Edge detection using adjusted Chebyshev polynomials on contrast-enhanced images by modified histogram equalization

作     者：Acharya, Kuldip Ghoshal, Dibyendu 

作者机构：Department of Computer Science and Engineering National Institute of Technology Agartala Barjala Tripura Jirania 799046 India Department of Electronics and Communication Engineering National Institute of Technology Agartala Barjala Tripura Jirania 799046 India 

出 版 物：《International Journal of Information Technology (Singapore)》 (Int. J. Inf. Technol.)

年 卷 期：2022年第14卷第6期

页      面：3031-3038页

主　　题：Chebyshev polynomial Contrast enhancement Edge detection Geometric mean Histogram equalization Thresholding Weighted standard deviation 

摘      要：Edge detection is a technique for determining the object lines in a digital image. Edge detection processes are considerably improved by noise reduction and contrast enhancement. This paper represents a new edge detection algorithm for images using the adjusted Chebyshev polynomial curve fitting method on contrast-enhanced images. In the first step, a new image enhancement method is proposed to enhance the contrast of an image, using the weighted standard deviation to compute a clipping threshold value to restrict the rate of over-enhancement, and a geometric mean is used to partition the histogram into equal parts. In the second step, the histogram of an enhanced image obtained from step one has been processed by a novel approach using a modified Chebyshev polynomial to compute a threshold value which is applied to the gradient image to detect the edges. Image quality metrics are calculated on the proposed method on both the original image and on the proposed enhanced image. The quantitative evaluation shows that EBCM is increased by 8.00%, PR is increased by 1.00%, and ESSIM is increased by 3% by the proposed edge detection method applied to the proposed enhanced image than on an original image. The proposed method is compared with the state-of-art and latest edge detection algorithms, and experimental results show that the proposed edge detection method’s overall performance is superior to that of state-of-the-art and recent edge detection methods. © 2022, The Author(s), under exclusive licence to Bharati Vidyapeeth s Institute of Computer Applications and Management.