CN111415405A

CN111415405A - Three-dimensional high-precision vision measurement method for workpiece with high light-reflecting surface

Info

Publication number: CN111415405A
Application number: CN202010182224.4A
Authority: CN
Inventors: 葛继; 宫正; 刘国营; 孙钰; 汝长海
Original assignee: Suzhou Inov Intelligent Technology Co ltd
Current assignee: Suzhou Inov Intelligent Technology Co ltd
Priority date: 2020-03-16
Filing date: 2020-03-16
Publication date: 2020-07-14

Abstract

The invention discloses a three-dimensional high-precision vision measurement method for a high-reflection surface workpiece, which comprises the following steps of (1) setting a structured light projection brightness value interval [ L ]_min,L_max]The number of groups of the projection structured light is N; (2) sequentially projecting N groups, and shooting 2 × M brightness values in each group by using a 3D camera

Wherein i is 0,1,2 …, N-1; (3) fusing the N groups of phase shift grating images into 1 group of 2 x M images by an image fusion method; (4) finally, reconstructing the fused 2 x M frame images into an accurate 3D point cloud picture of the object to be detected; wherein M and N are both positive integers. The invention has the advantages of: by means of high-speed structural light brightness sequential switching and high-speed image fusion, the problems of poor 3D point cloud reconstruction quality and low measurement precision caused by high reflection of light on the surface of a workpiece are solved.

Description

Three-dimensional high-precision vision measurement method for workpiece with high light-reflecting surface

Technical Field

The invention relates to the field of 3D, in particular to a three-dimensional high-precision vision measurement method for a workpiece with a high light-reflecting surface.

Background

Three-dimensional measurement techniques have been widely used in (1) surveying and mapping projects, such as topographic survey, highway surveying and mapping, site surveying and mapping, and cultural relic repair; (2) national security, such as anti-terrorism, mobile reconnaissance, forest fire monitoring; (3) entertainment industries such as 3D game development, virtual reality; (4) reverse engineering; (5) product quality management and the like.

With the rapid development of modern manufacturing industry, the measurement and detection of high-speed high-precision 3D topography features of large-scale parts have gradually become a new direction of attention in the field of three-dimensional measurement, such as the three-dimensional measurement of profiles of large objects such as airplanes and engine blades in the field of aerospace, automobiles and ships in the field of transportation, which also puts higher requirements on the three-dimensional topography measurement.

Taking the automobile industry as an example, automobile part and whole automobile part manufacturers gradually improve the product quality control standard to meet the aggravation of industry competition and the severe requirements of the market on the product quality, and the manufacturers can accurately detect and measure the quality (including appearance, spatial form and position geometric dimension and the like) of each product, the existing three-dimensional measurement mainly comprises (1) an off-line measurement system represented by a three-coordinate measuring instrument, in a three-dimensional measurable spatial range, the three-coordinate measuring instrument detects a workpiece point by point according to a measuring head system and returns point data on the surface of the workpiece, and various geometric shapes, dimensions and the like are finally calculated through a software system.

The traditional technology has the following technical problems:

however, the existing 3D vision measurement technology and system based on structured light still have disadvantages, for example, when there is a severe reflected light on the surface of the workpiece to be measured, the detail features will be lost or the precision will be greatly reduced.

Disclosure of Invention

The invention aims to provide a three-dimensional high-precision vision measurement method for a workpiece with a high-reflection surface, a high-speed structural light brightness sequence switching and high-speed image fusion method, and solves the problems of poor 3D point cloud reconstruction quality and low measurement precision caused by high reflection of the surface of the workpiece.

In order to solve the technical problem, the invention provides a three-dimensional high-precision vision measurement method of a high-reflection surface workpiece, which comprises the following steps:

(1) setting the projection brightness value interval of the structured light [ L ]_min,L_max]The number of groups of the projection structured light is N;

(2) sequentially projecting N groups, and shooting 2 × M brightness values in each group by using a 3D camera

Wherein i is 0,1,2 …, N-1;

(3) fusing the N groups of phase shift grating images into 1 group of 2 x M images by an image fusion method;

(4) finally, reconstructing the fused 2 x M frame images into an accurate 3D point cloud picture of the object to be detected;

wherein M and N are both positive integers.

