(\mathbf{x}) = \arg\min_{i \in \{1, \dots, C\}} r_i(\mathbf{x})$
Knowl 7 — Conditional Random Field (CRF) Formulation for Contextual Object Detection
equation
For an image represented as a graph $G=(S, E)$ with nodes $S$ (corresponding to image regions or pixels) and edges $E$ (connecting neighboring nodes), let $\mathbf{x} = \{\mathbf{x}_i\}_{i \in S}$ denote observed image features and $\mathbf{y} = \{y_i\}_{i \in S}$ denote class labels (with $y_i \in \{-1, +1\}$ in binary detection). The conditional posterior probability distribution $P(\mathbf{y} \mid \mathbf{x})$ is modeled in a discriminative framework as:

$P(\mathbf{y} \mid \mathbf{x}) = \frac{1}{Z} \exp\left( \sum_{i \in S} A_i(y_i, \mathbf{x}) + \sum_{i \in S} \sum_{j \in N_i} I_{ij}(y_i, y_j, \mathbf{x}) \right)$

where $Z$ is the partition function normalizing the distribution, $N_i$ denotes the set of spatial neighbors of node $i$ , $A_i(y_i, \mathbf{x})$ is the unary association potential, and $I_{ij}(y_i, y_j, \mathbf{x})$ is the pairwise interaction potential.

The association potential measures the likelihood that node $i$ takes label $y_i$ given feature representation $\mathbf{h}_i(\mathbf{x})$ and trained parameter vector $\mathbf{w}$ :

$A_i(y_i, \mathbf{x}) = \exp\left( y_i \mathbf{w}^T \mathbf{h}_i(\mathbf{x}) \right)$

The interaction potential models contextual dependencies between adjacent node labels $y_i$ and $y_j$ based on pairwise relational features $\boldsymbol{\mu}_{ij}(\mathbf{x})$ (constructed by subtracting or concatenating local node descriptors) and trained parameter vector $\mathbf{v}$ :

$I_{ij}(y_i, y_j, \mathbf{x}) = \exp\left( y_i y_j \mathbf{v}^T \boldsymbol{\mu}_{ij}(\mathbf{x}) \right)$

Knowl 8 — Public Benchmark Datasets for Remote Sensing Object Detection

data/table

Five standardized public benchmarking datasets are established for evaluating and comparing optical RSI object detection algorithms:

Dataset	Target Classes	Image Count	Resolution / GSD	Source	Ground Truth Annotations
NWPU VHR-10	10	800	0.08 m – 2.0 m	Google Earth (715 RGB) Vaihingen (85 CIR)	3,772 axis-aligned bounding boxes across 10 classes (airplane, ship, storage tank, baseball diamond, tennis court, basketball court, ground track field, harbor, bridge, vehicle).
SZTAKI-INRIA	1	9	VHR satellite/aerial	Google Earth, IKONOS, QuickBird (aerial: Budapest, Szada; satellite: Manchester, Bodensee, Normandy, Cote d'Azur)	665 building footprints manually delineated as oriented rectangular footprints (RGB only).
TAS Aerial Car	1	30	High resolution	Google Earth ( $792 \times 636$ pixels)	1,319 vehicle bounding boxes (average car window approx. $45 \times 45$ pixels).
OIRDS	1	$\sim 900$	0.0838 m – 0.3048 m	Aircraft-mounted cameras	$\sim 1,800$ annotated vehicle instances with ground sample distance metadata.
IITM Road	1	200	1.0 m	Wikimapia satellite imagery ( $512 \times 512$ pixels)	Centerline and binary ground truth road maps divided across developed (100) and emerging (100) country categories.

These datasets cover both single-class detection benchmarks (buildings, cars, roads) and multi-class generic geospatial targets (NWPU VHR-10), enabling quantitative comparisons under varying ground sample distances and imaging conditions.

Knowl 9 — Standard Quantitative Evaluation Metrics for Geospatial Object Detection
definition
Evaluation of geospatial object detectors relies on three standardized performance metrics based on the counts of True Positives ( $TP$ ), False Positives ( $FP$ ), and False Negatives ( $FN$ ):

Precision and Recall

$\text{Precision} = \frac{TP}{TP + FP}, \quad \text{Recall} = \frac{TP}{TP + FN}$

Pixel-Level vs. Object-Level Evaluation Criteria
- Pixel-level: Evaluates label agreement pixel-by-pixel against reference ground truth masks.
- Object-level: A predicted bounding box or polygon is designated as a True Positive ( $TP$ ) if its Intersection-over-Union (IoU) overlap ratio $a_o$ with a ground truth instance exceeds a predefined threshold $\lambda$ (commonly $\lambda = 0.5$ ): $a_o = \frac{\text{area}(\text{detection} \cap \text{ground\_truth})}{\text{area}(\text{detection} \cup \text{ground\_truth})} > \lambda$ If multiple predicted bounding boxes overlap a single ground truth object with $a_o > \lambda$ , only one detection is scored as a True Positive, and all remaining overlapping detections are marked as False Positives ( $FP$ ).
F-Measure ( $F_\beta$ )

The weighted harmonic mean combining Precision and Recall: $F_\beta = \frac{(1 + \beta^2) \cdot \text{Precision} \cdot \text{Recall}}{\beta^2 \cdot \text{Precision} + \text{Recall}}$ Setting $\beta = 1$ gives equal weighting ( $F_1$ -score), while $\beta < 1$ prioritizes Precision over Recall.

Average Precision (AP)

Average Precision computes the area under the Precision-Recall Curve (PRC) across the full recall interval $[0, 1]$ , providing an aggregate metric of detection quality independent of specific operational thresholds.
Knowl 10 — Frontier Challenges: Deep Feature Learning and Weakly Supervised Learning in RSIs
theoretical result
Two critical paradigms address key limitations of traditional shallow/handcrafted detection pipelines in optical remote sensing:

1. Deep Learning-Based Feature Representation

Deep convolutional architectures automatically learn hierarchical feature representations directly from raw pixels, capturing high-level semantic abstractions that surpass handcrafted descriptors (HOG, SIFT, LBP) in complex backgrounds. Key open challenges in remote sensing include:
- Data Dependency: Deep networks require large-scale diverse datasets to avoid overfitting, whereas domain-specific annotated remote sensing benchmarks remain limited.
- Inference Complexity: High computational latency during full-scene feature extraction and window scanning limits real-time satellite and aerial surveillance.
2. Weakly Supervised Learning (WSL) for Geospatial Detection

Supervised object detectors depend on exhaustive, labor-intensive pixel-level or bounding-box annotations. Weakly Supervised Learning trains detectors using only image-level binary tags (indicating category presence/absence per scene) without bounding boxes. Key open challenges in remote sensing include:
- Multi-Class Co-occurrence: Existing RSI WSL frameworks predominantly tackle isolated single-class problems, whereas wide-swath RSIs typically contain multiple co-occurring object categories in complex cluttered environments.
- Localization Ambiguity: Distinguishing target objects from background clutter remains less accurate under weak supervision compared to fully supervised detectors, requiring advanced negative bootstrapping and saliency-guided spatial localization.

Coverage note — None was omitted; all major survey contributions—the comprehensive 4-part taxonomy, template matching methods, knowledge-based paradigms, OBIA workflows, feature representations and classifier formulations, benchmark datasets, quantitative evaluation metrics, and open research challenges—have been fully captured.

