Machine vision-based remote sensor light shading hood honeycomb defect automatic detection method

Abstract

The invention discloses a machine vision-based remote sensor light shading hood honeycomb defect automatic detection method. The method comprises the following steps of: 1, light shading hood honeycomb image acquisition: a light shading hood honeycomb image is acquired; 2, preprocessing: the acquired light shading hood honeycomb image is preprocessed so as to reduce the noises of the image; 3, edge straight line segment feature extraction: feature extraction is performed on the preprocessed light shading hood honeycomb image, so that light shading hood honeycomb edge straight line segment features are obtained; 4, feature vector establishment and positive and negative sample screening: feature description is performed on the obtained light shading hood honeycomb edge straight line segment features, light shading hood honeycomb feature vectors are established, the feature vectors of a normal honeycomb and a defective honeycomb are selected as positive samples and negative samples respectively; and 5, artificial neural network establishment and training. With the method of the invention adopted, the production efficiency of a remote sensor light shading hood can be effectively improved, and the labor intensity of quality inspection personnel can be reduced.

Description

technical field [0001] The invention relates to the technical field of automatic detection based on machine vision, in particular to an automatic detection method for honeycomb defect of a remote sensor shading cover based on machine vision. Background technique [0002] The remote sensor hood is installed on the front end of the remote sensor optical lens to shield harmful light, suppress the halo of the picture, and prevent stray light from entering the lens; the common structure of the hood is a square pyramid, and the inner wall of the hood adopts a closely arranged honeycomb structure form. [0003] In order to achieve multiple extinction effects, the shading cover adopts a physical extinction method to increase the light absorption of the inner surface of the shading cover and reduce the reflection; on the one hand, a closely arranged honeycomb structure is often used on the inner wall of the shading cover. A regular hexagonal prism with a thick wall and an opening at...

AI Technical Summary

Problems solved by technology

[0004] In order to ensure the quality of spraying, it is required that the honeycomb structure is not allowed to have defects such as lodging and edge curling before spraying; in the field of automatic spraying, the automatic detection technology for smooth curved surfaces is relatively mature, but the automatic detection of complex honeycomb surfaces has not yet been reported. ; In order to further improve the production efficiency of the remote sensor shading cover and reduce the labor intensity of the quality inspection personnel, an automatic detection method for honeycomb defects of the remote sensor shading cover is needed

Images