Land Use Change Detection Using Deep Siamese Neural Networks and Weakly Supervised Learning

A weakly supervised change detection method is proposed for remotely sensed multi-temporal images, by utilizing a Siamese neural network architecture. The architecture of the combination two multi-filter multi-scale deep convolutional networks (MFMS DCNN). Initially, trained image-level semantic labels image pairs in dataset. features are obtained using to generate difference (DI). Then, PCA and K-means algorithms has been used produce map pair images. Experiments were carried out datasets. this paper offers better results comparison both supervised- unsupervised-based state-of-the-art models techniques.