References

1.Ahmadi, S., Zoej, M.V., Ebadi, H., Moghaddam, H.A., Mohammadzadeh, A., 2010. Automatic urban building boundary extraction from high resolution aerial images using an innovative model of active contours. Int. J. Appl. Earth Observ. Geoinform. 12, 150-157.
2.Akçay, H.G., Aksoy, S., 2010. Building detection using directional spatial constraints. In: Proc. IEEE Int. Geosci. Remote Sens. Symp., pp. 1932-1935.
3.Albrecht, F., 2010. Uncertainty in image interpretation as reference for accuracy assessment in object-based image analysis. In: Proceedings of the Ninth International Symposium on Spatial Accuracy Assessment in Natural Resources and Environmental Sciences, pp. 13-16.
4.Andreopoulos, A., Tsotsos, J.K., 2013. 50 Years of object recognition: Directions forward. Comput. Vis. Image Understand. 117, 827-891.
5.Ardila, J.P., Bijker, W., Tolpekin, V.A., Stein, A., 2012. Context-sensitive extraction of tree crown objects in urban areas using VHR satellite images. Int. J. Appl. Earth Obs. Geoinf. 15, 57-69.
6.Ari, C., Aksoy, S., 2014. Detection of compound structures using a Gaussian mixture model with spectral and spatial constraints. IEEE Trans. Geosci. Remote Sens. 52, 6627-6638.
7.Aytekın, Ö., Erener, A., Ulusoy, İ., Düzgün, Ş., 2012. Unsupervised building detection in complex urban environments from multispectral satellite imagery. Int. J. Remote Sens. 33, 2152-2177.
8.Aytekin, Ö., Zöngür, U., Halici, U., 2013. Texture-based airport runway detection. IEEE Geosci. Remote Sens. Lett. 10, 471-475.
9.Baatz, M., Schäpe, A., 2000. Multiresolution segmentation: an optimization approach for high quality multi-scale image segmentation. in: Strobl, J., Blaschke, T., Griesebner, G. (Eds.), Angewandte Geographische Informations-Verarbeitung XII. Wichmann Verlag, Heidelberg, pp. 12-23.
10.Bai, X., Zhang, H., Zhou, J., 2014. VHR object detection based on structural feature extraction and query expansion. IEEE Trans. Geosci. Remote Sens. 52, 6508-6520.
11.Baker, B.A., Warner, T.A., Conley, J.F., McNeil, B.E., 2013. Does spatial resolution matter? A multi-scale comparison of object-based and pixel-based methods for detecting change associated with gas well drilling operations. Int. J. Remote Sens. 34, 1633-1651.
12.Baltsavias, E., 2004. Object extraction and revision by image analysis using existing geodata and knowledge: current status and steps towards operational systems. ISPRS J. Photogramm. Remote Sens. 58, 129-151.
13.Barsi, A., Heipke, C., 2003. Artificial neural networks for the detection of road junctions in aerial images. Int. Arch. Photogramm. Remote Sens. Spat. Inf. Sci. 34, 113-118.
14.Barzohar, M., Coope, D.B., 1996. Automatic finding of main roads in aerial images by using geometric-stochastic models and estimation. IEEE Trans. Pattern Anal. Mach. Intell. 18, 707-721.
15.Benedek, C., Descombes, X., Zerubia, J., 2012. Building development monitoring in multitemporal remotely sensed image pairs with stochastic birth-death dynamics. IEEE Trans. Pattern Anal. Mach. Intell. 34, 33-50.
16.Benedek, C., Shadaydeh, M., Kato, Z., Szirányi, T., Zerubia, J., 2015. Multilayer Markov Random Field models for change detection in optical remote sensing images. ISPRS J. Photogramm. Remote Sens. 107, 22-37.
17.Benz, U.C., Hofmann, P., Willhauck, G., Lingenfelder, I., Heynen, M., 2004. Multi-resolution, object-oriented fuzzy analysis of remote sensing data for GIS-ready information. ISPRS J. Photogramm. Remote Sens. 58, 239-258.
18.Bhagavathy, S., Manjunath, B.S., 2006. Modeling and detection of geospatial objects using texture motifs. IEEE Trans. Geosci. Remote Sens. 44, 3706-3715.
19.Bi, F., Zhu, B., Gao, L., Bian, M., 2012. A visual search inspired computational model for ship detection in optical satellite images. IEEE Geosci. Remote Sens. Lett. 9, 749-753.
20.Bishop, C.M., 1995. Neural networks for pattern recognition. Oxford university press.
21.Blanzieri, E., Melgani, F., 2008. Nearest neighbor classification of remote sensing images with the maximal margin principle. IEEE Trans. Geosci. Remote Sens. 46, 1804-1811.
22.Blaschke, T., 2003. Object-based contextual image classification built on image segmentation. In: Proc. IEEE Workshop on Advances in Techniques for Analysis of Remotely Sensed Data, pp. 113-119.
23.Blaschke, T., 2010. Object based image analysis for remote sensing. ISPRS J. Photogramm. Remote Sens. 65, 2-16.
24.Blaschke, T., Burnett, C., Pekkarinen, A., 2004. Image Segmentation Methods for Object-based Analysis and Classification. in: Jong, S.M.D., Meer, F.D.V.d. (Eds.), Remote Sensing Image Analysis: Including The Spatial Domain. Springer Netherlands, pp. 211-236.
25.Blaschke, T., Hay, G.J., Kelly, M., Lang, S., Hofmann, P., Addink, E., Feitosa, R.Q., van der Meer, F., van der Werff, H., van Coillie, F., 2014. Geographic object-based image analysis–towards a new paradigm. ISPRS J. Photogramm. Remote Sens. 87, 180-191.
26.Blaschke, T., Hay, G.J., Weng, Q., Resch, B., 2011. Collective Sensing: Integrating Geospatial Technologies to Understand Urban Systems—An Overview. Remote Sens. 3, 1743-1776.
27.Blaschke, T., Lang, S., Hay, G.J., 2008. Object-based image analysis: spatial concepts for knowledge-driven remote sensing applications. Springer, Heidelberg, Berlin, New York.
28.Bontemps, S., Bogaert, P., Titeux, N., Defourny, P., 2008. An object-based change detection method accounting for temporal dependences in time series with medium to coarse spatial resolution. Remote Sens. Environ. 112, 3181–3191.
29.Bovolo, F., Bruzzone, L., Marconcini, M., 2008. A novel approach to unsupervised change detection based on a semisupervised SVM and a similarity measure. IEEE Trans. Geosci. Remote Sens. 46, 2070-2082.
30.Cao, L., Luo, J., Liang, F., Huang, T.S., 2009. Heterogeneous feature machines for visual recognition. In: Proc. IEEE Int. Conf. Comput. Vision, pp. 1095-1102.
31.Capobianco, L., Garzelli, A., Camps-Valls, G., 2009. Target detection with semisupervised kernel orthogonal subspace projection. IEEE Trans. Geosci. Remote Sens. 47, 3822-3833.
32.Chaudhuri, D., Kushwaha, N., Samal, A., 2012. Semi-automated road detection from high resolution satellite images by directional morphological enhancement and segmentation techniques. IEEE J. Sel. Topics Appl. Earth Observ. Remote Sens. 5, 1538-1544.
33.Chaudhuri, D., Samal, A., 2008. An automatic bridge detection technique for multispectral images. IEEE Trans. Geosci. Remote Sens. 46, 2720-2727.
34.Chen, G., Hay, G.J., 2012. Object-based change detection. Int. J. Remote Sens. 33, 4434-4457.
35.Chen, Y., Nasrabadi, N.M., Tran, T.D., 2011a. Hyperspectral image classification using dictionary-based sparse representation. IEEE Trans. Geosci. Remote Sens. 49, 3973-3985.
36.Chen, Y., Nasrabadi, N.M., Tran, T.D., 2011b. Simultaneous joint sparsity model for target detection in hyperspectral imagery. IEEE Geosci. Remote Sens. Lett. 8, 676-680.
37.Chen, Y., Nasrabadi, N.M., Tran, T.D., 2011c. Sparse representation for target detection in hyperspectral imagery. IEEE J. Sel. Topics Appl. Earth Observ. Remote Sens. 5, 629-640.
38.Chen, Y., Nasrabadi, N.M., Tran, T.D., 2013. Hyperspectral image classification via kernel sparse representation. IEEE Trans. Geosci. Remote Sens. 51, 217-231.
39.Cheng, G., Guo, L., Zhao, T., Han, J., Li, H., Fang, J., 2013a. Automatic landslide detection from remote-sensing imagery using a scene classification method based on BoVW and pLSA. Int. J. Remote Sens. 34, 45-59.
40.Cheng, G., Han, J., Guo, L., Liu, T., 2015a. Learning Coarse-to-Fine Sparselets for Efficient Object Detection and Scene Classification. In: Proc. IEEE Int. Conf. Comput. Vision Pattern Recognit., pp. 1173-1181.
41.Cheng, G., Han, J., Guo, L., Liu, Z., Bu, S., Ren, J., 2015b. Effective and efficient midlevel visual elements-oriented land-use classification using VHR remote sensing images. IEEE Trans. Geosci. Remote Sens. 53, 4238-4249.
42.Cheng, G., Han, J., Guo, L., Qian, X., Zhou, P., Yao, X., Hu, X., 2013b. Object detection in remote sensing imagery using a discriminatively trained mixture model. ISPRS J. Photogramm. Remote Sens. 85, 32-43.
