Search

Deep learning (DL) methods usually need to collect a large amount of labeled data to extract deep features. However, due to the difficulty of obtaining numerous labeled data from synthetic aperture radar (SAR) images, unsupervised feature learning has been focused on SAR image processing. In this article, we propose a 3-D sparse model (3-DSM) to extract deep sparse features from SAR images in an unsupervised way. Concretely, 3-DSM learns the convolution kernels by minimizing the error between the features and the constructed sparse maps, without labeled samples. Thus, the discriminative features can be extracted in an unsupervised way by the learned convolution kernels and are able to capture the main structural information of SAR images. Furthermore, to the best of our knowledge, 3-DSM first specifies the sparsity of convolution kernels, with each convolution kernel exhibiting its independence from the others and the redundancy of convolution kernels being diminishing. It means that each convolution kernel extracts its unique structural features of SAR images. Consequently, in the feature extraction, 3-D sparsities have been specified, including width, height, and depth, with the acquisition of discriminative less-redundant features. The effectiveness of 3-DSM is demonstrated by the feature extraction and segmentation of the simulated and real SAR images. [ABSTRACT FROM AUTHOR]

The automatic extraction of lake water is one of the research hotspots in the field of remote sensing image processing. Due to the small interclass variance between lakes and other ground objects, and the complex texture characteristics of lake boundaries, existing methods often have problems such as over-segmentation and inaccurate boundary segmentation when segmenting lake water bodies. To alleviate these problems, this article designs an end-to-end semantic segmentation network [noise-canceling transformer network (NT-Net)] for the automatic extraction of lake water bodies from remote sensing images. Aiming at the problem of over-segmentation caused by nonlake objects, an interference attenuation module is designed in the network. This module can model the key features that are distinguishable and suitable for segmenting lake water by analyzing the difference in feature representation between lakes and other ground objects, thus suppressing the feature representation of nonlake objects. To more accurately segment the lake boundary, a multilevel transformer module is designed. This module can capture the context association of boundary information and enhance the feature representation of boundary information by using the self-attention mechanism. The comparative experimental results show that, compared with the current mainstream semantic segmentation networks, the method in this article has advantages in extracting lake water bodies comprehensively and coherently. [ABSTRACT FROM AUTHOR]

Recently, convolutional neural networks (CNNs) have achieved state-of-the-art performance in infrared small target detection. However, the limited number of public training data restricts the performance improvement of CNN-based methods. To handle the scarcity of training data, we propose a method that can generate synthetic training data for infrared small target detection. We adopt the generative adversarial network framework where synthetic background images and infrared small targets are generated in two independent processes. In the first stage, we synthesize infrared images by transforming visible images into infrared ones. In the second stage, target masks are implanted on the transformed images. Then, the proposed intensity modulation network synthesizes realistic target objects that can be diversely generated from further image processing. Experimental results on the recent public dataset show that, when we train various detection networks using the dataset composed of both real and synthetic images, detection networks yield better performance than using real data only. [ABSTRACT FROM AUTHOR]

Cloud and cloud shadow segmentation is one of the most important issues in remote sensing image processing. Most of the remote sensing images are very complicated. In this work, a dual-branch model composed of transformer and convolution network is proposed to extract semantic and spatial detail information of the image, respectively, to solve the problems of false detection and missed detection. To improve the model’s feature extraction, a mutual guidance module (MGM) is introduced, so that the transformer branch and the convolution branch can guide each other for feature mining. Finally, in view of the problem of rough segmentation boundary, this work uses different features extracted by the transformer branch and the convolution branch for decoding and repairs the rough segmentation boundary in the decoding part to make the segmentation boundary clearer. Experimental results on the Landsat-8, Sentinel-2 data, the public dataset high-resolution cloud cover validation dataset created by researchers at Wuhan University (HRC_WHU), and the public dataset Spatial Procedures for Automated Removal of Cloud and Shadow (SPARCS) demonstrate the effectiveness of our method and its superiority to the existing state-of-the-art cloud and cloud shadow segmentation approaches. [ABSTRACT FROM AUTHOR]

