z-logo
open-access-imgOpen Access
Image segmentation application combined with DRLSE and moving mesh method
Author(s) -
Wan-lung Lee,
Hongcan Shi
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/2024/1/012019
Subject(s) - computer science , image segmentation , level set method , artificial intelligence , computer vision , segmentation , segmentation based object categorization , scale space segmentation , process (computing) , image (mathematics) , pixel , level set (data structures) , image processing , grid , mathematics , geometry , operating system
Aiming at the time-consuming problem of level set method in processing image segmentation, this paper presents an image segmentation algorithm combining DRLSE and moving mesh method. This method is applied to segment images into multiple blocks in different initial regions. When a large image needs to be segmented into multiple blocks, the level set method has to calculate various data at each pixel. So, there are problems such as low-efficiency and time-consuming processing. In order to solve this problem, we propose a moving mesh method to process image segmentation based on the level set method. By calculating the monitor function about the image gradient, the grid encryption is automatically realized with the change of image gradient. This makes the calculation of the image segmentation more accurate. In order to improve efficiency, our method will reduce the grid to 1/4. Then the level set method is applied to achieve the result of computational acceleration. In order to verify the performance and accuracy of the new method, our method segmented some images several times. Experiments show that our method can quickly and automatically segment some images into multiple blocks without affecting the results.

The content you want is available to Zendy users.

Already have an account? Click here to sign in.
Having issues? You can contact us here