In one embodiment, M is 12.

In one embodiment, the left and right cameras inside the 3D camera each take M pictures.

In one embodiment, in step (3), the images taken by the left camera inside the 3D camera are merged into M images, and the images taken by the right camera are merged into M images.

In one embodiment, the image fusion method employs an extreme learning machine.

In one embodiment, the image fusion method adopts a support vector machine or a back propagation method.

In one embodiment, in the step (4), the fused 24 frames of images are reconstructed into an accurate 3D point cloud picture of the object to be detected by a triangulation method.

Based on the same inventive concept, the present application also provides a computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of any of the methods when executing the program.

Based on the same inventive concept, the present application also provides a computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of any of the methods.

Based on the same inventive concept, the present application further provides a processor for executing a program, wherein the program executes to perform any one of the methods.

The invention has the beneficial effects that:

by means of high-speed structural light brightness sequential switching and high-speed image fusion, the problems of poor 3D point cloud reconstruction quality and low measurement precision caused by high reflection of light on the surface of a workpiece are solved.

Drawings

FIG. 1 is a flow chart of a three-dimensional high-precision vision measuring method of a high-reflectivity surface workpiece according to the invention.

Detailed Description

The present invention is further described below in conjunction with the following figures and specific examples so that those skilled in the art may better understand the present invention and practice it, but the examples are not intended to limit the present invention.

Aiming at a workpiece with a high-light-reflection surface, which kind of suitable brightness is difficult to determine to project light bars so as to obtain high-quality workpiece point cloud data, the invention provides a high-speed structural light brightness sequence switching and high-speed image fusion method, which comprises the following specific implementation steps (a specific flow chart is shown in fig. 1):

(2) n sets of 24 shots (12 shots from each of the left and right cameras within the 3D camera) are projected sequentially with brightness values of

Wherein i is 0,1,2 …, N-1.

(3) N groups of phase-shift grating images are fused into 1 group of 24 images (12 images shot by a left camera and 12 images shot by a right camera inside a 3D camera) by an image fusion method, wherein the fusion method adopts an Extreme learning Machine (E L M) method, and the fusion speed is thousands times of that of the most common methods such as Support Vector Machines (SVMs) and back propagation methods under the condition of equal fusion precision.

(4) And finally, reconstructing the fused 24 frames of images into an accurate 3D point cloud picture of the object to be detected by a triangular transformation method.

The above-mentioned embodiments are merely preferred embodiments for fully illustrating the present invention, and the scope of the present invention is not limited thereto. The equivalent substitution or change made by the technical personnel in the technical field on the basis of the invention is all within the protection scope of the invention. The protection scope of the invention is subject to the claims.

Claims

1. A three-dimensional high-precision vision measurement method for a workpiece with a high-light-reflection surface is characterized by comprising the following steps:

Wherein i is 0,1,2 …, N-1;

wherein M and N are both positive integers.

2. The method for three-dimensional high-precision visual measurement of highly reflective surface workpieces according to claim 1 wherein M is 12.

3. The method for three-dimensional high-precision vision measurement of a highly reflective surface workpiece according to claim 1, wherein M images are taken by each of left and right cameras inside the 3D camera.

4. The method for three-dimensional high-precision visual measurement of a workpiece with a highly reflective surface according to claim 1, wherein in step (3), the images taken by the left camera and the images taken by the right camera inside the 3D camera are merged into M images.

5. The method for three-dimensional high-precision visual measurement of a highly reflective surface workpiece according to claim 1, wherein the image fusion method employs an extreme learning machine.

6. The method for three-dimensional high-precision visual measurement of a highly reflective surface workpiece according to claim 1, wherein the image fusion method employs a support vector machine or a back propagation method.

7. The method for three-dimensional high-precision visual measurement of highly reflective surface workpieces according to claim 1, wherein in step (4), the fused 24 frames of images are reconstructed into an accurate 3D point cloud image of the object to be detected by a triangulation method.

8. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the steps of the method of any of claims 1 to 7 are implemented when the program is executed by the processor.

9. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 7.

10. A processor, characterized in that the processor is configured to run a program, wherein the program when running performs the method of any of claims 1 to 7.