Search

An approach of weighted Wishart distance learning, shorted for W2-based distance learning, is proposed for polarimetric synthetic aperture radar (PolSAR) image classification. It aims to adjust the Wishart distance by enhancing discrimination as well as exploiting spatial information. The proposed distance learning keeps samples within the same category close and separates samples from the different classes far apart. It is effectively implemented by solving a linear programming. Input of W2-based distance learning is called weighted Wishart feature, which is designed specifically for PolSAR data to describe the Wishart distribution, achieve regional consistency, and reduce speckle noise. Weight is calculated according to an adaptive window, where homogeneous samples are derived based on a connected region and extracted edge information. With this feature, W2-based distance learning is a whole scheme to adjust the Wishart distance. Furthermore, our experiments with benchmark data sets suggest that the proposed scheme provides both improved performance in terms of visual effect and classification accuracy. The achieved overall accuracy is better by more than 7% compared to other state-of-art methods. [ABSTRACT FROM PUBLISHER]

Spectral–spatial-based classification methods demonstrate satisfying performance for hyperspectral imagery (HSI) classification. In this letter, in order to make full use of spectral and contexture information with simultaneously considering within-class information, we propose a new algorithm for HSI classification based on within-class collaborative representation and column generation (CG) strategy. The proposed accelerated homogeneous patch mean kernel (HPMK) can automatically assign a homogeneous patch for the target sample and represent the similarities between training set and assigned homogeneous patch in kernel feature space based on CG strategy. Further, for including intra-class information and improve classification efficiency, within-class collaborative representation classification (WCRC) is incorporated into new feature space to enhance the classification performance. Experiments on two real HSI data sets demonstrate that the proposed algorithm presents satisfying results in terms of classification accuracy and efficiency. [ABSTRACT FROM PUBLISHER]

Sparse representation-based classification (SRC) and collaborative representation-based classification (CRC) have been proposed for image classification. Sparsity is overemphasized in SRC, but ignored in CRC, which has low representation complexity. In this article, a new local collaborative representation scheme is proposed for the classification of hyperspectral imagery (HSI). First, for a test sample with an unknown label, a few significant atoms are selected based on their contributions in representing the test sample, where the-norm is replaced by the-norm like CRC to reduce the complexity. Next, the test sample is represented by the active atoms. Finally, the class label of the test sample is determined based on the contribution of each class of active atoms. Additionally, in order to improve the robustness of the active atoms, both the test sample and the samples in its neighbourhood as context information are considered in the selection of active atoms and the class determination. Experimental results on three real HSI data sets confirm the effectiveness and accuracy of the proposed method. [ABSTRACT FROM AUTHOR]

An image segmentation method based on optimized spatial texture information is proposed in this article. Spatial information, including the relative position of neighbouring pixels and texture features of the multiscale neighbourhood, is incorporated into the similarity measure of the fuzzy c-means (FCM) clustering algorithm, in which the Gaussian kernel is adopted to diminish the local incorrect segmentation. The FCM clustering is spatially adjusted and optimized by the particle swarm optimization (PSO) algorithm. The purpose of optimization is to obtain the appropriate control parameters influencing spatial information, which can improve segmentation results. Experimental results demonstrate that the proposed method achieves better segmentation performance and is capable of effectively segmenting synthetic images and synthetic aperture radar (SAR) images. [ABSTRACT FROM PUBLISHER]

This letter presents a novel unsupervised change detection approach for multi-temporal synthetic aperture radar (SAR) image. The proposed method applies the semi-supervised Laplacian support vector machine (SVM) to classify the changed areas and the unchanged areas. A pseudo-training set, which is necessary for initializing the SVM, is generated according to the saliency similarity detection. The pseudo-training set is composed by the labelled changed pixels and the labelled unchanged pixels. The Laplacian SVM explores the prior information of the available labelled samples and combines unlabelled samples to enhance its discrimination. Experimental results on several real SAR image data sets confirm the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

