CN100562761C - Intelligent LED lighting quality online inspection system - Google Patents

Abstract

Intelligent light emitting diode (LED) vehicle light illumination quality on-line detection system, comprise industrial computer, display, operation keyboard, camera bellows top opening part connects a car light mold, and the car light face down is fixedly placed in the car light mold, and the camera bellows bottom is equipped with image sensor; Each image sensor links to each other with image collection card respectively, and image collection card links to each other with industrial computer.During detection, be divided into several zonules, calculate each zonule gray average and regional location, set the gray scale tolerance value, be made into standard form earlier to the LED car light of the standard image that pans, and with panoramic image; Detected LED car light is calculated each zonule gray average and regional location according to same procedure, be made into the comparison template; Check the gray average on the corresponding region one by one and compare, if margin tolerance with interior be qualified, otherwise be defective.

Description

Intelligent LED lighting quality online inspection system

technical field

The invention relates to the quality detection of an automobile lighting system, in particular to a detection system and a detection method for on-line detection of the quality of a light-emitting diode (LED) vehicle light using computer vision detection technology.

Background technique

The new and intelligent LED car lights are one of the most dazzling products in the world of automotive electronics. In the automotive lighting system, the headlight assembly is a combined lighting component and a key component for driving safety. There are corresponding standard specifications for the manufacture and installation of lighting components, and they are strictly adjusted before leaving the factory to ensure that the lights meet the standards. The quality inspection of the car light assembly includes the inspection of lighting performance, light source power and parts assembly quality. But so far, domestic automobile manufacturers have only carried out single-item inspections of lighting performance for the quality inspection of LED rear light assemblies, and they have used manual visual inspection methods, which directly affects the inspection quality and production efficiency of products. Can not adapt to the needs of the rapid development of the automobile industry.

Contents of the invention

In order to overcome the above disadvantages, the present invention designs an intelligent LED lighting quality online detection system, the basic idea is: introduce advanced computer vision detection technology into the quality detection of LED rear light assembly, by setting system parameters, it can Realize online dynamic detection of lighting quality for different types of lights, with a high degree of flexibility and intelligence. The specific plan is: according to the principle of visual inspection technology, design an effective image acquisition system, on the basis of obtaining the LED rear light lighting image, research and use digital image processing technology, analyze the lighting quality of the car light in real time, and judge the lighting of the LED car light failure and its area.

The intelligent online light-emitting diode (LED) lighting quality inspection system includes an industrial computer, a display, and an operating keyboard. In the car lamp mold, an image sensor is installed at the bottom of the dark box; each image sensor is connected to an image acquisition card, and the image acquisition card is connected to an industrial computer.

Detection method: computer control is used throughout the whole process, and an effective image acquisition system is designed based on the principle of visual detection technology. Divide it into several small areas, calculate the average gray value and area position of each small area, set the gray tolerance value, and make a standard template; light up the LED lights to be detected to take a panoramic image, and use the same principle to Divide the panoramic image into several small areas, calculate the average gray value and area position of each small area, and make a comparison template; check the average gray value on the corresponding area one by one and make a comparison, if it is within the tolerance range, it is qualified, otherwise is unqualified.

The invention adopts advanced virtual instrument technology and stable digital image processing algorithm, so that the system can complete the online automatic detection of various types of vehicle lights through computer screen setting or software adjustment. It can accurately judge the failure of LED car lighting and its area, with fast detection speed and high reliability. It provides a modern quality inspection method for the rapid development of the automobile industry.

Description of drawings

The present invention will be described in detail below in conjunction with the accompanying drawings and embodiments.

Accompanying drawing 1 is the schematic diagram of system principle of the present invention;

Accompanying drawing 2 is the flow chart of making standard formwork of the present invention;

Accompanying drawing 3 is the flow chart of making comparison template of the present invention.

1. Car light, 2. Car light mold, 3. Image sensor, 4. Camera obscura, 5. Image acquisition card, 6. Display, 7. Industrial computer.

Detailed ways

The industrial computer uses a high-performance industrial microcomputer, so that the detection system can work normally in the production workshop environment, and at the same time can achieve high-speed processing of image information to meet the requirements of detection speed; the image sensor is selected for industrial use with high sensitivity and high resolution. The digital camera can realize the detection of the lighting quality of different specifications of lights, and is equipped with a high-performance, high-reliability image acquisition card to complete the high-speed acquisition and high-speed preprocessing of image information. The quality of the image has a great relationship with the lighting. In order to ensure the reliability of the detection, the present invention uses an interference suppression power supply to ensure the high quality and stability of the lighting system.

The opening on the top of the camera obscura 4 is connected with a car lamp mold 2, and the car lamp 1 is fixedly placed in the car lamp tire with the front facing down. Two digital cameras 3 are installed at the bottom of the camera obscura, and each digital camera is connected to an image acquisition card 5 respectively. , the image acquisition card is connected with the industrial computer 7 .

Correctly adjust the camera so that the brightness of each area of the LED lights in the captured image is basically uniform. The model and parameters of the detected LED lights are input by the operating keyboard, and the computer saves the information.

The first step is to make a standard template: first light up the standard LED car lights, two digital cameras take panoramic images, pass them to the image acquisition card, and the computer saves the information; then intercept a lighting area of the LED car lights, and the computer according to the settings Parameters, to judge whether the area belongs to the image information of the bright area or the dark area is not obvious, and can carry out necessary enhancement processing; calculate the gray value and area position of the area through software programming calculation; then set the area Tolerance value of the gray level; Repeat the above steps to obtain the position, mean value of gray level, tolerance value and other information of all lighting areas and save them in the computer.

The second step is to make a comparison template: first light up the LED lights to be detected, two digital cameras take a panoramic image, pass it to the image acquisition card, and save the information, according to the position of the lighting area selected by the standard template, the intercept ratio For the image at the corresponding position of the template, judge whether to strengthen the image according to the parameters set by the standard template. If it is within the tolerance range, it is qualified; otherwise, it is unqualified. The computer screen 6 will display all test results.

Repeat the above steps to obtain and save information such as the position of all lighting areas, the average gray value, and whether they are qualified or not. For LED lights of different specifications and models, you only need to input the model of the lights, set new parameters, and repeat the above method to complete the inspection of the lighting quality of all lights.

Claims (2)

1. An intelligent light-emitting diode lighting quality online detection system, including industrial computer, display, and operating keyboard, characterized in that: there is a dark box, the opening above the dark box is connected with a car light mold, and the front of the car light is fixedly placed In the car lamp mold, image sensors are installed at the bottom of the dark box; each image sensor is connected to an image acquisition card, and the image acquisition card is connected to an industrial computer. 2 According to claim 1, the intelligent on-line inspection system for the lighting quality of light-emitting diode vehicle lights is controlled by a computer throughout the process, and an effective image acquisition system is designed based on the principle of visual inspection technology. And take a panoramic image, divide the panoramic image into several small areas, calculate the average gray value and area position of each small area, set the gray tolerance value, and make a standard template; light up the detected LED lights, shoot For the panoramic image, according to the position of the lighting area selected by the standard template, the image corresponding to the position of the comparison template is intercepted, and the parameters set by the standard template are used to determine whether to enhance the image. The average gray value is compared with the average value of the standard template. If it is within the tolerance range, it is qualified, otherwise it is unqualified; repeat the above steps to obtain information such as the position of all lighting areas, the average gray value, and whether it is qualified or not and save it.

Images