Title: An intelligent image denoising method using weighted multi-scale CB morphological filter algorithm

Authors: Yongjie Tan; Jie Qin

Addresses: School of Computer Science and Technology, Zhoukou Normal University, Zhoukou, Henan, 466100, China ' School of Computer Science and Technology, Zhoukou Normal University, Zhoukou, Henan, 466100, China

Abstract: In order to improve the accuracy of paper disease recognition in paper making process, a paper image denoising method based on multi-scale contour bougie (CB) element morphological filter is proposed. The small-scale structural elements in CB morphological filtering algorithm have better detail protection ability, and the large-scale structural elements have stronger noise suppression ability. By selecting several structural elements to filter the image, and then fusing the filtered images at different scales, the final denoising image can be obtained. The simulation results on the holes paper disease image with Gauss noise and salt and pepper noise show that the PSNR reaches more than 43 dB and 38 dB respectively, which proves that this method can suppress the noise in the image and keep the image details well.

Keywords: image denoising; CB morphological filter; multi-scale structural elements; weighted fusion.

DOI: 10.1504/IJITM.2022.126701

International Journal of Information Technology and Management, 2022 Vol.21 No.4, pp.359 - 368

Received: 09 May 2019
Accepted: 28 Dec 2019
Published online: 03 Nov 2022 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article