智能技术学报

文章详情

稿件标题: Change Detection in Synthetic Aperture Radar Images Based on Fuzzy Restricted Boltzmann Machine
稿件作者: Na Li , Jiao Shi, Yu Lei
关键字词: Fuzzy restricted boltzmann machine (FRBM); image change detection; Synthetic Aperture Radar (SAR) image
文章摘要: Image Change Detection is a process to identify the changes of two images of the same scene that were taken in di erent times.In this paper, we propose a novel change detection approach based on fuzzy Restricted Boltzmann Machine (FRBM). In order to adapt to the network structure of FRBM, we made some adjustments to the original Back Propagation (BP) and that is called fuzzy Back Propagation (FBP). The proposed approach applies FRBM as unsupervised feature learning algorithm and FBP as supervised fine-tuning algorithm to train data. The feature of FRBM is the parameters governing the model are replaced by fuzzy numbers. FRBMapplied to change detection can reduce the e ect of speckle noise in Synthetic Aperture Radar Images (SAR) and have better representation capability than traditional RBM. Experiments on real data sets and theoretical analysis show the proposed method can obtain promising results and outperforms some other methods.
收录刊物: 2017年2卷2期
稿件基金:
浏览次数: 90
下载次数: 38
点击下载