43.Cheng, G., Han, J., Zhou, P., Guo, L., 2014a. Multi-class geospatial object detection and geographic image classification based on collection of part detectors. ISPRS J. Photogramm. Remote Sens. 98, 119-132.
44.Cheng, G., Han, J., Zhou, P., Guo, L., 2014b. Scalable multi-class geospatial object detection in high-spatial-resolution remote sensing images. In: Proc. IEEE Int. Geosci. Remote Sens. Symp., pp. 2479-2482.
45.Cheng, G., Han, J., Zhou, P., Yao, X., Zhang, D., Guo, L., 2014c. Sparse coding based airport detection from medium resolution Landsat-7 satellite remote sensing images. In: Proc. Int. Workshop Earth Observ. Remote Sens. Appl., pp. 226-230.
46.Cheng, G., Zhou, P., Han, J., Guo, L., Han, J., 2015c. Auto-encoder-based shared mid-level visual dictionary learning for scene classification using very high resolution remote sensing images. IET Computer Vision 9, 639-647.
47.Clinton, N., Holt, A., Scarborough, J., Yan, L., Gong, P., 2010. Accuracy assessment measures for object-based image segmentation goodness. Photogramm. Eng. Remote Sens. 76, 289-299.
48.Congalton, R.G., Green, K., 2009. Assessing the accuracy of remotely sensed data: principles and practices. Taylor and Francis, London.
49.Contreras, D., Blaschke, T., Tiede, D., Jilge, M., 2015. Monitoring recovery after earthquakes through the integration of remote sensing, GIS, and ground observations: the case of L’Aquila (Italy). Cartogr. Geogr. Inf. Sci. 43, 115-133.
50.Contreras, D., Blaschke, T., Tiede, D., Jilge, M., 2016. Monitoring recovery after earthquakes through the integration of remote sensing, GIS, and ground observations: the case of L’Aquila (Italy). Cartogr. Geogr. Inf. Sci. 43, 115-133.
51.Corbane, C., Najman, L., Pecoul, E., Demagistri, L., Petit, M., 2010. A complete processing chain for ship detection using optical satellite imagery. Int. J. Remote Sens. 31, 5837-5854.
52.Cover, T.M., Hart, P.E., 1967. Nearest neighbor pattern classification. IEEE Trans. Inf. Theory 13, 21-27.
53.Cramer, M., 2010. The DGPF test on digital airborne camera evaluation-overview and test design. Photogramm. Eng. Remote Sens. 2010, 73-82.
54.D'Oleire-Oltmanns, S., Marzolff, I., Tiede, D., Blaschke, T., 2014. Detection of gully-affected areas by applying object-based image analysis (OBIA) in the region of Taroudannt, Morocco. Remote Sens. 6, 8287-8309.
55.Dalal, N., Triggs, B., 2005. Histograms of oriented gradients for human detection. In: Proc. IEEE Int. Conf. Comput. Vision Pattern Recognit., pp. 886-893.
56.Das, S., Mirnalinee, T., Varghese, K., 2011. Use of salient features for the design of a multistage framework to extract roads from high-resolution multispectral satellite images. IEEE Trans. Geosci. Remote Sens. 49, 3906-3931.
57.De Morsier, F., Tuia, D., Borgeaud, M., Gass, V., Thiran, J.-P., 2013. Semi-supervised novelty detection using svm entire solution path. IEEE Trans. Geosci. Remote Sens. 51, 1939-1950.
58.De Pinho, C.M.D., Fonseca, L.M.G., Korting, T.S., De Almeida, C.M., Kux, H.J.H., 2012. Land-cover classification of an intra-urban environment using high-resolution images and object-based image analysis. Int. J. Remote Sens. 33, 5973-5995.
59.Deselaers, T., Alexe, B., Ferrari, V., 2012. Weakly supervised localization and learning with generic knowledge. Int. J. Comput. Vis. 100, 275-293.
60.Dissanska, M., Bernier, M., Payette, S., 2009. Object-based classification of very high resolution panchromatic images for evaluating recent change in the structure of patterned peatland. Can. J. Remote Sens. 35, 189-215.
61.Doleire-Oltmanns, S., Eisank, C., Dragut, L., Blaschke, T., 2013. An object-based workflow to extract landforms at multiple scales from two distinct data types. IEEE Geosci. Remote Sens. Lett. 10, 947-951.
62.Dong, Y., Du, B., Zhang, L., 2015. Target Detection Based on Random Forest Metric Learning. IEEE J. Sel. Topics Appl. Earth Observ. Remote Sens. 8, 1830-1838.
63.Doxani, G., Karantzalos, K., Tsakiri-Strati, M., 2012. Monitoring urban changes based on scale-space filtering and object-oriented classification. Int. J. Appl. Earth Obs. Geoinf. 15, 38-48.
64.Doxani, G., Siachalou, S., Tsakiri-Strati, M., 2008. An object-oriented approach to urban land cover change detection. Int. Arch. Photogramm. Remote Sens. Spat. Inf. Sci. 37, 1655-1660.
65.Drăguţ, L., Blaschke, T., 2006. Automated classification of landform elements using object-based image analysis. Geomorphology 81, 330-344.
66.Drăguţ, L., Csillik, O., Eisank, C., Tiede, D., 2014. Automated parameterisation for multi-scale image segmentation on multiple layers. ISPRS J. Photogramm. Remote Sens. 88, 119-127.
67.Drăguţ, L., Eisank, C., 2012. Automated object-based classification of topography from SRTM data. Geomorphology 141, 21-33.
68.Drăguţ, L., Tiede, D., Levick, S.R., 2010. ESP: A tool to estimate scale parameter for multiresolution image segmentation of remotely sensed data. Int. J. Geogr. Inf. Sci. 24, 859-871.
69.Du, B., Zhang, L., 2014. A discriminative metric learning based anomaly detection method. IEEE Trans. Geosci. Remote Sens. 52, 6844-6857.
70.Dudani, S.A., 1976. The distance-weighted k-nearest-neighbor rule. IEEE Trans. Syst., Man, Cybern., 325-327.
71.Durieux, L., Lagabrielle, E., Nelson, A., 2008. A method for monitoring building construction in urban sprawl areas using object-based analysis of Spot 5 images and existing GIS data. ISPRS J. Photogramm. Remote Sens. 63, 399-408.
72.Duro, D.C., Franklin, S.E., Dubé, M.G., 2012. A comparison of pixel-based and object-based image analysis with selected machine learning algorithms for the classification of agricultural landscapes using SPOT-5 HRG imagery. Remote Sens. Environ. 118, 259-272.
73.Eikvil, L., Aurdal, L., Koren, H., 2009. Classification-based vehicle detection in high-resolution satellite images. ISPRS J. Photogramm. Remote Sens. 64, 65-72.
74.Eisank, C., Drăguţ, L., Blaschke, T., 2011. A generic procedure for semantics-oriented landform classification using object-based image analysis. Geomorphometry, 125-128.
75.Esch, T., Thiel, M., Bock, M., Roth, A., Dech, S., 2008. Improvement of image segmentation accuracy based on multiscale optimization procedure. IEEE Geosci. Remote Sens. Lett. 5, 463-467.
76.Fauvel, M., Chanussot, J., Benediktsson, J.A., 2006. Decision fusion for the classification of urban remote sensing images. IEEE Trans. Geosci. Remote Sens. 44, 2828-2838.
77.Feizizadeh, B., Tiede, D., Rezaei Moghaddam, M.H., Blaschke, T., 2014. Systematic evaluation of fuzzy operators for object-based landslide mapping. South-Eastern European Journal of Earth Observation and Geomatics 3, 219-222.
78.Fischler, M.A., Elschlager, R.A., 1973. The representation and matching of pictorial structures. EEE Trans. Comput., 67-92.
79.Flanders, D., Hall-Beyer, M., Pereverzoff, J., 2003. Preliminary evaluation of eCognition object-based software for cut block delineation and feature extraction. Can. J. Remote Sens. 29, 441-452.
80.Freund, Y., 1995. Boosting a weak learning algorithm by majority. Inf. Comput. 121, 256-285.
81.Freund, Y., Schapire, R.E., 1996. Experiments with a new boosting algorithm. In: Proc. Int. Conf. Mach. Learn., pp. 148-156.
82.Freund, Y., Schapire, R.E., 1997. A decision-theoretic generalization of on-line learning and an application to boosting. J. Comput. Syst. Sci. 55, 119-139.
83.Gao, Y., Mas, J.F., Kerle, N., Navarrete Pacheco, J.A., 2011. Optimal region growing segmentation and its effect on classification accuracy. Int. J. Remote Sens. 32, 3747-3763.
84.Ghosh, S., Bruzzone, L., Patra, S., Bovolo, F., Ghosh, A., 2007. A context-sensitive technique for unsupervised change detection based on Hopfield-type neural networks. IEEE Trans. Geosci. Remote Sens. 45, 778-789.
85.Goodin, D.G., Anibas, K.L., Bezymennyi, M., 2015. Mapping land cover and land use from object-based classification: an example from a complex agricultural landscape. Int. J. Remote Sens. 36, 4702-4723.
86.Grabner, H., Nguyen, T.T., Gruber, B., Bischof, H., 2008. On-line boosting-based car detection from aerial images. ISPRS J. Photogramm. Remote Sens. 63, 382-396.
87.Haala, N., Brenner, C., 1999. Extraction of buildings and trees in urban environments. ISPRS J. Photogramm. Remote Sens. 54, 130-137.
