Synthetic aperture radar (SAR) interferometry is a high-resolution microwave remote sensing imaging method. Over the past two decades, many researchers working on remote sensing have applied this technology in various disciplines, including environmental monitoring, disaster monitoring, and elevation mapping. However, due to the existence of many influencing factors in the acquisition stage, such as atmospheric humidity and temperature, the reflected wave signals from the ground will be disturbed when received by remote sensing satellites. The presence of noise in interferograms is inevitable. Therefore, the accuracy of interferometric SAR phase denoising and coherence estimation has a decisive impact on the validity of subsequent processing results. In this paper, we pioneer the use of a nested U-net as a feature extractor for interferometric SAR phase and coherence. In addition, we build a phase filter and a coherence estimator by using the residual learning module. With the aim of determining the unique non-local similarity of InSAR images, we use non-local convolution and channel attention mechanisms to extract features in different dimensions of the interferogram. Through quantitative and qualitative experiments, the proposed method performs better in phase denoising and coherence estimation than state-of-the-art methods.

Synthetic Aperture Radar Interferometry (InSAR) has grown significantly over the past few decades, which were mainly used in remote sensing applications. Most InSAR applications (e.g., terrain mapping and monitoring) utilized a key technique called phase unwrapping Phase unwrapping obtained the absolute phase from the wrapped phase for the subsequent application. However, the collected wrapped phase inevitably contains noise due to the influence of factors such as atmosphere and temperature in the InSAR acquisition stage. This noise made it challenging to obtain the absolute phase from the wrapped phase. This study proposed a deep learning framework (PUnet) for phase unwrapping form InSAR data. pUnet was a robust framework using U-net as the basic structure combined with an attention mechanism and positional encoding, facilitating accurate phase unwrapping from the wrapped phase. Through comparative experiments with typical phase unwrapping algorithms, we demonstrated that pUnet could obtain absolute phases with high accuracy and robustness than from the wrapped phase under various levels of noise.

The reason for the existence of small targets and sub-pixel targets in hyperspectral images is that the low spatial resolution leads to the phenomenon of mixed pixels, which makes some targets exist in the form of small targets and sub-pixel targets in the image. Moreover, the traditional anomaly detection algorithm introduces the regularization parameter into hyperspectral anomaly detection, but the regularization parameter needs to be manually set and adjusted. However, since the anomaly detection is carried out without prior information, in real situations, it is difficult to obtain the prior information of the target, so the regularization parameter may be difficult to determine. In addition, considering that traditional algorithms are not sensitive to small target pixel targets, a hyperspectral anomaly detection network based on self encoder is designed for hyperspectral images with small targets and sub-pixel targets.

Self-supervised depth estimation approaches present excellent results that are comparable to those of the fully supervised approaches, by employing view synthesis between the target and reference images in the training data. ResNet, which serves as a backbone network, has some structural deficiencies when applied to downstream fields, because its original purpose was to cope with classification problems. The low-texture area also deteriorates the performance. To address these problems, we propose a set of improvements that lead to superior predictions. First, we boost the information flow in the network and improve the ability to learn spatial structures by improving the network structures. Second, we use a binary mask to remove the pixels in low-texture areas between the target and reference images to more accurately reconstruct the image. Finally, we input the target and reference images randomly to expand the dataset and pre-train it on ImageNet, so that the model obtains a favorable general feature representation. We demonstrate state-of-the-art performance on an Eigen split of the KITTI driving dataset using stereo pairs.

Voxel-based structures in 3D detection have achieved rapid advancement due to their superior capability for feature extraction. However, the accuracy is usually low because the point cloud is divided into a grid. In order to overcome the above problems and improve detection accuracy, we propose a flexible two-stage 3D object detection architecture, which adopts two branches to refine generated proposals, aggregating voxel features and raw point features simultaneously. We also design a new gating mechanism to achieve fusion features from different levels. In addition, we propose a novel feature aggregation module to reduce the semantic gap between the features of the two types. First, a transformer based on raw points is employed as an encoder to aggregate the contextual information. Then, the point-based channel-wise self-attention mechanism serves as a decoder to aggregate the global features. Experiment results on the KITTI 3D dataset and Waymo Open datest demonstrate that our approach outperforms the state-of-the-art methods and exhibits excellent scalability.

