Forward-looking sonar image target segmentation method and system and electronic equipment

Abstract

The invention provides a forward-looking sonar image target segmentation method and system and electronic equipment, and the method comprises the steps: 1, carrying out the preprocessing of a collected forward-looking sonar target image through a mathematical morphology reconstruction method; 2, segmenting the preprocessed image by using a level set algorithm added with a distance constraint term to obtain a final target boundary segmentation result; according to the method, on the basis of a traditional C-V level set segmentation algorithm, a segmentation preprocessing process based on morphological reconstruction is added, and the problem that boundary segmentation precision is not high is solved. Meanwhile, the method can weaken the influence of the initial contour setting position on the final segmentation result to a certain extent. The evaluation process proves that the underwater target boundary segmentation result of the method is more accurate, the effect is more ideal, and the method is more suitable for the target segmentation process of the forward-looking sonar image.

Description

technical field [0001] The invention belongs to the technical field of image processing, and in particular relates to a method, system and electronic equipment for segmenting a forward-looking sonar image target. Background technique [0002] The underwater small target boundary segmentation algorithm is a hot issue in the underwater image processing process in recent years, and it is also an important step in the target detection process. The segmentation accuracy of this part directly affects the accuracy of subsequent target recognition related steps, and is directly related to the discrimination of underwater target types. Forward-looking sonar is often used to detect and identify underwater targets. Therefore, it is very important to ensure the accuracy of target boundary segmentation in forward-looking sonar images. [0003] Due to the shortcomings of low signal-to-noise ratio, uneven brightness, and blurred target boundaries in the images collected by forward-looking...

AI Technical Summary

Problems solved by technology

[0003] Due to the shortcomings of the images collected by the forward-looking sonar, such as low signal-to-noise ratio, uneven brightness, and blurred target boundaries, the processing effect of the traditional level set segmentation method on the forward-looking sonar images often fails to achieve the expected effect.

Images