Synthetic aperture radar (SAR) change detection (CD) can be broadly classified into two categories: noncoherent intensity CD and coherent change detection (CCD). The former methods, belonging to the field of image processing, are theoretically suitable for all SAR images since they only utilize the information of SAR magnitude; however, the detection precision cannot be guaranteed. The latter methods can show effective performances for most SAR image pairs using identical collection geometrics because of the basis of the probability and statistics theory. In this article, we propose a novel change estimator to combine the advantages of the two kinds of algorithms in a theoretical derivation way. An intensity-based estimator, inspired by the derivation of the coherence estimator in CCD, is first proposed. It is a new maximum-likelihood (ML) change estimator maximizing the probability distribution function (pdf) of the ratio change statistic instead of SAR complex data. In addition, the simple linear iterative cluster (SLIC) algorithm is introduced to make the new estimator adjustable because changes at different degrees can be extracted with varying settings of the superpixels number, which is further demonstrated in real SAR images. Finally, experiments in SAR image pairs of different statistical characteristics show that the proposed estimator can yield higher contrast SAR CD images than the other five common change statistics and obtain better CD maps than the other four classic thresholding methods. [ABSTRACT FROM AUTHOR]

In this article, a method is presented for automatic scaling of the F-layer from ionograms based on an image processing technique for the extraction of curvilinear structures. The algorithm obtains the ordinary and extraordinary traces and determines the F2 critical frequency. The performance was tested using a wide data set of ionograms recorded by the Advanced Ionospheric Sounder/Istituto Nazionale di Geofisica e Vulcanologia (AIS/INGV) ionosonde located at Universidad Nacional de Tucumán, Tucuman, Argentina, and the results are compared with manual scaling and also with Autoscala method. Results from these tests show that the method is feasible and can be the seed for the development of a robust automated scaling system. [ABSTRACT FROM AUTHOR]

High/very-high-resolution (HR/VHR) multitemporal images are important in remote sensing to monitor the dynamics of the Earth’s surface. Unsupervised object-based image analysis provides an effective solution to analyze such images. Image semantic segmentation assigns pixel labels from meaningful object groups and has been extensively studied in the context of single-image analysis, however not explored for multitemporal one. In this article, we propose to extend supervised semantic segmentation to the unsupervised joint semantic segmentation of multitemporal images. We propose a novel method that processes multitemporal images by separately feeding to a deep network comprising of trainable convolutional layers. The training process does not involve any external label, and segmentation labels are obtained from the argmax classification of the final layer. A novel loss function is used to detect object segments from individual images as well as establish a correspondence between distinct multitemporal segments. Multitemporal semantic labels and weights of the trainable layers are jointly optimized in iterations. We tested the method on three different HR/VHR data sets from Munich, Paris, and Trento, which shows the method to be effective. We further extended the proposed joint segmentation method for change detection (CD) and tested on a VHR multisensor data set from Trento. [ABSTRACT FROM AUTHOR]

This article deals with the problem of computing surface concentrations for two types of river ice from digital images acquired during freeze-up. It presents the results of attempting to solve this problem using several state-of-the-art semantic segmentation methods based on deep convolutional neural networks (CNNs). This task presents two main challenges—very limited availability of labeled training data and presence of noisy labels due to the great difficulty of visually distinguishing between the two types of ice, even for human experts. The results are used to analyze the extent to which some of the best deep learning methods currently in existence can handle these challenges. The code and data used in the experiments are made publicly available to facilitate further work in this domain. [ABSTRACT FROM AUTHOR]

With an increasing number of remote sensing images (RSIs), the automatic region of interest (ROI) extraction based on convolutional neural networks (CNNs) has attracted much interest in recent years. Although fully supervised CNN-based methods have shown superiority in the field of object extraction, pixelwise annotations are expensive and time-consuming. Moreover, due to the unstable distribution of ROIs in complex landscapes, the ratios of foreground and background areas are quite different in RSIs. Training CNNs with such unbalanced data sets lead to over-fitting and low accuracy. In this article, we propose a framework that combines multiview learning and attention mechanism (MLAM) to solve the above mentioned problems. First, we develop a CNN-based weakly supervised method with a weight-balanced loss function to solve the problems caused by an unbalanced data set. It also helps to generate imagewise saliency maps by computing the gradient maps with respect to the input images. Then, we design a multiview strategy to dramatically reduce the missing inspection. Finally, we design a feedback attention mechanism based on the stage neighbor binary pattern to further modify the extraction result. In summary, the proposed framework achieves pixelwise ROI extraction under imagewise annotations through data-driven ML and a knowledge-driven visual attention mechanism. We evaluate the performance of the MLAM framework on two challenging data sets with complex backgrounds. The experimental results indicate that the proposed framework can achieve better performance than other eight ROI extraction models for unbalanced remote sensing data sets. [ABSTRACT FROM AUTHOR]