88.Haapanen, R., Ek, A.R., Bauer, M.E., Finley, A.O., 2004. Delineation of forest/nonforest land use classes using nearest neighbor methods. Remote Sens. Environ. 89, 265-271.
89.Han, J., Zhang, D., Cheng, G., Guo, L., Ren, J., 2015. Object detection in optical remote sensing images based on weakly supervised learning and high-level feature learning. IEEE Trans. Geosci. Remote Sens. 53, 3325-3337.
90.Han, J., Zhou, P., Zhang, D., Cheng, G., Guo, L., Liu, Z., Bu, S., Wu, J., 2014. Efficient, simultaneous detection of multi-class geospatial targets based on visual saliency modeling and discriminative learning of sparse coding. ISPRS J. Photogramm. Remote Sens. 89, 37-48.
91.Hariharan, B., Malik, J., Ramanan, D., 2012. Discriminative decorrelation for clustering and classification. In: Proc. Eur. Conf. Comput. Vis., pp. 459-472.
92.Hay, G.J., Blaschke, T., Marceau, D.J., Bouchard, A., 2003. A comparison of three image-object methods for the multiscale analysis of landscape structure. ISPRS J. Photogramm. Remote Sens. 57, 327-345.
93.Hay, G.J., Castilla, G., Wulder, M.A., Ruiz, J.R., 2005. An automated object-based approach to the multiscale image segmentation of forest scenes. Int. J. Appl. Earth Observ. Geoinform. 7, 339–359.
94.Heitz, G., Koller, D., 2008. Learning spatial context: Using stuff to find things. In: Proc. Eur. Conf. Comput. Vis., pp. 30-43.
95.Hinton, G.E., Salakhutdinov, R.R., 2006. Reducing the dimensionality of data with neural networks. Science 313, 504-507.
96.Hofmann, A.D., Maas, H.-G., Streilein, A., 2002. Knowledge-based building detection based on laser scanner data and topographic map information. Int. Arch. Photogramm. Remote Sens. Spat. Inf. Sci. 34, 169-174.
97.Hofmann, P., Blaschke, T., Strobl, J., 2011. Quantifying the robustness of fuzzy rule sets in object-based image analysis. Int. J. Remote Sens. 32, 7359-7381.
98.Hu, J., Razdan, A., Femiani, J.C., Cui, M., Wonka, P., 2007. Road network extraction and intersection detection from aerial images by tracking road footprints. IEEE Trans. Geosci. Remote Sens. 45, 4144-4157.
99.Huang, X., Zhang, L., 2009. Road centreline extraction from high-resolution imagery based on multiscale structural features and support vector machines. Int. J. Remote Sens. 30, 1977-1987.
100.Huang, Y., Wu, Z., Wang, L., Tan, T., 2014. Feature coding in image classification: A comprehensive study. IEEE Trans. Pattern Anal. Mach. Intell. 36, 493-506.
101.Huertas, A., Nevatia, R., 1988. Detecting buildings in aerial images. Comput. Vis. Graph. Image Process. 41, 131-152.
102.Hung, C., Bryson, M., Sukkarieh, S., 2012. Multi-class predictive template for tree crown detection. ISPRS J. Photogramm. Remote Sens. 68, 170-183.
103.Hussain, M., Chen, D., Cheng, A., Wei, H., Stanley, D., 2013. Change detection from remotely sensed images: From pixel-based to object-based approaches. ISPRS J. Photogramm. Remote Sens. 80, 91-106.
104.Im, J., Jensen, J.R., Tullis, J.A., 2008. Object-based change detection using correlation image analysis and image segmentation. Int. J. Remote Sens. 29, 399-423.
105.Inglada, J., 2007. Automatic recognition of man-made objects in high resolution optical remote sensing images by SVM classification of geometric image features. ISPRS J. Photogramm. Remote Sens. 62, 236-248.
106.Irvin, R.B., McKeown, D.M., 1989. Methods for exploiting the relationship between buildings and their shadows in aerial imagery. IEEE Trans. Syst., Man, Cybern. 19, 1564-1575.
107.Jain, A.K., Ratha, N.K., Lakshmanan, S., 1997. Object detection using Gabor filters. Pattern Recog. 30, 295-309.
108.Jain, A.K., Zhong, Y., Dubuisson-Jolly, M.-P., 1998. Deformable template models: A review. Signal processing 71, 109-129.
109.Janssen, L., Middelkoop, H., 1992. Knowledge-based crop classification of a Landsat Thematic Mapper image. Int. J. Remote Sens. 13, 2827-2837.
110.Jin, X., Davis, C.H., 2007. Vehicle detection from high-resolution satellite imagery using morphological shared-weight neural networks. Image Vis. Comput. 25, 1422-1431.
111.Jing, Y., An, J., Liu, Z., 2011. A novel edge detection algorithm based on global minimization active contour model for oil slick infrared aerial image. IEEE Trans. Geosci. Remote Sens. 49, 2005-2013.
112.Jungho, I., Quackenbush, L.J., Li, M., Fang, F., 2014. Optimum Scale in Object-Based Image Analysis. in: Weng, Q. (Ed.), Scale Issues in Remote Sensing. John Wiley & Sons, Inc., pp. 197-214.
113.Karantzalos, K., Paragios, N., 2009. Recognition-driven two-dimensional competing priors toward automatic and accurate building detection. IEEE Trans. Geosci. Remote Sens. 47, 133-144.
114.Kasetkasem, T., Varshney, P.K., 2002. An image change detection algorithm based on Markov random field models. IEEE Trans. Geosci. Remote Sens. 40, 1815-1823.
115.Kembhavi, A., Harwood, D., Davis, L.S., 2011. Vehicle detection using partial least squares. IEEE Trans. Pattern Anal. Mach. Intell. 33, 1250-1265.
116.Kim, M., Madden, M., Warner, T., 2008. Estimation of optimal image object size for the segmentation of forest stands with multispectral IKONOS imagery. in: Blaschke, T., Lang, S., Hay, G.J. (Eds.), Object-Based Image Analysis-Spatial concepts for knowledge driven remote Sensing applications. Springer, Berlin, Heidelberg, pp. 291-307.
117.Kim, M., Warner, T.A., Madden, M., Atkinson, D.S., 2011. Multi-scale GEOBIA with very high spatial resolution digital aerial imagery: scale, texture and image objects. Int. J. Remote Sens. 32, 2825-2850.
118.Kim, T., Park, S.-R., Kim, M.-G., Jeong, S., Kim, K.-O., 2004. Tracking road centerlines from high resolution remote sensing images by least squares correlation matching. Photogramm. Eng. Remote Sens. 70, 1417-1422.
119.Kumar, S., Hebert, M., 2003. Discriminative random fields: A discriminative framework for contextual interaction in classification. In: Proc. IEEE Int. Conf. Comput. Vision, pp. 1150-1157.
120.Lafferty, J., McCallum, A., Pereira, F.C., 2001. Conditional random fields: Probabilistic models for segmenting and labeling sequence data. In: Proc. Int. Conf. Mach. Learn., pp. 282-289.
121.Laptev, I., Mayer, H., Lindeberg, T., Eckstein, W., Steger, C., Baumgartner, A., 2000. Automatic extraction of roads from aerial images based on scale space and snakes. Mach. Vis. Appl. 12, 23-31.
122.Lazebnik, S., Schmid, C., Ponce, J., 2006. Beyond bags of features: Spatial pyramid matching for recognizing natural scene categories. In: Proc. IEEE Int. Conf. Comput. Vision Pattern Recognit., pp. 2169-2178.
123.Le Hégarat-Mascle, S., André, C., 2009. Use of Markov random fields for automatic cloud/shadow detection on high resolution optical images. ISPRS J. Photogramm. Remote Sens. 64, 351-366.
124.LeCun, Y., Bengio, Y., Hinton, G., 2015. Deep learning. Nature 521, 436-444.
125.Lefèvre, S., Weber, J., Sheeren, D., 2007. Automatic building extraction in VHR images using advanced morphological operators. In: Proc. Urban Remote Sensing Joint Event, pp. 1-5.
126.Lei, Z., Fang, T., Huo, H., Li, D., 2012. Rotation-invariant object detection of remotely sensed images based on texton forest and hough voting. IEEE Trans. Geosci. Remote Sens. 50, 1206-1217.
127.Lei, Z., Fang, T., Huo, H., Li, D., 2014. Bi-Temporal Texton Forest for Land Cover Transition Detection on Remotely Sensed Imagery. IEEE Trans. Geosci. Remote Sens. 52, 1227-1237.
128.Leitloff, J., Hinz, S., Stilla, U., 2010. Vehicle detection in very high resolution satellite images of city areas. IEEE Trans. Geosci. Remote Sens. 48, 2795-2806.
129.Leninisha, S., Vani, K., 2015. Water flow based geometric active deformable model for road network. ISPRS J. Photogramm. Remote Sens. 102, 140-147.
130.Leon, J., Woodroffe, C.D., 2011. Improving the synoptic mapping of coral reef geomorphology using object-based image analysis. Int. J. Geogr. Inf. Sci. 25, 949-969.
131.Lhomme, S., He, D.C., Weber, C., Morin, D., 2009. A new approach to building identification from very-high-spatial-resolution images. Int. J. Remote Sens. 30, 1341-1354.
132.Li, E., Femiani, J., Xu, S., Zhang, X., Wonka, P., 2015a. Robust Rooftop Extraction From Visible Band Images Using Higher Order CRF. IEEE Trans. Geosci. Remote Sens. 53, 4483-4495.
133.Li, F.F., Perona, P., 2005. A bayesian hierarchical model for learning natural scene categories. In: Proc. IEEE Int. Conf. Comput. Vision Pattern Recognit., pp. 524-531.