Hyperspectral imagery (HSI) classification is a rapidly growing and highly active research area in the field of hyperspectral community. The method that combines both spatial and spectral information for hyperspectral image classification has made a great advance. The focus of spectral-spatial classification is how to extract discriminating features and how to combine the spectral and spatial information effectively. In this paper, we propose a new spectral-spatial method to solve these two problems. The first part is a series of support vector machines (SVMs) that are cascaded to form the enhanced features, where the predicted information of the preceding layer can provide correction information for the subsequent layer, and then a new superpixel segmentation method is adopted to introduce multi-scale spatial information, which can generate multi-scale superpixels at one time and is as accurate as the state-of-the-art methods. The final label is determined by majority voting of different scales. [ABSTRACT FROM AUTHOR]

Object detection has played an important role in the remote sensing (RS) image understanding domain, but it still suffers from challenges, including the complexity of the image background, the variety of objects and differences of target scales. Although many studies applying deep learning methods to object detection have been presented, effective methods of multiscale object detection in RS images are still lacking. In this paper, we propose a novel object detection approach based on the Multiscale-SELU-DenseNet (MSE-DenseNet) and the dynamic anchor assignment (DAA) strategy. We fuse a highly modified DenseNet and a feature pyramid network into the MSE-DenseNet to extract multi-level feature descriptions of RS images, and then we use them to detect objects on various scales. To further improve the detection performance, we propose a DAA strategy for assigning prior anchor boxes. The experimental results show that our proposed method outperforms state-of-the-art detection methods on the VHR-10 dataset and maintains a real-time detection speed. [ABSTRACT FROM AUTHOR]

This paper proposes a new algorithm, for polarimetric synthetic aperture radar (PolSAR) classification, based on a stacked auto-encoder and scattering energy. Previous approaches to PolSAR classification predominantly consider only the single pixel of distribution of the polarimetric data and scattering characteristics, and ignore other kinds of image features like the relationship of the local pixels. Besides, because of the complexities of PolSAR data, it is difficult to compute the derivatives that are needed for back-propagation in deep-learning classifiers. To overcome these difficulties, we propose a new approach that combines the scattering power and stacks sparse auto-encoder (Scattering SSAE) for PolSAR classification. Firstly, orientation compensation is used to compensate the polarization orientation angle, reducing the impact of polarimetric angle noise. Secondly, Freeman-Durden decomposition is adopted to extract three basic scattering powers: surface, double bounce and volume. Each PolSAR image pixel is transformed into these scattering powers, yielding a new kind of feature from the PolSAR data. Finally, using the three kinds of scattering power as inputs, we combine local spatial information using a patch-based approach, and use a deep learning architecture to achieve classification. We compare our method against several other state-of-the-art methods using ground-truthed test-data, and show that the Scattering SSAE method achieves higher accuracy than other methods on most categories. [ABSTRACT FROM AUTHOR]

In very recent years, deep learning based methods have been widely introduced for the classification of hyperspectral images (HSI). However, these deep models need lots of training samples to tune abundant parameters which induce a heavy computation burden. Therefore, most of these algorithms need to be accelerated with high-performance graphics processing units (GPU). In this paper, a new deep model-densely connected deep random forest (DCDRF) is proposed to classify the hyperspectral images. This model is composed of multiple forward connected random forests. The DCDRF has following merits: 1) It obtains satisfactory classification accuracy with a small number of training samples, 2) It can be run efficiently on the central processing unit (CPU), 3) Only a few parameters are involved during the training. Experimental results based on three hyperspectral images demonstrate that the proposed method can achieve better classification performance than the conventional deep learning based methods. [ABSTRACT FROM AUTHOR]

