RPCA Infrared Small Target Detection Based on Local Entropy Reference in Preprocessing

• Published in: 2021 International Conference on Control Science and Electric Power Systems (CSEPS) pp. 323 - 332
• Main Authors: Cao, Wenhuan, Xue, Xirui
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.05.2021
• Subjects: Entropy; Image reconstruction; Image segmentation; infrared small target detection; local entropy; Object detection; Power systems; preprocessing scheme selection Introduction; Principal component analysis; RPCA
• DOI: 10.1109/CSEPS53726.2021.00072

Based on the non-local similarity of the image, the low rank block image is obtained by image block reconstruction, which is the basic method to apply the robust principal component analysis (RPCA) algorithm to single frame infrared small target detection. This paper introduces the application process of RPCA algorithm in infrared small target detection of single frame image, and analyzes the influence of various blocking methods under different image backgrounds. In order to solve the problem that it is difficult to select the window and sliding step of image segmentation in complex background, a selection method based on the larger value of the minimum local entropy of image segmentation is proposed. The experimental results show that by calculating the local entropy of the image block, taking the larger value of the minimum local entropy as the reference, choosing the RPCA algorithm preprocessing scheme can make the small target detection of single frame infrared image achieve better results, and make up for the lack of RPCA algorithm application experience of engineers.