134.Li, X., Cheng, X., Chen, W., Chen, G., Liu, S., 2015b. Identification of forested landslides using LiDar data, object-based image analysis, and machine learning algorithms. Remote Sens. 7, 9705-9726.
135.Li, X., Myint, S.W., Zhang, Y., Galletti, C., Zhang, X., Ii, B.L.T., 2014. Object-based land-cover classification for metropolitan Phoenix, Arizona, using aerial photography. Int. J. Appl. Earth Obs. Geoinf. 33, 321-330.
136.Li, X., Shao, G., 2013. Object-based urban vegetation mapping with high-resolution aerial photography as a single data source. Int. J. Remote Sens. 34, 771-789.
137.Li, X., Zhang, S., Pan, X., Dale, P., Cropp, R., 2010. Straight road edge detection from high-resolution remote sensing images based on the ridgelet transform with the revised parallel-beam Radon transform. Int. J. Remote Sens. 31, 5041-5059.
138.Li, Y., Wang, S., Tian, Q., Ding, X., 2015c. Feature representation for statistical-learning-based object detection: A review. Pattern Recog. 48, 3542-3559.
139.Li, Z., Itti, L., 2011. Saliency and gist features for target detection in satellite images. IEEE Trans. Image Process. 20, 2017-2029.
140.Li, Z., Khananian, A., Fraser, R.H., Cihlar, J., 2001. Automatic detection of fire smoke using artificial neural networks and threshold approaches applied to AVHRR imagery. IEEE Trans. Geosci. Remote Sens. 39, 1859-1870.
141.Lienhart, R., Kuranov, A., Pisarevsky, V., 2003. Empirical analysis of detection cascades of boosted classifiers for rapid object detection. In: Proc. DAGM Symp., pp. 297-304.
142.Lin, C., Nevatia, R., 1998. Building detection and description from a single intensity image. Comput. Vis. Image Understand. 72, 101-121.
143.Lin, Y., He, H., Yin, Z., Chen, F., 2015. Rotation-invariant object detection in remote sensing images based on radial-gradient angle. IEEE Geosci. Remote Sens. Lett. 12, 746-750.
144.Liow, Y.-T., Pavlidis, T., 1990. Use of shadows for extracting buildings in aerial images. Comput. Vis. Graph. Image Process. 49, 242-277.
145.Lisita, A., Sano, E., Durieux, L., 2013. Identifying potential areas of Cannabis sativa plantations using object-based image analysis of SPOT-5 satellite data. Int. J. Remote Sens. 34, 5409-5428.
146.Liu, G., Sun, X., Fu, K., Wang, H., 2013a. Aircraft recognition in high-resolution satellite images using coarse-to-fine shape prior. IEEE Geosci. Remote Sens. Lett. 10, 573-577.
147.Liu, G., Sun, X., Fu, K., Wang, H., 2013b. Interactive geospatial object extraction in high resolution remote sensing images using shape-based global minimization active contour model. Pattern Recog. Lett. 34, 1186-1195.
148.Liu, L., Shi, Z., 2014. Airplane detection based on rotation invariant and sparse coding in remote sensing images. Optik-Int. J. Light Electron Opt. 125, 5327-5333.
149.Liu, Q., Liao, X., Carin, L., 2008. Detection of unexploded ordnance via efficient semisupervised and active learning. IEEE Trans. Geosci. Remote Sens. 46, 2558-2567.
150.Lizarazo, I., 2014. Accuracy assessment of object-based image classification: another STEP. Int. J. Remote Sens. 35, 6135-6156.
151.Lowe, D.G., 2004. Distinctive image features from scale-invariant keypoints. Int. J. Comput. Vis. 60, 91-110.
152.Ma, L., Crawford, M.M., Tian, J., 2010. Local manifold learning-based-nearest-neighbor for hyperspectral image classification. IEEE Trans. Geosci. Remote Sens. 48, 4099-4109.
153.Macfaden, S.W., O'Neil-Dunne, J.P.M., Royar, A.R., Lu, J.W.T., Rundle, A.G., 2012. High-resolution tree canopy mapping for New York City using LIDAR and object-based image analysis. J. Appl. Remote Sens. 6, 1-23.
154.MacLean, M.G., Congalton, R.G., 2012. Map accuracy assessment issues when using an object-oriented approach. In: Proceedings of the American Society for Photogrammetry and Remote Sensing 2012 Annual Conference, pp. 1-5.
155.Maillard, P., Cavayas, F., 1989. Automatic map-guided extraction of roads from SPOT imagery for cartographic database updating. Int. J. Remote Sens. 10, 1775-1787.
156.Malek, S., Bazi, Y., Alajlan, N., AlHichri, H., Melgani, F., 2014. Efficient framework for palm tree detection in UAV images. IEEE J. Sel. Topics Appl. Earth Observ. Remote Sens. 7, 4692-4703.
157.Mallinis, G., Koutsias, N., Tsakiri-Strati, M., Karteris, M., 2008. Object-based classification using Quickbird imagery for delineating forest vegetation polygons in a Mediterranean test site. ISPRS J. Photogramm. Remote Sens. 63, 237-250.
158.Martha, T.R., Kerle, N., Jetten, V., van Westen, C.J., Kumar, K.V., 2010. Characterising spectral, spatial and morphometric properties of landslides for semi-automatic detection using object-oriented methods. Geomorphology 116, 24-36.
159.Martha, T.R., Kerle, N., Van Westen, C.J., Jetten, V., Kumar, K.V., 2011. Segment optimization and data-driven thresholding for knowledge-based landslide detection by object-based image analysis. IEEE Trans. Geosci. Remote Sens. 49, 4928-4943.
160.Martha, T.R., Kerle, N., Westen, C.J.V., Jetten, V., Kumar, K.V., 2012. Object-oriented analysis of multi-temporal panchromatic images for creation of historical landslide inventories. ISPRS J. Photogramm. Remote Sens. 67, 105-119.
161.Mayer, H., 1999. Automatic object extraction from aerial imagery—a survey focusing on buildings. Comput. Vis. Image Understand. 74, 138-149.
162.Mayer, H., Hinz, S., Bacher, U., Baltsavias, E., 2006. A test of automatic road extraction approaches. Int. Arch. Photogramm. Remote Sens. Spat. Inf. Sci. 36, 209-214.
163.McGlone, J.C., Shufelt, J., 1994. Projective and object space geometry for monocular building extraction. In: Proc. IEEE Int. Conf. Comput. Vision Pattern Recognit., pp. 54-61.
164.McKeown, D.M., Denlinger, J.L., 1988. Cooperative methods for road tracking in aerial imagery. In: Proc. IEEE Int. Conf. Comput. Vision Pattern Recognit., pp. 662-672.
165.Mena, J.B., 2003. State of the art on automatic road extraction for GIS update: a novel classification. Pattern Recog. Lett. 24, 3037-3058.
166.Mikolajczyk, K., Schmid, C., 2001. Indexing based on scale invariant interest points. In: Proc. IEEE Int. Conf. Comput. Vision, pp. 525-531.
167.Ming, D., Li, J., Wang, J., Zhang, M., 2015. Scale parameter selection by spatial statistics for GeOBIA: Using mean-shift based multi-scale segmentation as an example. ISPRS J. Photogramm. Remote Sens. 106, 28-41.
168.Mishra, N.B., Crews, K.A., 2014. Mapping vegetation morphology types in a dry savanna ecosystem: integrating hierarchical object-based image analysis with Random Forest. Int. J. Remote Sens. 35, 1175-1198.
169.Mokhtarzade, M., Zoej, M.V., 2007. Road detection from high-resolution satellite images using artificial neural networks. Int. J. Appl. Earth Observ. Geoinform. 9, 32-40.
170.Moon, H., Chellappa, R., Rosenfeld, A., 2002. Performance analysis of a simple vehicle detection algorithm. Image Vis. Comput. 20, 1-13.
171.Moskal, L.M., Styers, D.M., Halabisky, M., 2011. Monitoring urban tree cover using object-based image analysis and public domain remotely sensed data. Remote Sens. 3, 2243-2262.
172.Mountrakis, G., Im, J., Ogole, C., 2011. Support vector machines in remote sensing: A review. ISPRS J. Photogramm. Remote Sens. 66, 247-259.
173.Moustakidis, S., Mallinis, G., Koutsias, N., Theocharis, J.B., Petridis, V., 2012. SVM-based fuzzy decision trees for classification of high spatial resolution remote sensing images. IEEE Trans. Geosci. Remote Sens. 50, 149-169.
174.Movaghati, S., Moghaddamjoo, A., Tavakoli, A., 2010. Road extraction from satellite images using particle filtering and extended Kalman filtering. IEEE Trans. Geosci. Remote Sens. 48, 2807-2817.
175.Myint, S.W., Gober, P., Brazel, A., Grossman-Clarke, S., Weng, Q., 2011. Per-pixel vs. object-based classification of urban land cover extraction using high spatial resolution imagery. Remote Sens. Environ. 115, 1145-1161.
176.Nebiker, S., Lack, N., Deuber, M., 2014. Building change detection from historical aerial photographs using dense image matching and object-based image analysis. Remote Sens. 6, 8310-8336.
177.Niu, X., 2006. A semi-automatic framework for highway extraction and vehicle detection based on a geometric deformable model. ISPRS J. Photogramm. Remote Sens. 61, 170-186.