In polarimetric synthetic aperture radar (PolSAR) image processing, the number of classes is an important factor for PolSAR image classification. Therefore, how to accurately estimate the number of PolSAR image classes is an important issue. In this article, we propose a novel unsupervised classification method which can accurately estimate the number of classes for PolSAR images. First, the PolSAR image is initialized into many small clusters by using the complementary information from Yamaguchi decomposition and distribution characteristics of data. Second, the improved clustering by fast search and find of density peaks, named as improved CFSFDP algorithm, is introduced to select the appropriate category number. Finally, to improve the representation of each category, the PolSAR data set is classified by an iterative fine-tuning process based on a complex K-Wishart function. The performance of the proposed classification approach is presented and analysed on three real data sets. The experimental results show that the proposed classification method can accurately estimate the category number and enhance the classification accuracy in comparison with other traditional methods. It is also shown that the data distribution characteristic has the additional information beyond the target scattering decomposition, and this information is important for the initialization. [ABSTRACT FROM AUTHOR]

It is well known that various features extraction approaches are utilized in polarimetric synthetic aperture (PolSAR) terrain classification for representing the data characteristic. It needs relevant and effective feature fusion algorithms to process complicated features. To address this issue, this article presents a multimodal sparse representation (MSR) framework based algorithm to fuse the different feature vectors from the complicated data space. Polarimetric data features, decomposition features, and the texture features from Pauli colour-coded image are selected to represent multimodal data in different observation modes. The corresponding multimodal manifold regularizations are added to MSR framework to approximate the data structure. Considering the independence and correlation of features, the intrinsic affinity matrices are calculated from this framework. They are processed via local preserve projection algorithm to project the multimodal features into a low dimensionally intrinsic feature space for subsequent classification. Three datasets are utilized in experiments, Western Xi'an, Flevoland, and San Francisco Bay datasets from the Radarsat-2 system in C-band. The effect of regularization parameters and different dimensional fused features are analysed in visualization and quantitation performance. The experiment results demonstrate that the effectiveness and validity of proposed method are superior to other state-of-art methods. [ABSTRACT FROM AUTHOR]

Band selection is widely used to identify relevant bands for land-cover classification of hyperspectral images. The combination of spectral and spatial information can improve the classification performance of hyperspectral images dramatically. Similarly, the fusion of spectral-spatial information should also improve the performance of band selection. In this article, two semi-supervised wrapper-based spectral-spatial band selection algorithms are proposed. The local spatial smoothness of hyperspectral imagery is used to improve the performance of band selection when limited labelled samples available. With superpixel segmentation, the first algorithm uses the statistical characteristics of classification map to predict the classification quality of all samples. Based on the Markov random field model, the second algorithm incorporates the spatial information by the minimization of spectral-spatial energy function. Four widely used real hyperspectral data sets are used to demonstrate the effectiveness of the proposed methods, when compared to cross-validation-based wrapper method, the accuracy is improved by 2% for different data sets. [ABSTRACT FROM AUTHOR]

Hyperspectral sensors often collect hundreds of bands at a time, so hyperspectral images can accurately characterize different land-cover types with abundant spectral information. However, these spectral bands also contain redundant information that needs to be removed. Band selection is one of the most widely used methods to remove noised or redundant bands. Because labelled samples are difficult to collect, most band selection methods adopt unsupervised ways to select diverse and representative bands. Still, noised bands are often selected because they usually have low correlation with other bands. In this article, objective image quality assessment is introduced to indicate the quality of every band, and combined with the redundancy measure, a new unsupervised band selection method is proposed. Three real hyperspectral images are used to demonstrate the effectiveness of the proposed algorithm. [ABSTRACT FROM PUBLISHER]

Spatial information has been widely used for hyperspectral image classification, which can dramatically improve the classification accuracy. Though band selection is an important pre-processing step for hyperspectral image processing, spatial information has not been well exploited in this field. In this article, we will exploit the spatial information for band selection. This article mainly includes two parts: algorithm design, and algorithm evaluation. In the first part, we propose an efficient band selection method by using the spatial structure information and spectral information. In the second part, we advocate the use of the local spatial filtering and the spectral-spatial classifier for evaluating the performance of band selection algorithms instead of the traditional pixel-wise classifiers. Comprehensive experiments over diverse publicly available benchmark data sets reveal some interesting results. [ABSTRACT FROM AUTHOR]