Hyperspectral images technology can greatly enhance the extraction ability of ground object information. It is a research hotspot and frontier field in the field of remote sensing in recent years, and has great application value and broad development prospects in many related fields This paper summarizes the methods of denoising and target detection in hyperspectral images.To help researchers better sort out the relevant algorithms.

The denoising of interferometric phase images attracts many researchers. Natural image denoising algorithms based on neural networks are often proposed in the development of deep learning methods. A neural network, In-CNN, is derived from an advanced natural image denoising network and proposed for interferometric phase image denoising. Preactivation and residual learning methods are combined and applied to the function of the neural network nodes. Considering the particularity of the interferometric phase image, we propose a neural network based on the rational application of the preactivation mode and feedforward mapping, which is different from previous natural image denoising networks. We also construct a training set for an interferometric phase image denoising neural network. We experimentally verify that our model performs better than state-of-the-art interferometric phase image denoising methods based on sparse representation and advanced natural image denoising networks. We discuss the complexity of traditional interferometric phase image denoising algorithms to demonstrate the efficiency of the proposed method. (c) 2021 SPIE and IS&T [DOI: 10.1117/1.JEI.30.2.023013]

Research findings relating to complex-valued images are applied in many fields. Because of external interference or equipment problems, a complex-valued image will collect noise during acquisition and generation. Noise reduction of complex-valued images is of great significance to the applications of measurement data. The main denoising methods can be divided into two categories. The first kind denoises both the real part and the imaginary part of a complex-valued image by employing an effective denoising method in the real domain. Currently, advanced denoising algorithms in the real domain include the nonlocal means image denoising method,1 block-matching and three-dimensional filtering (3D),2 and a denoising algorithm based on nonparametric Bayesian dictionary learning.3 Islam and Chong applied the wavelet transform method to denoise interferometric synthetic aperture radar (InSAR) images.4 Zhang et al.5 achieved noise reduction of InSAR images with block-matching and 3D filtering (BM3D). Baselice et al. proposed a Bayesian magnetic reso Noise reduction is an essential preprocess in applications of complex-valued images. A method of complex-valued image denoising based on complex-valued dictionary learning and an adaptive complex-valued dictionary filter (OCDL-ACDF) is proposed. Our dictionary is first trained by the online dictionary learning method. Then, to further reduce the noise contained in the dictionary atoms, we design a complex-valued dictionary filter based on the feature similarity between the atoms of redundant dictionaries. By combining the advantages of online dictionary learning and denoising methods of real-valued images, an effective complex-valued dictionary is obtained. The orthogonal matching tracking method, which is a greedy algorithm, is used in the process of sparse coding. The simulation experiments show that the denoising effect of the proposed method is not only better than the current advanced algorithms but also effective at avoiding overfitting. The detail fidelity was also relatively high. (c) 2021 SPIE and IS&T [DOI: 10.1117/1 .JEI.30.2.023027]

For the task of monocular depth estimation, self-supervised learning supervises training by calculating the pixel difference between the target image and the warped reference image, obtaining results comparable to those with full supervision. However, the problematic pixels in low-texture regions are ignored, since most researchers think that no pixels violate the assumption of camera motion, taking stereo pairs as the input in self-supervised learning, which leads to the optimization problem in these regions. To tackle this problem, we perform photometric loss using the lowest-level feature maps instead and implement first- and second-order smoothing to the depth, ensuring consistent gradients ring optimization. Given the shortcomings of ResNet as the backbone, we propose a new depth estimation network architecture to improve edge location accuracy and obtain clear outline information even in smoothed low-texture boundaries. To acquire more stable and reliable quantitative evaluation results, we introce a virtual data set in the self-supervised task because these have dense depth maps corresponding to pixel by pixel. We achieve performance that exceeds that of the prior methods on both the Eigen Splits of the KITTI and VKITTI2 data sets taking stereo pairs as the input.

