22 October 2010 Detection algorithm of infrared small target based on improved SUSAN operator
Author Affiliations +
Abstract
The methods of detecting small moving targets in infrared image sequences that contain moving nuisance objects and background noise is analyzed in this paper. A novel infrared small target detection algorithm based on improved SUSAN operator is put forward. The algorithm selects double templates for the infrared small target detection: one size is greater than the small target point size and another size is equal to the small target point size. First, the algorithm uses the big template to calculate the USAN of each pixel in the image and detect the small target, the edge of the image and isolated noise pixels; Then the algorithm uses the another template to calculate the USAN of pixels detected in the first step and improves the principles of SUSAN algorithm based on the characteristics of the small target so that the algorithm can only detect small targets and don't sensitive to the edge pixels of the image and isolated noise pixels. So the interference of the edge of the image and isolate noise points are removed and the candidate target points can be identified; At last, the target is detected by utilizing the continuity and consistency of target movement. The experimental results indicate that the improved SUSAN detection algorithm can quickly and effectively detect the infrared small targets.
© (2010) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Xingmiao Liu, Xingmiao Liu, Shicheng Wang, Shicheng Wang, } "Detection algorithm of infrared small target based on improved SUSAN operator", Proc. SPIE 7658, 5th International Symposium on Advanced Optical Manufacturing and Testing Technologies: Optoelectronic Materials and Devices for Detector, Imager, Display, and Energy Conversion Technology, 76581W (22 October 2010); doi: 10.1117/12.867973; https://doi.org/10.1117/12.867973
PROCEEDINGS
6 PAGES


SHARE
Back to Top