The whole-sky imager has been increasingly used for ground-based cloud automatic observation. Many approaches based on image processing have been applied to detect or classify clouds in whole-sky images (WSIs). However, most of the studies only focus on image segmentation for cloud detection or image classification for cloud recognition separately. The cloud detection only does the binary segmentation (sky and cloud) without cloud types, while the cloud recognition only gives the single image-level label without cloud coverage. In this paper, a fine-grained cloud detection and recognition task with a solution is proposed to fill the gap, which can simultaneously detect and classify clouds in a WSI. It can be regarded as a pixel-level fine-grained dense prediction for images. First, a new data set is built with pixel-level annotation of nine different types. Then, a solution based on supervised learning is proposed, in which the pixel-level prediction problem is converted to a superpixel classification problem. Multiview features are extracted, including color, inside texture, neighbor texture, and global relation, to represent the superpixels. Moreover, a class-specific feature space transformation method based on metric learning and subspace alignment is proposed to overcome the challenge brought by the high similarity among cloud types and the feature shifting. Finally, several experiments have verified that our approach is effective to the challenging new task and also outperforms some other methods in the normal tasks of cloud detection and cloud classification, respectively. [ABSTRACT FROM AUTHOR]

Semantic segmentation requires methods capable of learning high-level features while dealing with large volume of data. Toward such goal, convolutional networks can learn specific and adaptable features based on the data. However, these networks are not capable of processing a whole remote sensing image, given its huge size. To overcome such limitation, the image is processed using fixed size patches. The definition of the input patch size is usually performed empirically (evaluating several sizes) or imposed (by network constraint). Both strategies suffer from drawbacks and could not lead to the best patch size. To alleviate this problem, several works exploited multicontext information by combining networks or layers. This process increases the number of parameters, resulting in a more difficult model to train. In this paper, we propose a novel technique to perform semantic segmentation of remote sensing images that exploits a multicontext paradigm without increasing the number of parameters while defining, in training time, the best patch size. The main idea is to train a dilated network with distinct patch sizes, allowing it to capture multicontext characteristics from heterogeneous contexts. While processing these varying patches, the network provides a score for each patch size, helping in the definition of the best size for the current scenario. A systematic evaluation of the proposed algorithm is conducted using four high-resolution remote sensing data sets with very distinct properties. Our results show that the proposed algorithm provides improvements in pixelwise classification accuracy when compared to the state-of-the-art methods. [ABSTRACT FROM AUTHOR]

When a seismometer network records an earthquake, operators will manually review the waveforms and identify the wave phases, a task known as phase picking. Manual phase picking is a time-consuming process that can be automated using machine learning; however, automatic methods have not yet achieved human-level performance, and open-source implementations of state-of-the-art algorithms are not always available. Convolutional networks have revolutionized the field of image processing, where the large amounts of readily available data make possible near-human performance in tasks such as classification and segmentation. Fortunately, phase picking is also an area where thousands of phases are manually picked, which makes convolutional networks a good fit for the processing of this type of data. In this paper, we describe Cospy, an open-source convolutional phase picker that uses a two-stage analysis in which the first stage segments a rough area around the phase, and the second stage regresses the precise location. Our approach was evaluated on the Northern California Earthquake Data Center (NCEDC) data set and, when targeting picks closer than 0.1 s, it achieved an $F_{1}$ -score of 93.13% for P phases and 91.07% for S phases. Our results show that convolutional networks are on track to achieve human-level performance on the task of seismic phase picking and can contribute to decreasing the need for manual analysis. An open-source implementation of the proposed approach, pretrained on the NCEDC data set, can be downloaded at https://github.com/stbnps/cospy. [ABSTRACT FROM AUTHOR]

Image segmentation techniques are challenging to apply to large-size remote sensing imagery. Indeed, if the data to be processed are larger than the computer’s available memory, it must be split into smaller pieces. Without precaution, segmentation errors appear along the edges of these pieces. The goal of this paper is to present a tilewise processing method to overcome this issue for superpixel segmentation, applied in particular to the simple linear iterative clustering algorithm. Incidentally, tilewise methods allow for several pieces of the image to be processed simultaneously, which enables the deployment of these methods in a parallel processing environment. Estimations of the speed-up when using multiple processors are provided. Then, it is demonstrated that the result of the tilewise segmentation is equivalent to the segmentation of the complete image, with respect to a number of global unsupervised segmentation criteria. Finally, experimental results on a large-size Sentinel-2 time series validate the method’s feasibility. [ABSTRACT FROM AUTHOR]

