Translator Disclaimer
27 November 2012 Distorted target recognition based on canny operator enhancing OT-MACH filter
Author Affiliations +
Abstract
The image pattern recognition can accurately identify and locate the target, but image pattern recognition is unable to accurately recognize the distorted targets (the targets rotated in plane or scale changed), which has restricted the development of the image pattern recognition. In order to solve the problem of inaccurate recognition for distorted target in cluttered background among the image pattern recognition, the distorted target images and the training images are edge extracted by canny operator. The Optimum Trade-off Maximum Average Correlation Height (OT-MACH) filter is synthesized with the edge extracted training images. The low frequency information of the distorted target images and the filter is enhanced. Then the edge extracted distorted target image is filtered by the OT-MACH filter. Thereby, the distortion tolerance of the OT-MACH filter is expanded. It can respond higher correlation peaks and have higher distortion tolerance to recognize various types of distorted targets in cluttered background. By this method, which the space edge extraction combines with frequency domain filtering, the scale distortion tolerance is 0.72~1.42 times; the rotation distortion tolerance can reach up to 70 degrees. In order to prove the feasibility of this method, a lot of computer simulation experiments have been done with the canny operator and the OT-MACH filter.
© (2012) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Jiyang Shang, Yu Zhang, Qibo Zhang, and Wensheng Wang "Distorted target recognition based on canny operator enhancing OT-MACH filter", Proc. SPIE 8559, Information Optics and Optical Data Storage II, 85590N (27 November 2012); https://doi.org/10.1117/12.999326
PROCEEDINGS
7 PAGES


SHARE
Advertisement
Advertisement
Back to Top