A Deep Siamese Network with Hybrid Convolutional Feature Extraction Module for Change Detection Based on Multi-sensor Remote Sensing Images

Abstract

Information extraction from multi-sensor remote sensing images has increasingly attracted attention with the development of remote sensing sensors. In this study, a supervised change detection method, based on the deep Siamese convolutional network with hybrid convolutional feature extraction module (OB-DSCNH), has been proposed using multi-sensor images. The proposed architecture, which is based on dilated convolution, can extract the deep change features effectively, and the character of “network in network” increases the depth and width of the network while keeping the computational budget constant. The change decision model is utilized to detect changes through the difference of extracted features. Finally, a change detection map is obtained via an uncertainty analysis, which combines the multi-resolution segmentation, with the output from the Siamese network. To validate the effectiveness of the proposed approach, we conducted experiments on multispectral images collected by the ZY-3 and GF-2 satellites. Experimental results demonstrate that our proposed method achieves comparable and better performance than mainstream methods in multi-sensor images change detection.

Change Detection Based on Artificial Intelligence: State-of-the-Art and Challenges

“…To generate change maps, the output feature maps in the two periods can be directly classified by the concatenation of channels [91,155,161] or can be used to produce difference maps using a certain distance metric [9], and then used for further change analysis [162,163]. To retain multi-scale change information, feature maps at different depths can be concatenated for change detection [164][165][166][167], and this works well.…”

A Survey of Change Detection Methods Based on Remote Sensing Images for Multi-Source and Multi-Objective Scenarios

“…In addition, similar to EF, the skip-connect structure also shows the potential to contemplate the correlation between phases by connecting layers within dual-branch CNN. It is worth noting that the results of logical operation, that is, difference, stacking, concentration, or other logical operations, can be treated as the end element for skip-connect [135], as shown in (d) of Figure 15. Of course, in addition to structural changes, some tricks that focus on global or local features are also conducive to the optimization of change detection model, such as atrous spatial pyramid pooling (ASPP) [136] and dilated convolution [135,137].…”

“…It is worth noting that the results of logical operation, that is, difference, stacking, concentration, or other logical operations, can be treated as the end element for skip-connect [135], as shown in (d) of Figure 15. Of course, in addition to structural changes, some tricks that focus on global or local features are also conducive to the optimization of change detection model, such as atrous spatial pyramid pooling (ASPP) [136] and dilated convolution [135,137]. At present, binary cross-entropy [138] and structural similarity index measures (SSIM) [139] are mainstream loss functions to highlight differences and evaluate the similarity of multi-temporal features.…”

Land Cover Change Detection With VHR Satellite Imagery Based on Multi-Scale SLIC-CNN and SCAE Features

“…Zhang et al [19] adopted a change detection framework including a denoising autoencoder and mapping network to perform the binary change detection. To extract change information from multi-sensor remote sensing images, Wang et al [20] proposed an OB-DSCNH module, deep change features were extracted effectively and the depth and width of the network were partly increased. For multi-label change detection, Khan et al [21] applied a deep neural network (DNN) architecture to conduct change detection in forests with long-term serial images, dividing the change results into unchanged, harvest events, or fire events.…”