178.Ojala, T., Pietikäinen, M., Mäenpää, T., 2002. Multiresolution gray-scale and rotation invariant texture classification with local binary patterns. IEEE Trans. Pattern Anal. Mach. Intell. 24, 971-987.
179.Ok, A.O., 2013. Automated detection of buildings from single VHR multispectral images using shadow information and graph cuts. ISPRS J. Photogramm. Remote Sens. 86, 21-40.
180.Ok, A.O., Senaras, C., Yuksel, B., 2013. Automated detection of arbitrarily shaped buildings in complex environments from monocular VHR optical satellite imagery. IEEE Trans. Geosci. Remote Sens. 51, 1701-1717.
181.Pacifici, F., Chini, M., Emery, W.J., 2009. A neural network approach using multi-scale textural metrics from very high-resolution panchromatic imagery for urban land-use classification. Remote Sens. Environ. 113, 1276-1292.
182.Peng, J., Liu, Y., 2005. Model and context-driven building extraction in dense urban aerial images. Int. J. Remote Sens. 26, 1289-1307.
183.Peng, J., Zhang, D., Liu, Y., 2005. An improved snake model for building detection from urban aerial images. Pattern Recog. Lett. 26, 587-595.
184.Phinn, S.R., Mumby, C.M.R., J., P., 2012. Multi-scale, object-based image analysis for mapping geomorphic and ecological zones on coral reefs. Int. J. Remote Sens. 33, 3768-3797.
185.Qian, Y., Ye, M., Zhou, J., 2013. Hyperspectral image classification based on structured sparse logistic regression and three-dimensional wavelet texture features. IEEE Trans. Geosci. Remote Sens. 51, 2276-2291.
186.Radoux, J., Bogaert, P., Fasbender, D., Defourny, P., 2011. Thematic accuracy assessment of geographic object-based image classification. Int. J. Geogr. Inf. Sci. 25, 895-911.
187.Schapire, R.E., Singer, Y., 1999. Improved boosting algorithms using confidence-rated predictions. Machine learning 37, 297-336.
188.Senaras, C., Ozay, M., Yarman Vural, F.T., 2013. Building detection with decision fusion. IEEE J. Sel. Topics Appl. Earth Observ. Remote Sens. 6, 1295-1304.
189.Shi, Z., Yu, X., Jiang, Z., Li, B., 2014. Ship detection in high-resolution optical imagery based on anomaly detector and local shape feature. IEEE Trans. Geosci. Remote Sens. 52, 4511-4523.
190.Shufelt, J., 1999. Performance evaluation and analysis of monocular building extraction from aerial imagery. IEEE Trans. Pattern Anal. Mach. Intell. 21, 311-326.
191.Shufelt, J.A., 1996. Exploiting photogrammetric methods for building extraction in aerial images. Int. Arch. Photogramm. Remote Sens. 31, B6.
192.Sirmaçek, B., Ünsalan, C., 2009. Urban-area and building detection using SIFT keypoints and graph theory. IEEE Trans. Geosci. Remote Sens. 47, 1156-1167.
193.Sirmacek, B., Ünsalan, C., 2011. A probabilistic framework to detect buildings in aerial and satellite images. IEEE Trans. Geosci. Remote Sens. 49, 211-221.
194.Solberg, A.H.S., 1999. Contextual data fusion applied to forest map revision. IEEE Trans. Geosci. Remote Sens. 37, 1234-1243.
195.Song, M., Civco, D., 2004. Road extraction using SVM and image segmentation. Photogramm. Eng. Remote Sens. 70, 1365-1371.
196.Stankov, K., He, D.-C., 2013. Building detection in very high spatial resolution multispectral images using the hit-or-miss transform. IEEE Geosci. Remote Sens. Lett. 10, 86-90.
197.Stankov, K., He, D.-C., 2014. Detection of buildings in multispectral very high spatial resolution images using the percentage occupancy hit-or-miss transform. IEEE J. Sel. Topics Appl. Earth Observ. Remote Sens. 7, 4069-4080.
198.Stilla, U., Geibel, R., Jurkiewicz, K., 1997. Building reconstruction using different views and context knowledge. Int. Arch. Photogramm. Remote Sens. 32, 129-136.
199.Stumpf, A., Kerle, N., 2011. Object-oriented mapping of landslides using Random Forests. Remote Sens. Environ. 115, 2564-2577.
200.Sugiyama, M., 2007. Dimensionality reduction of multimodal labeled data by local fisher discriminant analysis. J. Mach. Learn. Res. 8, 1027-1061.
201.Sun, H., Sun, X., Wang, H., Li, Y., Li, X., 2012. Automatic target detection in high-resolution remote sensing images using spatial sparse coding bag-of-words model. IEEE Geosci. Remote Sens. Lett. 9, 109-113.
202.Sun, X., Wang, H., Fu, K., 2010. Automatic detection of geospatial objects using taxonomic semantics. IEEE Geosci. Remote Sens. Lett. 7, 23-27.
203.Tang, J., Deng, C., Huang, G.-B., Zhao, B., 2015. Compressed-domain ship detection on spaceborne optical image using deep neural network and extreme learning machine. IEEE Trans. Geosci. Remote Sens. 53, 1174-1185.
204.Tanner, F., Colder, B., Pullen, C., Heagy, D., Eppolito, M., Carlan, V., Oertel, C., Sallee, P., 2009. Overhead imagery research data set-an annotated data library & tools to aid in the development of computer vision algorithms. In: Proc. IEEE Applied Imagery Pattern Recognition Workshop, pp. 1-8.
205.Tao, C., Tan, Y., Cai, H., Tian, J., 2011. Airport detection from large IKONOS images using clustered SIFT keypoints and region information. IEEE Geosci. Remote Sens. Lett. 8, 128-132.
206.Tao, L., Sun, F., Yang, S., 2012. A fast and robust sparse approach for hyperspectral data classification using a few labeled samples. IEEE Trans. Geosci. Remote Sens. 50, 2287-2302.
207.Tchoku, C., Karnieli, A., Meisels, A., Chorowicz, J., 1996. Detection of drainage channel networks on digital satellite images. Int. J. Remote Sens. 17, 1659-1678.
208.Trinder, J.C., Wang, Y., 1998. Knowledge-based road interpretation in aerial images. Int. Arch. Photogramm. Remote Sens. 32, 635-640.
209.Tuermer, S., Kurz, F., Reinartz, P., Stilla, U., 2013. Airborne vehicle detection in dense urban areas using HoG features and disparity maps. IEEE J. Sel. Topics Appl. Earth Observ. Remote Sens. 6, 2327-2337.
210.Tzotsos, A., Karantzalos, K., Argialas, D., 2011. Object-based image analysis through nonlinear scale-space filtering. ISPRS J. Photogramm. Remote Sens. 66, 2-16.
211.Ünsalan, C., Sirmacek, B., 2012. Road network detection using probabilistic and graph theoretical methods. IEEE Trans. Geosci. Remote Sens. 50, 4441-4453.
212.Vapnik, V.N., Vapnik, V., 1998. Statistical learning theory. Wiley New York.
213.Viola, P., Jones, M., 2001. Rapid object detection using a boosted cascade of simple features. In: Proc. IEEE Int. Conf. Comput. Vision Pattern Recognit., pp. 511-518.
214.Walker, J.S., Blaschke, T., 2008. Object-based land-cover classification for the Phoenix metropolitan area: optimization vs. transportability. Int. J. Remote Sens. 29, 2021-2040.
215.Walker, J.S., Briggs, J.M., 2007. An object-oriented approach to urban forest mapping in Phoenix. Photogramm. Eng. Remote Sens. 73, 577-583.
216.Walter, V., 2004. Object-based classification of remote sensing data for change detection. ISPRS J. Photogramm. Remote Sens. 58, 225-238.
217.Wang, F., 1993. A knowledge-based vision system for detecting land changes at urban fringes. IEEE Trans. Geosci. Remote Sens. 31, 136-145.
218.Wang, F., Newkirk, R., 1988. A knowledge-based system for highway network extraction. IEEE Trans. Geosci. Remote Sens. 26, 525-531.
219.Wang, H., Nie, F., Huang, H., Ding, C., 2013. Heterogeneous visual features fusion via sparse multimodal machine. In: Proc. IEEE Int. Conf. Comput. Vision Pattern Recognit., pp. 3097-3102.
220.Wang, J., Song, J., Chen, M., Yang, Z., 2015. Road network extraction: a neural-dynamic framework based on deep learning and a finite state machine. Int. J. Remote Sens. 36, 3144-3169.
221.Wang, M., Zhang, S., 2011. Road extraction from high-spatial-resolution remotely sensed imagery by combining multi-profile analysis and extended Snakes model. Int. J. Remote Sens. 32, 6349-6365.
222.Weber, J., Lefèvre, S., 2008. A multivariate hit-or-miss transform for conjoint spatial and spectral template matching. In: Proc. IEEE Int. Conf. Image and Signal Processing, pp. 226-235.
223.Weber, J., Lefèvre, S., 2012. Spatial and spectral morphological template matching. Image Vis. Comput. 30, 934-945.
224.Wegner, J.D., Hänsch, R., Thiele, A., Soergel, U., 2011a. Building detection from one orthophoto and high-resolution InSAR data using conditional random fields. IEEE J. Sel. Topics Appl. Earth Observ. Remote Sens. 4, 83-91.
