SAR Image Change Detection Method Based on Pulse-Coupled Neural Network
The study proposes a new algorithm for change detection of SAR images based on segmentation to improve the accuracy of the SAR image change detection. The ratio method is used to acquire the difference image (DI). Then, the global dictionary is applied to address the image denoising problem. Finally, change mask is obtained by pulse-coupled neural network (PCNN). The results of the experiment show that the proposed method improves accuracy.
KeywordsChange detection Remote sensing image Global dictionary Pulse-coupled neural network (PCNN)
This work was supported in part by International Cooperative Research and Personnel Training Projects of the Ministry of the Ministry of Education of the People’s Republic of China [Grant number DICE2014-2029].
- Bergeaud, F., & Mallat, S. (1996). Matching pursuit: adaptive representations of images and sounds. Computational and Applied Mathematics, 15(2), 97–109.Google Scholar
- Li, H. & Bai, Z. (2008). A new PCNN-based method for segmentation of SAR images. Control, Automation, Robotics and Vision, 2008. ICARCV 2008. 10th International Conference on IEEE 1635–1639Google Scholar
- Ma, Y., Li, H., & Gu, H. (2011). A study of fast change detection algorithm based on feature library of remote sensing imagery. Image and Data Fusion (ISIDF), International Symposium on IEEE 1–3.Google Scholar
- Zhao, Z., Yan, Q., Liu, Z., & Luo, C. (2014). Study on method of geohazard change detection based on integrating remote sensing and GIS. IOP Conference Series Earth and Environmental Science, 17, 682–691.Google Scholar