CN205538710U - Inductance quality automatic check out system based on machine vision - Google Patents

Abstract

The utility model discloses an inductance quality automatic detection system based on machine vision, which relates to the technical field of inductance detection. The LED light source is matched with the CCD camera, the CCD camera is connected to the image acquisition card, the image acquisition card is connected with the computer system with image processing software installed, the computer system is connected with the CRT display, and the computer system is connected with the control system for controlling the inductance; the computer system consists of The image processing unit, the edge positioning unit, the damage detection unit and the loose wire detection unit are composed, the image processing unit is connected with the edge positioning unit, the edge positioning unit is respectively connected with the damage detection unit, the loose wire detection unit, the damage detection unit, the loose wire detection unit and Control system connection. The utility model realizes the integration of production, detection and packaging, the detection is stable and reliable, the production efficiency is improved, the labor intensity of workers is reduced, the production cost is reduced, and it is easy to popularize and use.

Description

An automatic inspection system for inductance quality based on machine vision

technical field

The utility model relates to the technical field of inductance detection, in particular to an automatic inductance quality detection system based on machine vision.

Background technique

Surface defects will directly affect product quality and performance. Therefore, in the inductance quality inspection system, the automatic detection of surface defects is an indispensable link. The traditional contact manual inspection method is not only cumbersome, labor-intensive, but also the detection speed It is slow, cannot eliminate human-made measurement errors, and may cause certain damage to the surface of the object during the detection process, which makes the traditional detection method fail to meet the ideal requirements.

With the further strengthening of industrial automation, relying on traditional manual inspection can no longer meet the speed requirements of production automation. Automatic online inspection has become an urgent need for many automatic manufacturers. For some appearance defects such as core breakage, relying on manual inspection cannot meet the requirements of automation speed and reduce production efficiency.

In order to solve the above problems, those skilled in the art are devoting themselves to developing an automatic detection system for inductance quality based on machine vision.

Utility model content

In view of the above-mentioned defects of the prior art, the technical problem to be solved by this utility model is to provide an automatic detection system for inductance quality based on machine vision. Efficiency, reduce the labor intensity of workers, reduce production costs, strong practicability, easy to popularize and use.

In order to achieve the above object, the utility model provides a kind of automatic detection system of inductance quality based on machine vision, including LED light source, CCD camera, image acquisition card, computer system, CRT display, image processing software device and control system, LED light source and The CCD camera is matched, the CCD camera is connected to an image acquisition card, the image acquisition card is connected to a computer system equipped with an image processing software device, the computer system is connected to a CRT display, and the computer system is connected to a control system for controlling the inductance.

As preferably, the computer system is composed of an image processing unit, an edge positioning unit, a damage detection unit and a loose wire detection unit, the image processing unit performs some preprocessing on the inductance picture obtained by the image acquisition card, and the image processing unit is connected to the edge positioning unit, The edge positioning unit locates the position of the inductor in the image, can accurately find the position of the workpiece, and facilitates the positioning and tracking of the subsequent processing program. The edge positioning unit is connected with the damage detection unit and the loose wire detection unit respectively, and can detect the damage and loosening of the inductor. Detection, the damage detection unit and the loose wire detection unit are connected with the control system, the damage detection unit includes an ROI area segmentation module and a Blob analysis module, the ROI area segmentation module is connected to the Blob analysis module, and the loose wire detection unit includes a threshold segmentation module and a Blob analysis module. The edge number statistics module, the threshold segmentation module is connected to the edge number statistics module.

As preferably, the CCD camera adopts costarSI-M310 industrial black-and-white camera, and the CCD camera adopts the computar MLM-3XMP optical lens that can meet the requirements of shooting 8mm-10mm in the field of view; the image acquisition card adopts PCI-1409 image acquisition card.

The beneficial effects of the utility model are: the integration of inductance production, detection and packaging is realized, the production efficiency is greatly increased, the labor intensity of the workers is reduced, the labor force is reduced for the enterprise, the production cost is reduced, and the artificial fatigue and other subjective factors are reduced. Incoming false detection and missed detection can prevent the outflow of defective products; the real-time data records of the system also provide a basis for enterprise product quality management and production tracking, which is convenient for enterprise managers to know the production situation of the workshop, find problems in time, and solve them in time. Avoiding waste is of great significance to the development of enterprises.

The conception, specific structure and technical effects of the present utility model will be further described below in conjunction with the accompanying drawings, so as to fully understand the purpose, characteristics and effects of the present utility model.