225.Wegner, J.D., Soergel, U., Rosenhahn, B., 2011b. Segment-based building detection with conditional random fields. In: Proc. Joint Urban Remote Sensing Event, pp. 205-208.
226.Weidner, U., Förstner, W., 1995. Towards automatic building extraction from high-resolution digital elevation models. ISPRS J. Photogramm. Remote Sens. 50, 38-49.
227.Wen, X., Shao, L., Fang, W., Xue, Y., 2015. Efficient feature selection and classification for vehicle detection. IEEE Trans. Circuits Syst. Video Technol. 25, 508-517.
228.Weng, Q., 2009. Remote Sensing and GIS Integration-Theories, Methods, and Applications. McGraw-Hill, New York.
229.Weng, Q., 2011. Advances in Environmental Remote Sensing: Sensor, Algorithms, and Applications. CRC Press, Boca Raton, FL, USA.
230.Woodcock, C.E., Strahler, A.H., 1987. The factor of scale in remote sensing. Remote Sens. Environ. 21, 311-332.
231.Wright, J., Yang, A.Y., Ganesh, A., Sastry, S.S., Ma, Y., 2009. Robust face recognition via sparse representation. IEEE Trans. Pattern Anal. Mach. Intell. 31, 210-227.
232.Xie, Z., Roberts, C., Johnson, B., 2008. Object-based target search using remotely sensed data: A case study in detecting invasive exotic Australian Pine in south Florida. ISPRS J. Photogramm. Remote Sens. 63, 647-660.
233.Xu, C., Duan, H., 2010. Artificial bee colony (ABC) optimized edge potential function (EPF) approach to target recognition for low-altitude aircraft. Pattern Recog. Lett. 31, 1759-1772.
234.Xu, S., Fang, T., Li, D., Wang, S., 2010. Object classification of aerial images with bag-of-visual words. IEEE Geosci. Remote Sens. Lett. 7, 366-370.
235.Yang, J.-M., Yu, P.-T., Kuo, B.-C., 2010. A nonparametric feature extraction and its application to nearest neighbor classification for hyperspectral image data. IEEE Trans. Geosci. Remote Sens. 48, 1279-1293.
236.Yang, L., Bi, G., Xing, M., Zhang, L., 2015. Airborne SAR Moving Target Signatures and Imagery Based on LVD. IEEE Trans. Geosci. Remote Sens. 53, 5958-5971.
237.Yang, L., Xing, M., Wang, Y., Zhang, L., Bao, Z., 2013. Compensation for the NsRCM and phase error after polar format resampling for airborne spotlight SAR raw data of high resolution. IEEE Geosci. Remote Sens. Lett. 10, 165-169.
238.Yang, Y., Newsam, S., 2010. Bag-of-visual-words and spatial extensions for land-use classification. In: Proc. ACM SIGSPATIAL Int. Conf. Adv. Geogr. Inform. Syst., pp. 270-279.
239.Yang, Y., Newsam, S., 2011. Spatial pyramid co-occurrence for image classification. In: Proc. IEEE Int. Conf. Comput. Vision, pp. 1465-1472.
240.Yang, Y., Newsam, S., 2013. Geographic image retrieval using local invariant features. IEEE Trans. Geosci. Remote Sens. 51, 818-832.
241.Yao, X., Han, J., Guo, L., Bu, S., Liu, Z., 2015. A coarse-to-fine model for airport detection from remote sensing images using target-oriented visual saliency and CRF. Neurocomputing 164, 162-172.
242.Yin, J., Li, H., Jia, X., 2015. Crater Detection Based on Gist Features. IEEE J. Sel. Topics Appl. Earth Observ. Remote Sens. 8, 23-29.
243.Yokoya, N., Iwasaki, A., 2015. Object Detection Based on Sparse Representation and Hough Voting for Optical Remote Sensing Imagery. IEEE J. Sel. Topics Appl. Earth Observ. Remote Sens. 8, 2053-2062.
244.Yu, Q., Gong, P., Clinton, N., Biging, G., Kelly, M., Schirokauer, D., 2006. Object-based detailed vegetation classification with airborne high spatial resolution remote sensing imagery. Photogramm. Eng. Remote Sens. 72, 799-811.
245.Zhan, Q., Molenaar, M., Tempfli, K., Shi, W., 2005. Quality assessment for geo-spatial objects derived from remotely sensed data. Int. J. Remote Sens. 26, 2953-2974.
246.Zhang, D., Han, J., Cheng, G., Liu, Z., Bu, S., Guo, L., 2015a. Weakly supervised learning for target detection in remote sensing images. IEEE Geosci. Remote Sens. Lett. 12, 701-705.
247.Zhang, J., Lin, X., Liu, Z., Shen, J., 2011a. Semi-automatic road tracking by template matching and distance transformation in urban areas. Int. J. Remote Sens. 32, 8331-8347.
248.Zhang, L., Zhang, L., Tao, D., Huang, X., 2011b. A multifeature tensor for remote-sensing target recognition. IEEE Geosci. Remote Sens. Lett. 8, 374-378.
249.Zhang, L., Zhang, L., Tao, D., Huang, X., 2014a. Sparse transfer manifold embedding for hyperspectral target detection. IEEE Trans. Geosci. Remote Sens. 52, 1030-1043.
250.Zhang, Q., Couloigner, I., 2006. Benefit of the angular texture signature for the separation of parking lots and roads on high resolution multi-spectral imagery. Pattern Recog. Lett. 27, 937-946.
251.Zhang, W., Sun, X., Fu, K., Wang, C., Wang, H., 2014b. Object detection in high-resolution remote sensing images using rotation invariant parts based model. IEEE Geosci. Remote Sens. Lett. 11, 74-78.
252.Zhang, W., Sun, X., Wang, H., Fu, K., 2015b. A generic discriminative part-based model for geospatial object detection in optical remote sensing images. ISPRS J. Photogramm. Remote Sens. 99, 30-44.
253.Zhang, Y., Du, B., Zhang, L., 2015c. A sparse representation-based binary hypothesis model for target detection in hyperspectral images. IEEE Trans. Geosci. Remote Sens. 53, 1346-1354.
254.Zhang, Y., Zhang, L., Du, B., Wang, S., 2015d. A Nonlinear Sparse Representation-Based Binary Hypothesis Model for Hyperspectral Target Detection. IEEE J. Sel. Topics Appl. Earth Observ. Remote Sens. 8, 2513-2522.
255.Zhao, T., Nevatia, R., 2003. Car detection in low resolution aerial images. Image Vis. Comput. 21, 693-703.
256.Zhao, Y.-Q., Yang, J., 2015. Hyperspectral image denoising via sparse representation and low-rank constraint. IEEE Trans. Geosci. Remote Sens. 53, 296-308.
257.Zhen, Z., Quackenbush, L.J., Stehman, S.V., Zhang, L., 2013. Impact of training and validation sample selection on classification accuracy and accuracy assessment when using reference polygons in object-based classification. Int. J. Remote Sens. 34, 6914-6930.
258.Zheng, Z., Zhou, G., Wang, Y., Liu, Y., Li, X., Wang, X., Jiang, L., 2013. A novel vehicle detection method with high resolution highway aerial image. IEEE J. Sel. Topics Appl. Earth Observ. Remote Sens. 6, 2338-2343.
259.Zhong, P., Wang, R., 2007. A multiple conditional random fields ensemble model for urban area detection in remote sensing optical images. IEEE Trans. Geosci. Remote Sens. 45, 3978-3988.
260.Zhou, J., Bischof, W.F., Caelli, T., 2006. Road tracking in aerial images based on human–computer interaction and Bayesian filtering. ISPRS J. Photogramm. Remote Sens. 61, 108-124.
261.Zhou, P., Cheng, G., Liu, Z., Bu, S., Hu, X., 2015a. Weakly supervised target detection in remote sensing images based on transferred deep features and negative bootstrapping. Multidimensional Systems and Signal Processing, DOI: 10.1007/s11045-11015-10370-11043.
262.Zhou, P., Zhang, D., Cheng, G., Han, J., 2015b. Negative Bootstrapping for Weakly Supervised Target Detection in Remote Sensing Images. In: Proc. IEEE Int. Conf. Multimedia Big Data, pp. 318-323.
263.Zhou, W., 2013. An object-based approach for urban land cover classification: integrating LiDAR height and intensity data. IEEE Geosci. Remote Sens. Lett. 10, 928-931.
264.Zhou, W., Huang, G., Troy, A., Cadenasso, M.L., 2009. Object-based land cover classification of shaded areas in high spatial resolution imagery of urban areas: A comparison study. Remote Sens. Environ. 113, 1769-1777.
265.Zhou, W., Troy, A., 2008. An object-oriented approach for analysing and characterizing urban landscape at the parcel level. Int. J. Remote Sens. 29, 3119-3135.
266.Zhu, C., Shi, W., Pesaresi, M., Liu, L., Chen, X., King, B., 2005. The recognition of road network from high-resolution satellite remotely sensed data using image morphological characteristics. Int. J. Remote Sens. 26, 5493-5508.
267.Zhu, C., Zhou, H., Wang, R., Guo, J., 2010. A novel hierarchical method of ship detection from spaceborne optical image based on shape and texture features. IEEE Trans. Geosci. Remote Sens. 48, 3446-3456.
268.Zhu, F., Shao, L., 2014. Weakly-supervised cross-domain dictionary learning for visual recognition. Int. J. Comput. Vis. 109, 42-59.
269.Zhu, H., Basir, O., 2005. An adaptive fuzzy evidential nearest neighbor formulation for classifying remote sensing images. IEEE Trans. Geosci. Remote Sens. 43, 1874-1889.

