LAUSR

Detecting changes in complicated remote sensing images have been gaining much attention. One of the main challenges lies in how to detect intrinsic changes robustly while avoiding the false alarms caused by various challenging factors, such as spatial illumination variations, small viewpoint differences, noises, and outliers between multitemporal remote sensing images. To reduce the influence of these factors, in this article, we propose an unsupervised joint learning model based on a total variation (TV) regularization and bipartite deep convolutional neural network, called the TV regularized bipartite network (TVRBN). In this model, parametric feature differences are initialized by a bipartite autoencoder. An objective function is defined as a parametric feature difference term integrated with a change predetection constraint term and a TV regularization to the change probability map with intrinsic changes. Then, the unified objective function is optimized jointly to learn the bipartite network parameters and detect an intrinsic change map. Due to the intrinsic difference constraints in feature space and TV regularization, the proposed TVRBN method can compute higher quality and smoother change maps, suppress nonintrinsic changes, and overcome small viewpoint differences in comparable images. Extensive experiments on both homogeneous and heterogeneous images demonstrate the robustness of the proposed method by comparing it with state-of-the-art methods.