Mathematical morphology adaptive structure element construction method

Abstract

The invention discloses a mathematical morphology adaptive structural element construction method, and belongs to the technical field of structural element construction methods. The method comprises the steps that firstly, local smoothing is conducted on an image through a neighborhood gray level difference change vector field, so that the interior of an image target has similar gray levels, and then whether a central pixel point of the local area is a boundary point s determined by defining a boundary degree function based on relative density, wherein the adaptive structure element is composed of all strong boundary points determined by the local area. On the basis of the self-adaptive structural elements, self-adaptive morphological expansion and corrosion and related derivative operations are constructed according to a classical morphological operation principle. The method has the beneficial effects that morphological operation is performed on the image by adopting the self-adaptive structural element, the occurrence of a target object boundary contour positioning offset phenomenon can be effectively inhibited, the accuracy of target area boundary contour positioning is kept, other effective information of the image is kept while the details of the image are weakened, and the distortion degree of the image is reduced.

Description

technical field [0001] The invention belongs to the technical field of structural element construction methods, and relates to a mathematical morphology self-adaptive structural element construction method, in particular to a morphological self-adaptive structural element construction method based on neighborhood gray scale difference change vector field and relative density. Background technique [0002] Mathematical morphology is a theory based on set, integral geometry and grid algebra to analyze the geometric structure characteristics of images. It uses structural elements with a certain shape and size to traverse the pixels in the image one by one and perform morphological operations to realize image Processing such as structural analysis and feature extraction. The morphological operation on the image can remove the irrelevant structure in the image, simplify the image data, and effectively extract the nonlinear features such as the shape and structure of the image obj...

AI Technical Summary

Problems solved by technology

However, in practical applications, due to the complex and diverse content of the image target, the shape, size and direction of the target are different, the classical morphological operation uses structural elements with fixed shape and size to process the image, which may easily lead to changes in the target feature information in the image. For example, the positioning of the target boundary contour shifts, new artificial targets appear, or small targets are lost, etc.

Images