A line structured light center line extraction method

Abstract

The invention belongs to the technical field of machine vision, and discloses a line structured light center line extraction method, which comprises the following steps of: cutting an image shot by acamera, extracting features, performing preprocessing and correcting a background; selecting a segmentation threshold value, and carrying out binarization processing on the corrected light bar image;carrying out Refinement processing by adopting a refinement algorithm, and carrying out closed operation on the image after refinement processing to obtain an image containing a single-pixel light barcenter line; Selecting m pixel points from the center line of the single-pixel light bar, converting the m pixel points into corresponding image coordinates, carrying out distortion correction on theimage coordinates, and carrying out fitting by utilizing the image coordinates subjected to distortion correction to obtain the center line of the line structured light. According to the method, theproblem of non-uniform image background can be solved, the background gray level is equalized, and the segmentation precision is improved, so that the structured light strip image with a certain pixelwidth is obtained, the extraction speed, the extraction precision and the accuracy of center line extraction are improved, and the real-time requirement of a visual detection system is met.

Description

technical field [0001] The invention relates to the technical field of machine vision, in particular to a method for extracting a center line of line structured light. Background technique [0002] Structured light is a system structure composed of a projector and a camera. It has the characteristics of non-contact, fast dynamic response, good system flexibility, low cost, and small size. It has been widely used in product processing quality control, rapid design, and automatic control. , reverse engineering and automatic control and many other fields. [0003] Structured light is based on the principle of laser triangulation for non-contact measurement. The principle is: use a laser to generate line laser light, irradiate the object to be measured, take a light strip image from a certain angle through a camera, and finally obtain three-dimensional information on the surface of the object based on the principle of optical triangulation. Therefore, the accuracy of obtaining ...

AI Technical Summary

Problems solved by technology

However, the center of gravity method scans according to each row. For curved stripes, its extraction accuracy will be affected to a certain extent, and the algorithm is complex, resulting in slow extraction, which cannot meet the measurement speed requirements of the 3D inspection system.

The direction template method is more complicated and has low operating efficiency.

Images