CD-SDN: Unsupervised Sensitivity Disparity Networks for Hyper-Spectral Image Change Detection

Deep neural networks (DNNs) could be affected by the regression level of learning frameworks and challenging changes caused external factors; their deep expressiveness is greatly restricted. Inspired fine-tuned DNNs with sensitivity disparity to pixels two states, in this paper, we propose a novel change detection scheme served (CD-SDN). The CD-SDN proposed for detecting bi-temporal hyper-spectral images captured AVIRIS sensor HYPERION over time. In CD-SDN, frameworks, unchanged network (USNet) changed (CSNet), are utilized as dominant part generation binary argument map (BAM) high assurance (HAM). Then approaches, arithmetic mean learning, employed re-estimate BAM. Finally, detected results merged HAM obtain final maps (BCMs). Experiments performed on three real-world hyperspectral image datasets, indicate good universality adaptability scheme, well its superiority other existing state-of-the-art algorithms.