当前位置:网站首页>The "eye" of industrial robot -- machine vision
The "eye" of industrial robot -- machine vision
2022-06-26 03:15:00 【Double vision】
Preface
Industrial robots “ Insight ”
As early as 20 century 80 In the s, the National Bureau of standards predicted that , Detection task of 80% Even 90% This will be done by a visual measurement system , This prediction has basically become a reality . In recent years , With the rapid development of machine vision technology , The rapidity of machine vision technology 、 Accuracy 、 Intelligent features have been widely used in all walks of life in modern industry .
and , In the current industry with intelligent manufacturing as the core 4.0 Against the background of the times , Made in China 2025 The strategic deployment is gradually deepened , The industrial robot industry market shows an explosive growth momentum , And act as an industrial robot “ Insight ” The machine vision system is also important .

Machine vision system is a non-contact optical sensing system , It integrates both hardware and software , It can automatically obtain information from the collected images or generate control actions .
in short , It is to simulate the visual function of the human eye with a computer , Extract information from images or image sequences , To deal with and understand , Final for inspection 、 Measurement and control .
A typical industrial machine vision system uses a camera to capture the image information of the measured object , Then the analog image information is converted into digital image signal and transmitted to the digital image processing system of the upper computer , In the processing system, according to the pixel distribution of the picture 、 brightness 、 Information such as marking points and colors , Calculate the position, posture, shape and other information of the measured target , Finally, according to the information obtained from the test, control the operation of driving the actuator to respond .
At present, machine vision has been applied to the sorting of agricultural products 、 Medical imaging 、 In the fields of product packaging inspection and industry .

Machine vision is widely used in industry , There are three main functions : Visual measurement 、 Visual guidance and visual inspection .
Visual measurement is aimed at some parts with high accuracy requirements , The accuracy shall be millimeter level or even micron level , It cannot be done by human eyes, but must be done by machine , For example, the size of high-precision screw holes needs to be measured by machine vision , Ensure the clearance and accuracy of the connection ;
Visual guidance requires that machine vision can quickly and accurately find the parts to be tested , And confirm its position , Guide the robot manipulator to grasp accurately , Such as visual guided random grab , Three dimensional images are obtained by scanning the parts randomly distributed in the tool box , Using pattern recognition , Obtain the best point for the manipulator to grasp the workpiece in the three-dimensional image , Guide grab to realize automatic production ;
Visual inspection is to use the machine vision system to check whether the products on the production line have quality problems , For its beauty 、 Test the comfort and performance , It is the link that replaces the most manual work , For example, machine vision has been used to detect the geometric shape of potatoes at home and abroad 、 Detection of surface defects , From grading potatoes according to these characteristics .
Machine vision has greatly improved the flexibility and automation of industrial production , And it can complete some tasks that are difficult to be completed manually in dangerous working environment , The use of labor is greatly reduced in mass production , And improve and ensure the quality of production .

It's worth pointing out , Machine vision technology has been widely used in automobile industry, a typical representative of modern industry .
As one of the largest and most important industries in the world, the automobile manufacturing industry has 100 Years of development history . Compared with other manufacturing industries , The automobile manufacturing industry has several typical characteristics : Production is huge and still growing rapidly ; The model updating speed is fast ; The manufacturing process is very complicated . The application of machine vision technology in automobile manufacturing industry , The operation quality and efficiency of the process are greatly improved , Reduced labor intensity .
It has been successfully applied in many automobile OEMs at home and abroad , Including automatic detection of vehicles and parts 、 Part 3D positioning 、 Body assembly / machining , Part traceability, etc . The application of machine vision technology has run through the whole automobile body manufacturing process , Including the development from initial raw material quality inspection to auto parts 100% on-line measurement , Then welding in the manufacturing process 、 Gluing 、 Control punching and other technological processes , Finally, the body assembly 、 Check the quality of the finished vehicle .

at present , Machine vision has been widely used in various fields of modern industry , In the future, the development space will be broader , Machine vision technology must become a higher guidance 、 faster 、 More stable in the era of automation industry “ Insight ”.
But I have to point out , The application of machine vision technology in modern industry is also restricted by certain factors , Targeted improvement and promotion are needed in the future . in addition , At present, the intelligence of computer software is far from enough , Deep learning neural network, which is popular in the field of computer vision, has less successful application in the field of machine vision , When the target object changes 、 Complex features 、 When the number of samples is not enough , Deep learning cannot be used , We should return to the traditional way , Then consider the strict requirements of real-time , Machine vision especially needs a new intelligent method which is widely used in most fields .
Industry is basically a three-dimensional component , After all, two-dimensional imaging is a morbid data acquisition of the actual situation in three-dimensional space , Around three-dimensional machine vision inspection 、 measurement 、 Robot guidance and other projects and applications will be more and more , Three-dimensional visual inspection will be more applied to the field of industrial production .
There will also be a climax in the development of machine vision in many other civil fields , Modern manufacturing 、 Medical care 、 Electronics 、 Warehousing and other fields will have a variety of new applications , Promote the development of intelligence in the new era ; Machine vision hardware system will also develop towards embedded , Operation efficiency and intelligence are getting higher and higher .
边栏推荐
- Dreamcamera2 video recording, playing without sound, recording function is normal, using a third-party application for video recording, playing with sound
- Additional: brief description of hikaricp connection pool; (I didn't go deep into it, but I had a basic understanding of hikaricp connection pool)
- 限制输入字符长度length英文1个字符中文2个字符
- Leetcode 176 The second highest salary (June 25, 2022)
- ORB-SLAM3论文概述
- 360 秒了解 SmartX 超融合基础设施
- 解析少儿编程的多元评价体系
- Review of the paper: unmixing based soft color segmentation for image manipulation
- Dreamcamera2 record key prompt sound into video during video recording
- Lumen Analysis and Optimization of ue5 global Lighting System
猜你喜欢

OpenAPI 3.0 specification - Food Guide

小 P 周刊 Vol.10

解析少儿编程的多元评价体系

UE5全局光照系统Lumen解析与优化

Fresh graduates talk about their graduation stories

MySQL增删查改(进阶)

Camtasia 2022 nouvelle vidéo d'ordinateur d'enregistrement ultra - clair

Vulhub replicate an ActiveMQ

MySQL数据库基础

Multi surveyor Gongshu Xiao sir_ The solution of using websocket error reporting under working directory
随机推荐
Matlab| short term load forecasting of power system based on BP neural network
R language Markov chain Monte Carlo: practical introduction
scrapy返回400
Notes on the 3rd harmonyos training in the studio
数字孪生智慧水务,突破海绵城市发展困境
UE5全局光照系統Lumen解析與優化
在UE内添加控制台工程(Programs)
Multi surveyor Gongshu Xiao sir_ The solution of using websocket error reporting under working directory
【QT】自定义控件-空气质量仪表盘
Install development cross process communication
2021-08-04
培育项目式Steam教育理念下的儿童创造力
Gd32 ADC acquisition voltage
The "more" option of Spreadtrum platform dearmcamera2 is removed
Interpreting Oracle
On virtual memory and oom in project development
经典模型——NiN&GoogLeNet
项目部署遇到的问题-生产环境
P2483-[模板]k短路/[SDOI2010]魔法猪学院【主席树,堆】
【论文笔记】Manufacturing Control in Job Shop Environments with Reinforcement Learning