To fully explore the complementary information from optical and synthetic aperture radar (SAR) imageries, they need firstly to be co-registered with high accuracy. Due to the vast radiometric and geometric disparity, the problem to match high resolution optical and SAR images is quite challenging. The present deep learning based methods have shown advantages over the traditional approaches, but the performance increment is not significant. In this article, we explore better network framework for high resolution optical and SAR image matching from three aspects. Firstly, we propose an effective multilevel feature fusion method which helps to take advantage of both the low-level fine-grained features for precious feature location and the high-level semantic features for better discriminative ability. Secondly, a feature channel excitation procedure is conducted using a novel multi-frequency channel attention module, which is able to make image features of different types and multiple levels effectively collaborate with each other, and produce image matching features with high diversity. Thirdly, the self-adaptive weighting loss is introduced, with which, each sample is assigned with an adaptive weighting factor, and therefore information buried in all nearby samples can be better exploited. Under a pseudo-Siamese architecture, the proposed optical and SAR image matching network (OSMNet) is trained and tested on a large and diverse high resolution optical and SAR dataset. Extensive experiments demonstrate that each component of the proposed deep framework helps to improve the matching accuracy. Also, the OSMNet shows overwhelming superior to the state-of-the-art handcrafted approaches on imageries of different land-cover types.

Built-up area extraction from polarimetric SAR (PolSAR) imagery has a close relationship with urban planning, disaster management, etc. Since the buildings have complex geometries and may be misclassified as forests due to the significant cross-polarized scattering, built-up area extraction from PolSAR data is still a challenging problem. This paper proposes a new urban extraction method for PolSAR data. First, a multiple-component model-based decomposition method, which was previously proposed by us, is applied to detect the urban areas using the scattering powers. Second, with the sub-aperture decomposition, a new average polarimetric coherence coefficient ratio is proposed to discriminate the urban and natural areas. Finally, these two preliminary detection results are fused on the decision level to improve the overall detection accuracy. We validate our method using one dataset acquired with the Phased Array type L-band Synthetic Aperture Radar (PALSAR) system. Experimental results demonstrate that the decomposed scattering powers and the proposed polarimetric coherence coefficient ratio are both capable of distinguishing urban areas from natural areas with accuracy about 83.1% and 80.1%, respectively. The overall detection accuracy can further increase to 86.9% with the fusion of two detection results.

Image matching is a primary technology to fuse the complementary information from optical and SAR images. Due to the high nonlinear radiometric and geometric relationship, the optical and SAR image matching task remains a widely unsolved challenge. In this study, we propose to use a Siamese convolutional neural network (CNN) architecture to learn pixelwise deep dense features. The proposed network is able to balance the learning of high-level semantic information and low-level fine-grained information, which is nonnegligible for feature matching task. Under the local searching framework, the loss function is defined based on the score map produced by the sum of squared differences (SSDs) between the learned pixelwise dense features of local optical and the SAR image patches, with a fast implementation in the frequency domain. The hardest negative mining strategy is adopted to increase the discrimination of the network. Extensive experiments are conducted on optical and SAR image pairs of different spatial resolution and different landcover types, verifying the superiority and robustness of the proposed method in terms of matching accuracy and matching precision.

The co-registration of optical and SAR imageries is the bottleneck in exploring the complementary information from the two multi-modal datasets. The difficulties lie in not only the complex radiometric relationship between them, but also the distinct geometrical models of the optical and SAR imaging systems, which cause it nontrivial to explicitly depict the spatial relationship between the corresponding image regions when elevation fluctuations exist. This article aims to investigate the optical flow technique for the pixelwise dense registration of the high resolution optical and SAR images, so as to get rid of the outlier removal and geometric mapping procedures, which have to be conducted in the classical image registration approaches that are based on sparse feature point matching. Herein, a deep optical flow framework is designed. Firstly, a dilated feature concatenation method is proposed to enhance the discriminability of the pixelwise features for similarity measurement. An effective network training strategy is used, based on a smoothed flow loss, and also a training dataset that contains simulated elevation fluctuations. Secondly, we propose a self-supervised optical flow finetuning strategy. It incorporates the strength of the blockwise matching approach, which produces better matching precision, into the proposed pixelwise matching procedure. In this way, the accuracy of the optical-SAR dense registration is substantially improved. Extensive experiments conducted on the 1m resolution optical-SAR image pairs of different land-cover types and distinct topographic conditions indicate that, the proposed OSFlowNet-Ft framework is quite robust, and has the potential to perform the optical-SAR image dense registration in practical applications. The Python code of the proposed deep optical flow network will be made available.

Due to the speckle noise and complex geometric distortions within SAR images, it is still a challenge to develop a stable method that can produce superpixels with both high boundary adherence and visual compactness with low computational costs at the same time. In this paper, we propose an adaptive superpixel generation approach with linear feature clustering and edge constraint for synthetic aperture radar (SAR) images, which consists of three stages. First, the local gradient ratio pattern of each pixel in SAR imagery is extracted as features, which was previously proposed by us for SAR target recognition and has been proven to be insensitive to speckle noise. Second, we propose to use the feature-ratio-based edge detector with Gauss-shaped window instead of the traditional rectangle-shaped window to obtain the edge strength map and final edges for SAR images. Finally, a modified normalized cut (Ncut)-based superpixel generation strategy is adopted using a distance metric that simultaneously measures both the feature similarity and space proximity. In this strategy, we approximate the similarity measure through a positive semidefinite kernel function rather than directly using the traditional eigen-based algorithm. Therefore, the objective functions of weighted local K-means and Ncuts can achieve the same optimum point by appropriately weighting each point in this feature space, which greatly reduces the computation cost. During the linear feature clustering, the coefficient of variation is used to automatically determine the tradeoff factor between the feature similarity and space proximity, which helps change the superpixel shape and size adaptively according to the image homogeneity. Furthermore, the edge information is also introduced to constrain the clustering for the sake of high boundary adherence. By bridging the local K-means clustering and Ncuts, as well as the benefits of edge constraint, our method not only produces superpixels with good boundary adherence but also captures the global image structure information. Experimental results with simulated and real SAR images demonstrate the effectiveness of our proposed method, which performs better than other state-of-the-art algorithms.

This article proposes an efficient and adaptive statistical superpixel merging approach with edge penalty for polarimetric synthetic aperture radar (PolSAR) image segmentation. Based on the initial superpixel over-segmentation result obtained by our previously proposed adaptive polarimetric superpixel generation algorithm (Pol-ASLIC), this work achieves efficient and accurate PolSAR image segmentation by merging superpixels using the statistical region merging (SRM) framework. This article proposes to define a new dissimilarity measure between superpixels, which takes the edge penalty into consideration, leading to a reasonable and accurate merging order for superpixel pairs. With regard to the merging predicate of superpixels, a polarimetric homogeneity measurement (HoM) is used to define the merging threshold, making the merging predicate and merging threshold adaptive to the PolSAR image content. Experimental results on three airborne and one spaceborne PolSAR data sets demonstrate that the proposed approach can effectively improve the computation efficiency and segmentation accuracy in comparison with state-of-the-art merging-based methods for PolSAR data. More importantly, the proposed approach is free of parameters and easy to use.