It is a classical task to automatically extract road networks from very high-resolution (VHR) images in remote sensing. This paper presents a novel method for extracting road networks from VHR remotely sensed images in complex urban scenes. Inspired by image segmentation, edge detection, and object skeleton extraction, we develop a multitask convolutional neural network (CNN), called RoadNet, to simultaneously predict road surfaces, edges, and centerlines, which is the first work in such field. The RoadNet solves seven important issues in this vision problem: 1) automatically learning multiscale and multilevel features [gained by the deeply supervised nets (DSN) providing integrated direct supervision] to cope with the roads in various scenes and scales; 2) holistically training the mentioned tasks in a cascaded end-to-end CNN model; 3) correlating the predictions of road surfaces, edges, and centerlines in a network model to improve the multitask prediction; 4) designing elaborate architecture and loss function, by which the well-trained model produces approximately single-pixel width road edges/centerlines without nonmaximum suppression postprocessing; 5) cropping and bilinear blending to deal with the large VHR images with finite-computing resources; 6) introducing rough and simple user interaction to obtain desired predictions in the challenging regions; and 7) establishing a benchmark data set which consists of a series of VHR remote sensing images with pixelwise annotation. Different from the previous works, we pay more attention to the challenging situations, in which there are lots of shadows and occlusions along the road regions. Experimental results on two benchmark data sets show the superiority of our proposed approaches. [ABSTRACT FROM AUTHOR]

Since it is usually difficult and time-consuming to obtain sufficient training samples by manually labeling, feature extraction, which investigates the characteristics of hyperspectral images (HSIs), such as spectral continuity and spatial locality of surface objects, to achieve the most discriminative feature representation, is very important for HSI classification. Meanwhile, due to the spatial regularity of surface materials, it is desirable to improve the classification performance of HSIs from the superpixel viewpoint. In this paper, we propose a novel local binary pattern (LBP)-based superpixel-level decision fusion method for HSI classification. The proposed framework employs uniform LBP (ULBP) to extract local image features, and then, a support vector machine is utilized to formulate the probability description of each pixel belonging to every class. The composite image of the first three components extracted by a principal component analysis from the HSI data is oversegmented into many homogeneous regions by using the entropy rate segmentation method. Then, a region merging process is applied to make the superpixels obtained more homogeneous and agree with the spatial structure of materials more precisely. Finally, a probability-oriented classification strategy is applied to classify each pixel based on superpixel-level guidance. The proposed framework "ULBP-based superpixel-level decision fusion framework" is named ULBP-SPG. Experimental results on two real HSI data sets have demonstrated that the proposed ULBP-SPG framework is more effective and powerful than several state-of the- art methods. [ABSTRACT FROM AUTHOR]

The problem of limited labeled training samples is challenging for the classification of remote sensing imagery. We develop a joint classification and segmentation algorithm to address this problem. Our algorithm combines semisupervised learning and conditional random fields (CRFs) into a single framework. The multimodal Gaussian maximum-likelihood classifier is used to estimate the probabilities for the unary potentials of the CRF. Unlike traditional methods based on random fields, region merging is concatenated with the CRF inference to reduce the number of nodes iteratively. Moreover, a semisupervised technique called self-training is used, which iteratively enlarges the training sample set and retrains the classifier. The selection of training samples is based on the region information, so that the risk of assigning wrong labels is largely reduced. The proposed algorithm is applied to hyperspectral image classification, and results on benchmark data sets show that the proposed algorithm significantly improves classification performance after using self-training, and outperforms state-of-the-art spectral-spatial methods for limited labeled training samples. [ABSTRACT FROM AUTHOR]

Simple linear iterative clustering (SLIC) algorithm was proposed for superpixel generation on optical images and showed promising performance. Several studies have been proposed to modify SLIC to make it applicable for polarimetric synthetic aperture radar (PolSAR) images, where the Wishart distance is adopted as the similarity measure. However, the superpixel segmentation results of these methods were not satisfactory in heterogeneous urban areas. Further, it is difficult to determine the tradeoff factor which controls the relative weight between polarimetric similarity and spatial proximity. In this research, an adaptive polarimetric SLIC (Pol-ASLIC) superpixel generation method is proposed to overcome these limitations. First, the spherically invariant random vector (SIRV) product model is adopted to estimate the normalized covariance matrix and texture for each pixel. A new edge detector is then utilized to extract PolSAR image edges for the initialization of central seeds. In the local iterative clustering, multiple cues including polarimetric, texture, and spatial information are considered to define the similarity measure. Moreover, a polarimetric homogeneity measurement is used to automatically determine the tradeoff factor, which can vary from homogeneous areas to heterogeneous areas. Finally, the SLIC superpixel generation scheme is applied to the airborne Experimental SAR and PiSAR L-band PolSAR data to demonstrate the effectiveness of this proposed superpixel generation approach. This proposed algorithm produces compact superpixels which can well adhere to image boundaries in both natural and urban areas. The detail information in heterogeneous areas can be well preserved. [ABSTRACT FROM PUBLISHER]