In recent years, sparse representation technology has made outstanding contributions in signal processing, image processing, target recognition, blind source separation, etc. Greedy algorithms are an important class of algorithms in the field of sparse representation, and matching pursuit algorithm and orthogonal matching pursuit algorithm are two typical greedy algorithms. The orthogonal matching pursuit algorithm has a great improvement in the convergence time compared with the matching pursuit algorithm, but the matching pursuit algorithm is still used in many current scenarios where greedy algorithm is applied. In order to understand the signal reconstruction effect of greedy algorithms and to analyze and compare the advantages and disadvantages of different algorithms, this paper briefly analyzes and compares these two typical greedy algorithms from the principle to simulation, and obtains the simulation results of matching pursuit algorithm and orthogonal matching pursuit algorithm and compares the advantages and disadvantages.

Abstract. The denoising of interferometric phase images attracts many researchers. Natural image denoising algorithms based on neural networks are often proposed in the development of deep learning methods. A neural network, In-CNN, is derived from an advanced natural image denoising network and proposed for interferometric phase image denoising. Preactivation and residual learning methods are combined and applied to the function of the neural network nodes. Considering the particularity of the interferometric phase image, we propose a neural network based on the rational application of the preactivation mode and feedforward mapping, which is different from previous natural image denoising networks. We also construct a training set for an interferometric phase image denoising neural network. We experimentally verify that our model performs better than state-of-the-art interferometric phase image denoising methods based on sparse representation and advanced natural image denoising networks. We discuss the complexity of traditional interferometric phase image denoising algorithms to demonstrate the efficiency of the proposed method.

针对高光谱图像异常检测中背景信息与异常目标信息难以有效区分,背景预测精度不佳的问题,提出一种新的基于背景重建的高光谱图像异常检测算法.通过字典学习方法获取高光谱图像背景光谱字典,并利用该字典对待检测图像进行稀疏重建,得到预测背景图像.将预测背景图像与原始图像做差后得到残差图像,进而利用局部RX检测算法对残差图像进行遍历,实现异常目标检测.通过对真实高光谱图像场景进行实验,证明了算法的有效性.

By merging high-resolution panchromatic (PAN) images with low-resolution multispectral (MS) images, high-resolution MS images with complementary information can be obtained, i.e., pansharpening. Multiresolution analysis (MRA) methods have attracted widespread attention in the pansharpening field. The spatial detail information injected into MS images is extracted from PAN images by MRA tools. Since such methods often suffer from spatial distortion and ringing artifacts, a restoration algorithm based on blind deblurring and iterative back-projection (IBP) is proposed in this paper. First, a blind deblurring method based on the Tikhonov regular constraint model is used to estimate the blurring filter. Second, spatial details extracted from PAN images are modulated into MS images using a high-pass modulation (HPM) framework, and then fusion images are spatially enhanced based on the modulation results and blurring filter. Finally, the IBP technique is used to project the reconstruction error back to iteratively update and optimize the desired high-resolution images. Experiments are performed on data sets acquired by different satellites at full and reduced resolution, and eight state-of-the-art MRA-based pansharpening methods are used for validation. Compared to the enhanced back-projection (EBP) algorithm, the proposed restoration method is better in improving spectral and spatial quality of MRA-based pansharpening. The results indicate the effectiveness and superiority of the proposed method.

Visualization has been spaciously used in various fields for assisting data analysis and so on. In this paper, we focus on the visualization of hyperspectral data and analyze the characteristics of them firstly. Then, an interactive visualization framework for spatial and spectral dimensions of hyperspectral data is studied by dimensionality reduction. For the sake of displaying the spatial distribution and the electromagnetic energy scattered in the spectral domain, we proposed a method for visualizing the isosurfaces of hyperspectral data which combining the information in spatial and spectral dimensions. In this method, by using the ray casting algorithm combining with the equivalent sampling, the hyperspectral data of a specified value are visualized with specified color. The results in the experiment section show that the proposed method displays the distribution of hyperspectral data with different values in an intuitional way in real-time. Furthermore, the data users can be benefited in noise analysis and hyperspectral classification by visualization.

