Research of unsupervised image change detection algorithm based on clustering characteristic of 2-D histogram

Change detection in images of a given scene acquired at different times is one of the most interesting topics of remote image processing, which finds important applications in many fields. In this paper, a novel image change detection algorithm was proposed based on the clustering characteristic of 2-D histogram. First, the best segmentation direction of 2-D histogram formed by pixel gray levels and the local average gray levels was ascertained by using LSM. Secondly, a kind of new 2-D entropy was defined to search the best threshold line and segment the 2-D histogram into unchanged and change region. Then the change area was detected based on the change region of 2-D histogram. Finally, the proposed algorithm was compared to traditional change detection algorithm by carrying out an experiment on a synthetic data set generated artificially. Theoretical analysis and experiment result show that the proposed algorithm is more accurate on detection precision, and faster on detection speed.