In this paper, we propose a new unsupervised change detection method designed to analyze multispectral remotely sensed image pairs. It is formulated as a segmentation problem to discriminate the changed class from the unchanged class in the difference images. The proposed method is in the category of the committee machine learning model that utilizes an ensemble of classifiers (i.e., the set of segmentation results obtained by several thresholding methods) with a dynamic structure type. More specifically, in order to obtain the final “change/no-change” output, the responses of several classifiers are combined by means of a mechanism that involves the input data (the difference image) under an iterative Bayesian–Markovian framework. The proposed method is evaluated and compared to previously published results using satellite imagery. [ABSTRACT FROM PUBLISHER]

Superpixels have been widely used in lots of computer vision and image processing tasks but rarely used in topographic map processing due to the complex distribution of geographic elements in this kind of images. We propose a novel superpixel-generating method based on guided watershed transform (GWT). Before GWT, the cues of geographic element distribution and boundaries between different elements need to be obtained. A linear feature extraction method based on a compound opposite Gaussian filter and a shear transform is presented to acquire the distribution information. Meanwhile, a boundary detection method, which based on the color-opponent mechanisms of the visual system, is employed to get the boundary information. Then, both linear features and boundaries are input to the final partition procedure to obtain superpixels. The experiments show that our method has the best performance in shape control, size control, and boundary adherence, among all the comparison methods, which are classic and state of the art. Furthermore, we verify the low complexity and low cost of memory in our method through experiments, which makes it possible to deal with large-scale topographic maps. [ABSTRACT FROM PUBLISHER]

A synthetic aperture radar (SAR) imaging system usually produces pairs of bright area and dark area when depicting the ground objects, such as a building or tree and its shadow. Many buildings (trees) are aggregated together to form urban areas (forests). It means that the pairs of bright and dark areas often exist in the aggregated scenes. Conventional unsupervised segmentation approaches usually segment the scenes (e.g., urban areas and forests) into different regions simply according to the gray values of the image. However, a more convincing way is to regard them as the consistent regions. In this paper, we aim at addressing this issue and propose a new SAR image segmentation approach via a hierarchical visual semantic and adaptive neighborhood multinomial latent model. In this approach, the hierarchical visual semantic of SAR images is proposed, which divides SAR images into aggregated, structural, and homogeneous regions. Based on the division, different segmentation methods are chosen for these regions with different characteristics. For the aggregated region, locality-constrained linear coding-based hierarchical clustering is used for segmentation. For the structural region, visual semantic rules are designed for line object location, and a geometric structure window-based multinomial latent model is proposed for segmentation. For the homogeneous region, a multinomial latent model with adaptive window selection is proposed for segmentation. Finally, these results are integrated together to obtain the final segmentation. Experiments on both synthetic and real SAR images indicate that the proposed method achieves promising performances in terms of the consistencies of the regions and the preservations of the edges and line objects. [ABSTRACT FROM AUTHOR]

Processing large very high-resolution remote sensing images on resource-constrained devices is a challenging task because of the large size of these data sets. For applications such as environmental monitoring or natural resources management, complex algorithms have to be used to extract information from the images. The memory required to store the images and the data structures of such algorithms may be very high (hundreds of gigabytes) and therefore leads to unfeasibility on commonly available computers. Segmentation algorithms constitute an essential step for the extraction of objects of interest in a scene and will be the topic of the investigation in this paper. The objective of the present work is to adapt image segmentation algorithms for large amounts of data. To overcome the memory issue, large images are usually divided into smaller image tiles, which are processed independently. Region-merging algorithms do not cope well with image tiling since artifacts are present on the tile edges in the final result due to the incoherencies of the regions across the tiles. In this paper, we propose a scalable tile-based framework for region-merging algorithms to segment large images, while ensuring identical results, with respect to processing the whole image at once. We introduce the original concept of the stability margin for a tile. It allows ensuring identical results to those obtained if the whole image had been segmented without tiling. Finally, we discuss the benefits of this framework and demonstrate the scalability of this approach by applying it to real large images. [ABSTRACT FROM PUBLISHER]

