Improved morphologic filtering method in image processing

Abstract

The invention relates to an improved morphological wave filtering method in imaging processing of the technical field of image processing, comprising steps of firstly defining an inner structure element and an outer structure element so as to form an annular structure element; scanning each pixel in an image from up to down and from left to right through the annular structure element, respectively dividing the annular structure element into eight directional windows at each pixel to be processed, respectively selecting a middle value of image pixel grey scale values in each directional window, selecting the minimal grey scale value from the middle values as a grey scale value of the current processed pixel point to obtain a tainted image; carrying out traditional expansion operation to the tainted image with a flat structure element to restore grey scale transplacement and domain shrinkage caused by tainting operation. The invention can effectively maintain geometrical features of a target in an image, removes bright noise and dark noise in the image by implementing operation processing sequentially, and improves processing efficiency and property.

Description

technical field [0001] The invention relates to a method in the technical field of image processing, in particular to an improved morphological filtering method in image processing. Background technique [0002] Image processing based on mathematical morphology is a relatively novel image processing technology emerging in recent years, which analyzes and describes the geometric shape of signals from the perspective of collection. Now, mathematical morphology has been widely used in image processing and analysis, image denoising, pattern recognition, artificial intelligence, machine vision, industrial inspection, biomedical engineering and other fields, playing a huge role and constantly developing and improving. Morphological filtering, in essence, can be decomposed into two basic problems of morphological transformation and structural element selection. Erosion and dilation are two basic operations of mathematical morphology, which can be combined to form many morphologica...

AI Technical Summary

Problems solved by technology

This overcomes the fact that the traditional morphological opening or closing operation can only remove one kind of noise at a time, and the performance of the improved morphological Top-Hat operator decreases when the noise density is large and the image has a gray gradient, and the median filter cannot keep the target in the image Disadvantages such as geometric properties

Images