LEARNING A TRANSFERABLE CHANGE DETECTION METHOD BY RECURRENT NEURAL NETWORK

In this paper, a novel change detection method learned from Recurrent Neural Network with transferable ability is proposed. The proposed method, which is based on an improved Long Short Term Memory (LSTM) model, aims at: 1) learning a novel change detection rule to distinguish changed regions with high accuracy; 2) analyzing a new target data with transferable ability from learned change rule; 3) learning the differencing information and detecting the changes independently without any classifiers. In the process of learning the change rule, a core memory cell is utilized to detect and record the differencing information in multi-temporal images; meanwhile, the memory cell can update the storage by iteration for optimization. Finally, experiments are performed on two multi-temporal datasets, and the results show superior performance on detecting changes with transferable ability.