The conditional random field (CRF) model is suitable for the image segmentation because this model relaxes the assumption of conditional independence of the observed data and models the data-dependent label interaction in the image modeling. However, this model has a limited ability to capture the global and local image information from the perspective of multiresolution analysis. Moreover, for synthetic aperture radar (SAR) image segmentation, SAR scattering statistics that are essential to SAR image processing are not considered in the CRF model. In this paper, we propose a hierarchical CRF (HIECRF) model for SAR image segmentation. The HIECRF model belongs to the discriminative models according to the semantic structure. While inheriting the advantages of the CRF model, the HIECRF model achieves the integration of the image features and SAR scattering statistics and captures the contextual structure information in the spatial and scale spaces. Moreover, we derive a hierarchical inference algorithm for the HIECRF model in virtue of the mean-field approximation (MFA) to provide the maximization of the posterior marginal (MPM) estimate of the HIECRF model. Then, by the bottom-up and the top-down recursions in the hierarchical inference procedure, the HIECRF model effectively exploits the global and local image information, including the contextual structures, the image features, and the scattering statistics, to achieve the MPM segmentation. The effectiveness of the HIECRF model is demonstrated by the application to the semisupervised segmentation of the simulated images and the real SAR images. [ABSTRACT FROM PUBLISHER]

Just over a decade has passed since the concept of morphological profile was defined for the analysis of remote sensing images. Since then, the morphological profile has largely proved to be a powerful tool able to model spatial information (e.g., contextual relations) of the image. However, due to the shortcomings of using the morphological profiles, many variants, extensions, and refinements of its definition have appeared stating that the morphological profile is still under continuous development. In this case, recently introduced theoretically sound attribute profiles (APs) can be considered as a generalization of the morphological profile, which is a powerful tool to model spatial information existing in the scene. Although the concept of the AP has been introduced in remote sensing only recently, an extensive literature on its use in different applications and on different types of data has appeared. To that end, the great amount of contributions in the literature that address the application of the AP to many tasks (e.g., classification, object detection, segmentation, change detection, etc.) and to different types of images (e.g., panchromatic, multispectral, and hyperspectral) proves how the AP is an effective and modern tool. The main objective of this survey paper is to recall the concept of the APs along with all its modifications and generalizations with special emphasis on remote sensing image classification and summarize the important aspects of its efficient utilization while also listing potential future works. [ABSTRACT FROM AUTHOR]

Segmentation of real-world remote sensing images is challenging because of the large size of those data, particularly for very high resolution imagery. However, a lot of high-level remote sensing methods rely on segmentation at some point and are therefore difficult to assess at full image scale, for real remote sensing applications. In this paper, we define a new property called stability of segmentation algorithms and demonstrate that piece- or tile-wise computation of a stable segmentation algorithm can be achieved with identical results with respect to processing the whole image at once. We also derive a technique to empirically estimate the stability of a given segmentation algorithm and apply it to four different algorithms. Among those algorithms, the mean-shift algorithm is found to be quite unstable. We propose a modified version of this algorithm enforcing its stability and thus allowing for tile-wise computation with identical results. Finally, we present results of this method and discuss the various trends and applications. [ABSTRACT FROM PUBLISHER]

Object detection is an important task in very high-resolution remote sensing image analysis. Traditional detection approaches are often not sufficiently robust in dealing with the variations of targets and sometimes suffer from limited training samples. In this paper, we tackle these two problems by proposing a novel method for object detection based on structural feature description and query expansion. The feature description combines both local and global information of objects. After initial feature extraction from a query image and representative samples, these descriptors are updated through an augmentation process to better describe the object of interest. The object detection step is implemented using a ranking support vector machine (SVM), which converts the detection task to a ranking query task. The ranking SVM is first trained on a small subset of training data with samples automatically ranked based on similarities to the query image. Then, a novel query expansion method is introduced to update the initial object model by active learning with human inputs on ranking of image pairs. Once the query expansion process is completed, which is determined by measuring entropy changes, the model is then applied to the whole target data set in which objects in different classes shall be detected. We evaluate the proposed method on high-resolution satellite images and demonstrate its clear advantages over several other object detection methods. [ABSTRACT FROM PUBLISHER]