An intelligent image detection method using improved canny edge detection operator. (25th October 2022)

• Record Type: Journal Article
• Authors: Wang, Qian; Chen, Wenxia; Peng, Haiyun
• Is Part Of: International journal of information technology and management. Volume 21:Number 4(2022)
• Page Start: 369
• Page End: 381
• Publication Date: 2022-10-25
• Subjects: intelligent image detection -- edge detection -- improved canny operator -- Gauss filtering -- feature statistical analysis -- adaptive; Management information systems -- Periodicals; Information technology -- Periodicals; Management -- Data processing -- Periodicals; 658.403805
• Languages: English
• ISSNs: 1461-4111

Abstract

In order to meet the requirement of edge detection of paper disease image in papermaking process, a paper disease image detection method based on improved Canny operator is proposed. Firstly, according to the principle of Gauss filtering and the method of feature statistical analysis, the filtering function and window are selected adaptively. Then, in the gradient solution, the traditional 2 × 2 neighbourhood is replaced by the 3 × 2 or 2 × 3 neighbourhood which enhances the weight of the intermediate pixel, and the accuracy of edge detection is improved by enhancing the influence of the intermediate pixel. Finally, the iterative averaging method is used to determine the optimal threshold and reduce the error rate of image edge segmentation. The experimental results show that this method can effectively detect the edges of paper disease area and has good edge continuity.