A high-precision extraction method and system of image real boundary

Abstract

The invention provides a high-precision extraction method and system for the real boundary of an image, which converts the original image into a binary image and performs boundary expansion; and uses the Suzuki contour tracking detection algorithm to obtain the outermost boundary contour line in the expanded binary image. The pixel center of the inner 1 pixel; set the corresponding coordinate offset to obtain the pixel coordinates of the outermost boundary pixel on the unexpanded binary image; traverse all the boundary pixels, and determine the next boundary pixel and the current one one by one. The positional relationship of the 8-nearest neighbors of the boundary pixels, adding the true boundary vertices to generate a true boundary vertex set with no pixel difference; removing the duplicate boundary vertices to obtain a non-redundant true boundary vertex set; judging the continuity of the true boundary vertex set, Obtains a non-redundant and contiguous set of true boundary vertices; when compressing the image boundary is selected, the compression saves the set of boundary vertices. The invention can solve the defect of Suzuki contour tracking detection, so as to obtain the real and complete outermost boundary of the image.

Description

technical field [0001] The invention belongs to the field of digital image processing, and in particular relates to a high-precision extraction method and system for the real boundary of an image. Background technique [0002] Boundary or edge extraction refers to a process for image contours in digital image processing. In a grayscale image, for the boundary, the place where the grayscale value changes most drastically is defined as the edge. In a binary image, the boundary between the white area (value 255) and the black area (value 0) is the border, where the outer border is defined as the space between any group of 1 pixels and the group of 0 pixels directly surrounding it The hole border is defined as the set of 1 boundary points between the hole and the group of 1 pixels directly surrounding it. The analysis and research of image real boundary extraction has extremely high application value in the fields of image stitching (especially high-precision stitching of a ce...

AI Technical Summary

Problems solved by technology

Through this technical solution, two problems existing in the Suzuki contour tracking detection method can be solved when tracking the outer boundary: (1) Sometimes the boundary points of some areas on the boundary may not be obtained (2) There is generally a 0.5 pixel distance deviation in obtaining the boundary points , the main purpose is to obtain the real and complete outermost boundary of the image, thereby significantly improving the integrity and accuracy of the image's outermost boundary extraction, and avoiding cumulative errors and seam lines when stitching multiple non-real boundary images Gap and other issues

Images