A biomimetic adaptive fuzzy edge detection method based on visual features

• Published in: Chinese Control Conference pp. 3902 - 3906
• Main Authors: Ren, Hongge, Liu, Weimin, Shi, Tao, Li, Fujin
• Format: Conference Proceeding Journal Article
• Language: English
• Published: TCCT 01.07.2016
• Subjects: Adaptation models; Adjustment; Algorithm design and analysis; Algorithms; Brightness; Edge detection; Feature extraction; Fuzzy; Global adaptive; Human eye vision; Image contrast; Image detection; Image edge detection; Local enhancement; Pal and King algorithm; Tri-Gaussian model; Visual systems; Visualization
• ISSN: 1934-1768
• DOI: 10.1109/ChiCC.2016.7553960

Aiming at the poor performance of the lower contrast images edge detection, a fuzzy edge image extraction algorithm based on human eye visual features is proposed. Based on the global and local adaptive adjustment feature of human visual perception system, the traditional Pal and King fuzzy edge detection algorithm is optimized. By setting a nonlinear mapping model, the overall brightness of the image is adaptively adjusted according to the statistical characteristics of the image. Through a bilateral filtering method combined with the tri-Gaussian model and the single Gaussian model, the local contrast enhancement of the brightness of the image is realized. The experimental results show that the optimization algorithm in retaining the important edge information of image edge detection is more consistent with human subjective vision and improves the adaptability and practicability of the edge detection.