Address:Building B10,Huachuang Animation Park,Jinshan Village,Shiji Town,Panyu District,Guangzhou,China

In-depth analysis of machine vision
发布时间:2019-03-12 08:59:00 点击:

Machine visionconcept

In simple terms,the machine vision is toreplace human eyes with machines in terms of measuring and judging.Machine visionsystem is mainly composed of image acquisition, image processing and analysis,the output or display.Computer software is mainly used to simulate human visualfunction, extracting information from images of the objective things andprocessing, finally by used in the actual detection, measurement and control.

The parts of themachine vision system

The imageprocessing unit- the brain

  Image processing unit is completed by image processingsoftware, including a large number of image processing algorithms.After acquiringimage, image processing unit processes,analysis and calculate digital image andthe output results with the algorithm .Machine vision competition betweenenterprises is the accuracy of the algorithm of competition in the finalanalysis, so that each enterprise will invest a lot of resources to developcore software.Only excellent machine vision image processing software test rapidlyand accurately, and reduce the dependence of the hardware system.Software isthe brain of a machine vision,so the machine can identify,test, and so on. onlyafter the acquiring image digitalization,

The light source

  The light source is one of important components in themachine vision system,and  an appropriatelight source is a prerequisite for normal operation of the machine visionsystem.Light source illuminationcan make the measuredobject and the background difference as obvious as possible, to acquire high-qualityand contrast images.

The lens- theretina

  The function of lens is the optical imaging.Althoughcameras, analysis software, and lighting is very important for machine visionsystem , the most critical components is industrial camera lens.If system exertsits function, the lens must meet the requirements.The parameters determining thelens performance mainly include the focal length, working distance, and thedepth of field, resolution, etc.Depth of field is the range of distance fromthe before best focus and after best focus,acquiring the best image.Field of viewis the maximum range can be observe and usually expressed by angle. generally thelarger the field of view, the larger the observation range.Working distance isdistance from the lens to the object, the longer the working distance thehigher the cost.

The camera- the eye

  Machine vision camera (" eyeball ") transmitssensor image projected by the lens to the machine equipment with storage,analysis, and (or) shows function.According to the chip type,camera can bedivided into the CCD camera and CMOS camera.CCD and CMOS is common twokinds of image process technology now, the main difference between them lies inthe different transmission way.

The imageacquisition unit- the optic nerve

  The most important element of image acquisition unit isthe image acquisition card, which is the interface of image acquisition andimage processing part .Generally the imageacquisition unit hasthe following function modules: image signal receiving and A/D conversionmodule, responsible for image signal amplification and digital.The acquisitioncard can collect color or black-and-white images.Input color signals can bedivided into the compound or RGB component signal.Camera control input/outputinterface is mainly responsible for coordinating the cameras synchronous orasynchronous photoing, regular time photo, etc.Bus interface is responsible forthe internal bus high-speed digital output data via a PC and is generally a PCIinterface, transferring rate up to 130 MBPS, highly qualifying for highprecision image real-time transmission.and taking up less CPU time.

The output unit

  Input and output unit outputs the results of imageprocessing and acts correspond to the result, such as eliminating wastes, alarmlamp alarming and displaying production information with the man-machineinterface after finished the image acquisition and processing,

 The application ofmachine vision domain

Visual technologyis one of the biggest advantages no damage toboth the observer and the observed because of observedobjects without contact, safe and reliable.In theory,Machine vision can alsoobserve the band range that the human eye cannot observe,such as infrared ray,microwave, ultrasonic, etc., so machine vision can take infrared image,microwave image, ultrasonic image, etc, through those kinds of sensors.In addition,machine vision can observe object over a long period of time, no time limit, withthe very high resolution precision and speed,but humans not.

  So, machine vision application field is very wide, canbe divided into four areas including the industrial, scientific research,military and civilian.Machine vision is widely used in intelligent integratedtest system for the instrument panel, Automatic Injury Control System for MetalPlate Surface, automobile body detection system, Quality Inspection of PaperMoney Printing, intelligent traffic management, Metallographic analysis,medical imaging analysis, assembly line production inspection and so on.

At present, nearly80% of industrialvision system is mainly used in the test, to improve the production efficiency,control the quality of the products in the process of production and collect productdata, etc.Machine vision automation equipment can work tirelesslyand repetitively and meet the requirements of some dangerous workingenvironment not suitable for manual operation, taking the place of human beings.In the Trend ofIndustrial 4.0,machine vision industry is flourishing, application of machine vision productsin other downstream industry still has great development space.