Method and system for generating edge seam path and edge seam topological structure of raster image

Abstract

The invention discloses a method and a system for generating edge seam path and edge seam topological structure of raster image, which belongs to technical field of image edge detection. The method comprises S1, selecting a raster image, and carrying out edge detection calculation by calculating a gradient magnitude diagram to obtain boundary points of the image; S2, selecting a boundary point of step S1, and carrying out neighborhood tracing according to gradient magnitude of pixels to obtain the second boundary point; S3, converting the edge seam gradient of the second boundary point in the vertical direction in reference to the first boundary point to the edge seam path; and S4, repeating the steps of S2 and S3 until the edge seam path of the whole raster image is completed; and recording information of the edge seam path and intersection points to generate the edge seam topological structure. The boundary generated by the method has high location accuracy, good continuity and abundant boundary information, and can meet the application requirement for boundary detection in each hot field of image processing.

Description

technical field [0001] The invention relates to the field of image processing, in particular to a method and system for generating a raster image edge seam path and edge seam topological structure. Background technique [0002] In image processing, color image boundary detection technology has extremely broad application prospects, and plays an important role in many hot fields such as image recognition, matching, 3D reconstruction, image retrieval, and image segmentation. [0003] There are many boundary detection algorithms, but after the candidate boundary pixels are generated, there are two edge seam tracking algorithms (four-neighborhood tracking) and eight-neighborhood tracking algorithms that are more effective in boundary linked list tracking algorithms. The edge seam tracking algorithm (four-neighborhood tracking) generally uses the direct calculation of the edge seam gradient in the horizontal and vertical directions to track the boundary, and the positioning is re...

AI Technical Summary

Problems solved by technology

The edge seam tracking algorithm (four-neighborhood tracking) generally uses the direct calculation of the edge seam gradient in the horizontal and vertical directions to track the boundary, and the positioning is relatively accurate. However, since there are only four neighborhood directions and the edge seam gradient is unstable, the generated The edge seam boundary is very easy to break; the eight-neighborhood tracking algorithm generally uses the calculation of the gradient magnitude of the pixel point, and tracks the gradient magnitude in eight directions to track the boundary, which has good continuity, but the positioning accuracy can only reach the pixel level, which is not enough accurate

Images