Repose angle detection method based on gradient image segmentation

Abstract

The invention discloses a repose angle detection method based on gradient image segmentation. The method comprises the following steps: A1, carrying out the two rounds of filtering of pixels, carryingout the filtering of a first round of filtering in a gradient direction, and filtering out horizontal and vertical pixel points in the gradient direction; in the second round, filtering is carried out according to the gradient amplitude, a threshold value is adaptively calculated for each row of the image, and pixels with excessively small gradient amplitude are filtered out; a2, performing linear detection by calculating the similarity between the pixels; and A3, finally clustering the straight lines, detecting the left edge and the right edge of the coal pile, and calculating the repose angle. By means of the repose angle detection method, repose angle detection is conducted on the bulk goods storage yard material pile, the stacking and taking range of the materials is pre-judged, and the materials are prevented from exceeding the stacking area in the storage yard process.

Description

technical field [0001] The invention relates to the technical field of computer vision, in particular to a method for detecting an angle of repose based on gradient image segmentation. Background technique [0002] The estimation of the angle of repose of the coal pile is based on the assumption that during the free-fall accumulation process of the coal pile, the projection of the coal pile on the image (which can be approximately regarded as the longitudinal section of the coal pile) is an isosceles triangle. Therefore, the key to calculating the angle of repose of the coal pile lies in the edge extraction of the longitudinal section of the coal pile. After decades of research, there have been many edge extraction algorithms. However, the existing mainstream edge extraction methods such as canny, ED, EDLine, LSD, etc., often rely on points with high thresholds as key points, and the subsequent edge connection work is based on key points. Point-based completion. Although m...

AI Technical Summary

Problems solved by technology

Although many people have proposed the method of adaptively calculating the threshold value, the threshold value is the global threshold value without exception, and in the actual stockyard image, the coal pile part is darker overall, and the surrounding environment often presents a higher contrast, making Part of the coal pile is difficult to detect

Images