Quantities of multi-temporal remote sensing (RS) images create favorable conditions for exploring the urban change in the long term. However, diverse multi-source features and change patterns bring challenges to the change detection in urban cases. In order to sort out the development venation of urban change detection, we make an observation of the literatures on change detection in the last five years, which focuses on the disparate multi-source RS images and multi-objective scenarios determined according to scene category. Based on the survey, a general change detection framework, including change information extraction, data fusion, and analysis of multi-objective scenarios modules, is summarized. Owing to the attributes of input RS images affect the technical selection of each module, data characteristics and application domains across different categories of RS images are discussed firstly. On this basis, not only the evolution process and relationship of the representative solutions are elaborated in the module description, through emphasizing the feasibility of fusing diverse data and the manifold application scenarios, we also advocate a complete change detection pipeline. At the end of the paper, we conclude the current development situation and put forward possible research direction of urban change detection, in the hope of providing insights to the following research.

The filtering of multi-pass synthetic aperture radar interferometry (InSAR) stack data is a necessary preprocessing step utilized to improve the accuracy of the object-based three-dimensional information inversion in urban area. InSAR stack data is composed of multi-temporal homogeneous data, which is regarded as a third-order tensor. The InSAR tensor can be filtered by data fusion, i.e., tensor decomposition, and these filters keep balance in the noise elimination and the fringe details preservation, especially with abrupt fringe change, e.g., the edge of urban structures. However, tensor decomposition based on batch processing cannot deal with few newly acquired interferograms filtering directly. The filtering of dynamic InSAR tensor is the inevitable challenge when processing InSAR stack data, where dynamic InSAR tensor denotes the size of InSAR tensor increases continuously due to the acquisition of new interferograms. Therefore, based on the online CANDECAMP/PARAFAC (CP) decomposition, we propose an online filter to fuse data and process the dynamic InSAR tensor, named OLCP-InSAR, which performs well especially for the urban area. In this method, CP rank is utilized to measure the tensor sparsity, which can maintain the structural features of the InSAR tensor. Additionally, CP rank estimation is applied as an important step to improve the robustness of Online CP decomposition - InSAR(OLCP-InSAR). Importing CP rank and outlier’s position as prior information, the filter fuses the noisy interferograms and decomposes the InSAR tensor to acquire the low rank information, i.e., filtered result. Moreover, this method can not only operate on tensor model, but also efficiently filter the new acquired interferogram as matrix model with the assistance of chosen low rank information. Compared with other tensor-based filters, e.g., high order robust principal component analysis (HoRPCA) and Kronecker-basis-representation multi-pass SAR interferometry (KBR-InSAR), and the widespread traditional filters operating on a single interferometric pair, e.g., Goldstein, non-local synthetic aperture radar (NL-SAR), non-local InSAR (NL-InSAR), and InSAR nonlocal block-matching 3-D (InSAR-BM3D), the effectiveness and robustness of OLCP-InSAR are proved in simulated and real InSAR stack data. Especially, OLCP-InSAR can maintain the fringe details at the regular building top with high noise intensity and high outlier ratio.

Multitemporal optical remote sensing image registration is still a challenging problem for current feature-based image registration algorithms due to the complex nonlinear discrepancies arising from diverse factors, including illumination, weather, and surface condition changes. To address the issue, this article attempts to combine the dual receptive field descriptors (DRFDs) constructed by a novel deep convolutional network. In addition, a novel inner loss function (ILF) that imposes constraints on the intermediate descriptors is adopted in order to consolidate the distinguishability of the descriptors when the overlapping areas of the input image patches are large. Subsequently, the dual feature distance maps (DFDMs) are built on the basis of the DRFDs and combined with features from accelerated segment test (FAST) key points for efficient and accurate correspondence establishment across the source image and the target image. Eventually, an iterative algorithm is proposed to remove the possible outliers. Experiments show that the combination of DRFDs trained with the ILF performs better than current learnable local descriptors, such as L2-Net, HardNet, and SOSNet. The image registration results using our method are more accurate than the methods based on learnable descriptors, such as L2-Net, HardNet, and SOSNet, and handcrafted descriptors, such as scale-invariant feature transform (SIFT), SURF, and ORB.