Infrared Image Edge Detection Based on Improved Canny Algorithm

Abstract

In recent years, the infrared image has been used frequently in medical, military, and industrial fields, and it has become increasingly important to extract a good target contour from the infrared image. Because of the imaging mechanism of the infrared image, there is a lot of noise in the image, which leads to the difficulty of edge extraction. By analyzing the application of the Canny edge detection algorithm in the infrared image, it is found that the detection results have a poor noise filtering effect and the loss of edge details. To solve this problem, this paper improves the Canny algorithm. The Gaussian filter is replaced with the bilateral filter for smoothing noise filtering, and the double global threshold segmentation algorithm is used to select adaptively the high and low thresholds to overcome the error caused by artificial experience setting thresholds. The experimental results show that compared with the traditional Canny algorithm, the improved algorithm can suppress noise better and retain more edge details in the process of infrared image edge detection.