Description of drawings

Fig. 1 is a system block diagram of the present utility model;

Fig. 2 is the overall flowchart of inductance production of the present utility model.

detailed description

With reference to Fig. 1-2, this specific embodiment adopts following technical scheme: A kind of inductance quality automatic detection system based on machine vision, comprises LED light source 1, CCD camera 2, image acquisition card 3, computer system 4, CRT display 5, image Processing software device 6 and control system 7, LED light source 1 cooperates with CCD camera 2, CCD camera 2 is connected with image acquisition card 3, image acquisition card 3 is connected with computer system 4 that image processing software device 6 is installed, computer system 4 is connected CRT display 5, computer system 4 is connected with the control system 7 of control inductance 8; Described computer system 4 is made up of image processing unit 401, edge positioning unit 402, damaged detection unit 403 and loose wire detection unit 404, and image processing unit 401 is connected The edge positioning unit 402 and the edge positioning unit 402 are respectively connected with the damage detection unit 403 and the loose wire detection unit 404 , and the damage detection unit 403 and the loose wire detection unit 404 are connected with the control system 7 .

It is worth noting that the damage detection unit 403 includes an ROI region segmentation module 405 and a Blob analysis module 406, and the ROI region segmentation module 405 is connected to the Blob analysis module 406; the loose line detection unit 404 includes a threshold value segmentation module 407 and edge number statistics Module 408 , the threshold segmentation module 407 is connected to the edge number statistics module 408 .

In addition, the LED light source 1 uses a blue square diffuse reflection light source to better highlight the defect characteristics to achieve the best effect; the CCD camera 2 uses a costarSI-M310 industrial black and white camera, which is mainly used to collect inductive images. The workpiece is focused on the CCD sensitive element through the optical lens, and different lenses can be selected according to the size of the workpiece and the working distance. This system uses a computar MLM-3XMP lens to meet the requirements of the field of view 8mm-10mm; the image acquisition card 3 uses PCI -1409 image acquisition card, used for synchronous control of automatic A/D conversion and the function of storing a large amount of data on the card, through computer software and synchronous pulse triggering, under the synchronous control of drive electrical appliances, A/D is performed on the video signal output by CCD Conversion, converted into image data, and then transmitted to the computer memory and video memory through the PCI bus in real time for the computer to perform various processing operations.

This specific embodiment uses machine vision technology for inductance detection. Machine vision is to use computers to simulate the visual function of the human eye, extract information from images or image sequences, and perform shape and motion recognition on three-dimensional scenes and objects in the objective world, replacing human eyes. Quantitatively detect whether the product quality is qualified and give the judgment result.

The working principle of this specific embodiment: the automatic detection system for inductance defects is to add a visual inspection system to the inductance production line to perform real-time detection and monitoring of inductance quality, so as to prevent defective products from flowing into the next process: the measured inductance 8 is placed on the conveyor belt according to a certain beat The CCD camera 2 that has been adjusted in position, combined with the image acquisition card 3 and the illumination of the LED light source 1, captures images in real time in accordance with the rhythm of the production line, and the images are projected to the optical imaging system. On the photosensitive array surface, the CCD camera 2 converts the optical image it receives into a video signal and outputs it to the image acquisition card 3, and the image acquisition card 3 converts the video signal into digital image information for the computer system 4 to process and the CRT monitor 5 to display. The system 4 uses various image processing algorithms to analyze the image of the data, judges whether it is qualified, and then sends the signal to the control system 7 to control the machine to perform the rejecting action.

The image processing software device 6 installed in the computer system 4 adopts a kind of general program development software Vision Builder for Automated Inspection developed by National Instruments (NI) as a development platform, because it itself includes multiple types of image processing, image transformation and Basic functional functions such as filtering, model matching and measurement, thus simplifying program development and greatly shortening the development cycle. VBAI is a configurable machine vision software that can be used to complete prototype creation, benchmark setting and application configuration, etc. In practical applications, VBAI can be upgraded to a powerful programming environment (such as LabVIEW) without programming. It has a built-in development interface, so it can quickly complete monitoring, guidance and identification applications; VBAI also includes the ability to establish complex pass/fail judgments , Digital I/0 device control and communication applications through serial devices.