This study clarifies the implicit potential deficiency caused by the sparse cardinality parameterkin Rong et al. (2014). In addition,k = β × W × M × N(0.9 ≤ β < 1) is suggested to avert this potential deficiency, whereβis a ratio controlling the amount of sparse cardinality,Wis the number of multispectral bands andM × Nis the size of panchromatic image. With the choice ofksuggested in this study, the low rank matrixLand sparse matrixSobtained by Go Decomposition (Zhou and Tao 2011) can be iteratively optimized and solved. Thus, instead of choosingkasW × M × Nin Rong et al. (2014), the potential deficiency thatLis directly obtained as an analytic solution can be averted. [ABSTRACT FROM AUTHOR]

High-frequency injection (HFI)-based methods are proved to be powerful in pansharpening multispectral (MS) images. In this article, based on one of the low-rank and sparse (LRS) decomposition algorithms, i.e. Go Decomposition (GoDec), a HFI-based pansharpening method exploiting spatial structure sharpness of both MS images and a low-frequency panchromatic (PAN) image component is proposed. The spectral and spatial measure of local perceived sharpness (S3) is employed to estimate sharpness of the corresponding MS and low-frequency PAN component blocks, and the sharper one is used to construct the spatial structure of the sharpened MS images. Experimental results with QuickBird, IKONOS and WorldView-2 data demonstrate that the proposed method is comparable with or even better than other popular methods. [ABSTRACT FROM AUTHOR]

The Yellow River Estuary area of China is under great pressure from both human intervention and natural processes. For analysis of the changes in this area, this article presents a novel change-detection method based on a local fit-search model and kernel-induced graph cuts in multitemporal synthetic aperture radar images. Change detection involves assigning a label to every pixel. This task is naturally formulated in terms of energy minimization, which can be effectively solved by graph cuts. The difference image is transformed implicitly by a kernel function so that an alternative to complex modelling of the original data makes the piecewise constant model become applicable for graph cuts formulation. An issue is that graph cuts are sensitive to the initial estimate. The local fit-search model is proposed to approximate to the local histogram while selecting an optimal threshold for the initial labelling, which leads to an effective constraint for graph cuts and computational benefits as well. Visual and quantitative analyses obtained on the Yellow River Estuary data set confirm the effectiveness of the proposed method and that it outperforms the other state-of-the-art methods of change detection. [ABSTRACT FROM AUTHOR]

Change detection in synthetic aperture radar (SAR) images has become increasingly important along with the development of SAR techniques. In this study, a novel locally fitting and expectation-maximization (EM) approach is proposed for unsupervised change detection tasks in SAR images. A difference image is generated first, and an in-depth study of the inherent characteristics of the histogram of the difference image is then made. Thus, the new approach is proposed corresponding to these characteristics. In this approach, the locally fitting model orientated to deal with the unchanged class is put forward to reach a best fit, and the semi-EM algorithm for the changed class is used to tackle the phenomenon of overlapping. Then, through the Bayesian decision rule, the optimal threshold is determined. Our contributions lie in two aspects. First, in the locally fitting model, the location of the optimal threshold can be determined, which leads to an accurate fit over a short interval. Second, the use of the semi-EM algorithm not only retains the efficacy of the EM algorithm to cope with overlapping for the changed class but also simplifies the computing process. Experiments on real data sets confirm the effectiveness of the proposed approach, which results in final maps very similar to the ground truth and is more effective in determining the optimal threshold in comparison with others. The experimental results also demonstrate its effectiveness when the changed areas are of different geometrical shapes. [ABSTRACT FROM PUBLISHER]