The space information networks (SIN) have a series of characteristics, such as strong heterogeneity, multiple types of resources, and difficulty in management. Aiming at the problem of resource allocation in SIN, this paper firstly establishes a hierarchical and domain-controlled SIN architecture based on software-defined networking (SDN). On this basis, the transmission, caching, and computing resources of the whole network are managed uniformly. The Asynchronous Advantage Actor-Critic (A3C) algorithm in deep reinforcement learning is introduced to model the process of resource allocation. The simulation results show that the proposed scheme can effectively improve the expected benefits of unit resources and improve the resource utilization efficiency of the SIN.

Extraction and injection of spatial details from the panchromatic (PAN) image and preservation of the spectral characteristics of the multispectral (MS) image are the key issues in pansharpening methods, which are of great significance for the application of image interpretation and target recognition. In this paper, a spectral-driven pansharpening method in which the injection detailed coefficients are estimated over each component segmented by the adaptive K-means algorithm is proposed. The pansharpening method relies on a multi-resolution framework, generalized Laplacian pyramid (GLP) technique, which is applied for the extraction of detail image. The fused image is used as the feedback element, its distance from the original MS image is used to adjust the number of segments adaptively so as to reduce the spectral distortion. Experiments carried out on GeoEye-1 and QuickBird data sets demonstrate the efficiency and effectiveness of our proposed method.

In order to improve the spatial resolution of multispectral (MS) images and reduce spectral distortion, a segmentation-cooperated pansharpening method using local adaptive spectral modulation (LASM) is proposed in this paper. By using the k-means algorithm for the segmentation of MS images, different connected component groups can be obtained according to their spectral characteristics. For spectral information modulation of fusion images, the LASM coefficients are constructed based on details extracted from images and local spectral relationships among MS bands. Moreover, we introduce a cooperative theory for the pansharpening process. The local injection coefficient matrix and LASM coefficient matrix are estimated based on the connected component groups to optimize the fusion result, and the parameters of the segmentation algorithm are adjusted according to the feedback from the pansharpening result. In the experimental part, degraded and real data sets from GeoEye-1 and QuickBird satellites are used to assess the performance of our proposed method. Experimental results demonstrate the validity and effectiveness of our method. Generally, the method is superior to several classic and state-of-the-art pansharpening methods in both subjective visual effect and objective evaluation indices, achieving a balance between the injection of spatial details and maintenance of spectral information, while effectively reducing the spectral distortion of the fusion image.

Due to the sparsity of hyperspectral images, the dictionary learning framework has been applied in hyperspectral endmember extraction. However, current endmember extraction methods based on dictionary learning are not robust enough in noisy environments. To solve this problem, this paper proposes a novel endmember extraction approach based on online robust dictionary learning, termed EEORDL. Because of the large scale of the hyperspectral image (HSI) data, an online scheme is introduced to reduce the computational time of dictionary learning. In the proposed algorithm, a new form of the objective function is introduced into the dictionary learning process to improve the robustness for noisy HSI data. The experimental results, conducted with both synthetic and real-world hyperspectral datasets, illustrate that the proposed EEORDL outperforms the state-of-the-art approaches under different signal-to-noise ratio (SNR) conditions, especially for high-level noise.

Maximum a posteriori estimation of Markov random fields (MRFs) is a popular research area in computer vision, and many algorithms have been proposed to deal with these types of problems. The phase-unwrapping problem is modeled as the optimization of MRFs in this research, and the binary algorithm, improved quadratic pseudo-Boolean optimization, is utilized to solve the phase-unwrapping problem. Both the interferometric phases generated from the commonly used computer simulated surfaces and also real terrains are researched in the experimental section, and the unwrapping results are compared. The proposed algorithm achieves unwrapping results comparable to the state-of-the-art unwrapping method but costs less time for large-scale phase images.