Fuzzy Image Corner Extraction Method for Mine Shaft Environment

Abstract

The invention relates to a method of extraction of a corner of a blurred image in a mine shaft environment. The method of the extraction of the corner of the blurred image in the mine shaft environment comprises the steps of carrying out area judgment on the blurred image by using an inner template, rapidly recognizing a flat domain and a corner domain, and avoiding further judgment of the flat domain; dividing an image in the corner domain into a background area and a foreground area by using a maximum entropy algorithm and combining area characteristics so as to select corresponding corner domain values according to different areas to carry out corner extraction; extracting a candidate corner by using an outer template according to a corner domain state of a detection point; carrying out a fake-removing operation on the candidate corner, removing burr points on the edges and faked corners which satisfy a corner response function and are arranged on a narrow band, and obtaining a final real corner. According to the experimental results, compared with an existing corner extraction algorithm, the method of the extraction of the corner of the blurred image in the mine shaft environment has a high extraction precision and good robustness under the premise of guaranteeing instantaneity when the method of the extraction of the corner of the blurred image in the mine shaft environment is applied to blurred mine shaft images with uneven illumination.

Description

technical field [0001] The invention relates to the technical fields of mine safety and digital image processing, in particular to a fuzzy image corner extraction method used in the mine shaft environment. Background technique [0002] Illumination is uneven in the mine shaft environment. With the development of image processing technology, real-time splicing of mine shaft video scenes and intelligent detection of fault points have attracted attention. Corner extraction for images is a basic in the field of digital image processing and computer vision. The key issue is to realize the registration and stitching of mine shaft images, scene analysis and fault detection. Specifically, corner point extraction refers to the process of extracting a small number of corner points in the image that contain rich information and important features of the image by using certain characteristics of the digital image, such as color, shape, and grayscale. [0003] In recent years, many rese...

AI Technical Summary

Problems solved by technology

Window filtering and template operation have a direct impact on the time complexity of the algorithm. Small templates and small windows have a small amount of calculation but increase the probability of extracting wrong corners, while larger templates and windows increase the calculation amount of the algorithm.

Existing corner point extraction algorithms cannot meet the real-time requirements of wellbore image processing due to long calculation time

Images