Skip to content
DSML.KZ
Publications

Talk · EPJ Web of Conferences · January 1, 2025

Hybrid quadratic diagonal algorithm for thinning contour lines

Narzillo Mamatov, Malika Jalelova, Vohid Fayziyev, Abdurashid Samijonov, Boymirzo Samijonov

Narzillo Mamatov
Open via DOIPDF

Abstract

One of the main issues of image analysis is the separation of contour lines. Nowadays, many effective methods for dividing contour lines have been developed. In solving some practical problems, the results obtained by contour separation methods will not be enough, that is, operations such as thinning, filling, and smoothing of contour lines are required. In this case, the development of an efficient contour thinning algorithm used for accurate separation of the shape of the object is an urgent issue. Contour thinning algorithms can reduce the amount of data to be processed and increase processing speed. Based on the literature analysis, the Zhang-Suen algorithm can be recognized as the most efficient among the contour thinning algorithms due to the efficiency and speed of preserving the shape of the objects in the images. However, this algorithm fails to thin some contour lines. Therefore, in this work, an improved quadratic diagonal algorithm based on the strengths of the Zhang-Suen algorithm is proposed. Also, the proposed algorithm is compared with the existing algorithms regarding error and time criteria in contour detection. By conducting experimental studies, the Hybrid Quadratic Diagonal Algorithm showed the smallest error compared to the algorithms obtained for testing in the experiment.

Digital Image Processing TechniquesMedical Image Segmentation TechniquesComputer Science and Engineeringbinary image processingshape preservation

More by Narzillo Mamatov