The main purpose of the image processing unit 401 in the computer system 4 is to perform some preprocessing on the inductance image obtained by the image acquisition card. The image processing algorithm mainly involves median filtering. The basic idea of the median filter is to use pixel point neighborhood The median value of the gray value is used to replace the gray value of the pixel. This method can preserve the edge details of the image while removing impulse noise and salt and pepper noise. This system uses a 3×3 function window to calculate ] as the center of the function window pixel median value steps are as follows: ①Arrange the pixel points according to the brightness value; ②Select the median value of the sorted pixel set as the new value of point [i, j]. After processing by the median filter tool, the bright pixels at both ends of the inductor core are weakened, and the copper wires that do not need to be detected in the middle are filtered out to facilitate subsequent detection.

Edge positioning unit 402: Since there is a slight position difference in the image collected each time of the workpiece, the position of the inductor in the image must be located. This system uses the positioning tool Find Straight Edge based on edge search to locate the x and y directions. The position of the workpiece can be accurately found, which is convenient for the positioning and tracking of the subsequent processing procedures.

Loose wire detection unit 404: ① ROI region segmentation: Inductor damage is manifested by the exposed magnetic core at both ends of the inductance, which is prone to damage defects such as missing corners and pits. Because the image performance characteristics of surface damage are relatively bright, that is, the gray value is relatively high. Therefore, the ROI area, which is the damaged area on the surface of the magnetic core, should be within the surface area of both ends of the inductor magnetic core, and should not exceed this area, otherwise the bright pixels in the background will cause interference.

②Blob analysis: The basis for judging the surface damage of the inductor magnetic core is to detect the area of the largest surface damage area after threshold segmentation of the image. If the area exceeds a certain threshold, the inductor is considered to have a damage defect. In this system, the Detect Objects tool is used to detect defect spots, which can effectively detect damaged defects and mark them.

Damage detection unit 403: ①Automatic threshold segmentation based on histogram analysis: gray histogram is one of the simplest and most useful tools in digital image processing, which is a function of gray value and describes the gray level content of an image And its distribution, through the histogram analysis of the inductance image, it can be known that the distribution of the inductance magnetic core, copper wire and background gray value is relatively uniform, and the valley point can be selected as the threshold for segmentation, which can better separate the magnetic core and copper at both ends of the inductance. The lines are separated from the background.

②Statistics of the number of edges: Loose wire defects show that the end of the copper wire wound by the inductor is prone to looseness, and there is a gap with the non-loose winding area. Edge detection can be performed at the magnetic cores at both ends of the inductor. If it is a good product, only There is a pixel jump point, if there is more than 1, it means that there is a loose thread situation; because there is a relatively small loose thread situation, there is often no gap at one end and a gap at the other end, so the upper, middle and lower Use the Find Edges tool to detect the number of pixel jump points at the three positions.

This specific implementation method successfully solves the problem of automatic appearance inspection on the inductance production line by using the non-contact measurement of machine vision technology, realizes the integrated production process of production, inspection, and packaging, greatly improves production efficiency, and reduces labor costs. The detection method of inductor loose wire and damaged defect is stable and reliable in actual detection, and has high engineering application value and broad market application prospect.

The preferred specific embodiments of the present utility model have been described in detail above. It should be understood that those skilled in the art can make many modifications and changes according to the concept of the present utility model without creative efforts. Therefore, all technical solutions that can be obtained by those skilled in the art based on the concept of the utility model through logical analysis, reasoning or limited experiments on the basis of the prior art should be within the scope of protection defined by the claims .

Claims (3)

1. an inductance quality automatic checkout system based on machine vision, it is characterized in that: include LED light source (1), CCD camera (2), image pick-up card (3), computer system (4), CRT monitor (5), image processing software device (6) and control system (7), LED light source (1) matches with CCD camera (2), CCD camera (2) connects image pick-up card (3), image pick-up card (3) is connected with the computer system (4) being provided with image processing software device (6), computer system (4) connects CRT monitor (5), computer system (4) is connected with the control system (7) controlling inductance (8)；Described computer system (4) is made up of graphics processing unit (401), edge positioning unit (402), damage testing unit (403) and loose ends detector unit (404), graphics processing unit (401) edge fit edge positioning unit (402), edge positioning unit (402) is connected with damage testing unit (403), loose ends detector unit (404) respectively, and damage testing unit (403), loose ends detector unit (404) are connected with control system (7).

A kind of inductance quality automatic checkout system based on machine vision, it is characterised in that: described CCD camera (2) uses and can meet the computar MLM-3XMP optical lens that visual field shooting 8mm-10mm requires.

A kind of inductance quality automatic checkout system based on machine vision, it is characterised in that: described image pick-up card (3) uses PCI-1409 image pick-up card.