Citation

MLA

Cheng, G., and J. Han. “A Survey on Object Detection in Optical Remote Sensing Images”. ISPRS Journal of Photogrammetry and Remote Sensing, vol. 117, 2016, pp. 11–28, https://doi.org/10.1016/j.isprsjprs.2016.03.014.

APA

Cheng, G., & Han, J. (2016). A survey on object detection in optical remote sensing images. ISPRS Journal of Photogrammetry and Remote Sensing, 117, 11–28. https://doi.org/10.1016/j.isprsjprs.2016.03.014

Chicago

Cheng, G., and J. Han. 2016. “A Survey on Object Detection in Optical Remote Sensing Images”. ISPRS Journal of Photogrammetry and Remote Sensing 117: 11–28. https://doi.org/10.1016/j.isprsjprs.2016.03.014.

Harvard

Cheng, G. and Han, J. (2016) “A survey on object detection in optical remote sensing images”, ISPRS Journal of Photogrammetry and Remote Sensing, 117, pp. 11–28. Available at: https://doi.org/10.1016/j.isprsjprs.2016.03.014.

Vancouver

1. Cheng G, Han J (2016) A survey on object detection in optical remote sensing images. ISPRS Journal of Photogrammetry and Remote Sensing 117:11–28

BibTeX

@article{Cheng_2016, title={A survey on object detection in optical remote sensing images}, volume={117}, ISSN={0924-2716}, url={http://dx.doi.org/10.1016/j.isprsjprs.2016.03.014}, DOI={10.1016/j.isprsjprs.2016.03.014}, journal={ISPRS Journal of Photogrammetry and Remote Sensing}, publisher={Elsevier BV}, author={Cheng, Gong and Han, Junwei}, year={2016}, month=July, pages={11–28} }

Metadata:Crossref

Access the Paper

This paper is available from its original source. Click below to access the PDF.

Open PDF

License: Authors

A Survey on Object Detection in Optical Remote Sensing Images

Abstract

Topics

Keywords

Table of Contents

Knowls

Knowl 1 — Taxonomy of Object Detection Methods in Optical Remote Sensing Images

Knowl 2 — Template Matching Frameworks in Remote Sensing

Rigid Template Matching

Deformable Template Matching

Knowl 3 — Knowledge-Based Hypothesis Generation and Context Verification

Knowl 4 — Object-Based Image Analysis (OBIA / GEOBIA) Paradigm

Segmentation and Multi-Resolution Segmentation (MRS)

Objective Scale Optimization

Segment Classification and Assessment

Knowl 5 — Machine Learning-Based Object Representation Features for RSIs

Knowl 6 — Sparse Representation and Sparse Representation-Based Classification (SRC)

Knowl 7 — Conditional Random Field (CRF) Formulation for Contextual Object Detection

Knowl 8 — Public Benchmark Datasets for Remote Sensing Object Detection

Knowl 9 — Standard Quantitative Evaluation Metrics for Geospatial Object Detection

Precision and Recall

Pixel-Level vs. Object-Level Evaluation Criteria

F-Measure (FβF_\betaFβ​)

Average Precision (AP)

Knowl 10 — Frontier Challenges: Deep Feature Learning and Weakly Supervised Learning in RSIs

1. Deep Learning-Based Feature Representation

2. Weakly Supervised Learning (WSL) for Geospatial Detection

References

Citation

BibTeX

Access the Paper

F-Measure ( $F_\beta$ )