Automatic teaching method of glued sensor

Abstract

The invention discloses an automatic teaching method for a glue coating sensor. At the position where the glue strip is to be detected, a light source projects red, green and blue light rays to the glue strip, and the corresponding grayscale image is synthesized by using the images in the three colors. And get the glue strip area; according to the integrity of the glue strip area and the position in the image, determine the camera with the best shooting angle, record it as the acquisition camera; store the acquisition camera at each detection position as the camera teaching information; Use the average grayscale of the glue strip area to judge the color of the light source; store the color information of the light source at each detection position as the teaching information of the light source; this method can also perform automatic parameter teaching, only the operator needs to set the initial parameters of the camera and the light source. After setting, the computer can be used to complete the subsequent teaching process, so that the image can achieve better contrast and improve the teaching efficiency.

Description

technical field [0001] The invention relates to the field of gluing visual inspection, in particular to an automatic teaching method for gluing sensors. Background technique [0002] With the development of industrial automation, the automatic gluing process has been widely used in various fields of industrial production. However, due to the differences in the production environment, the difference in the material of the sprayed glue, and the condensation of the glue at the nozzle of the glue gun, the glue application process will not appear. Uneven glue or glue discharge will eventually cause the glue strip on the parts to be missed, broken, too wide or too narrow, which cannot meet the production quality requirements. Therefore, it is necessary to test the glue quality; the existing glue test The sensor is usually placed on the glue gun, and multiple cameras are set in the range of 360°. For example, three cameras are used, and the interval between them is 120° to realize ...

AI Technical Summary

Problems solved by technology

[0002] With the development of industrial automation, automatic gluing process is widely used in various fields of industrial production, but due to the differences in production environment, the material difference of sprayed glue and the condensation of glue at the glue nozzle of the glue gun, the gluing process will appear out of order. The glue or glue is not uniform, which will eventually lead to missing, broken, too wide or too narrow glue strips on the parts, which cannot meet the production quality requirements. Therefore, it is necessary to test the glue quality; the existing glue inspection The sensor is usually set on the glue gun, and multiple cameras are set in the range of 360°. For example, three cameras are used, and the interval between two cameras is 120°, so as to realize the all-round detection of the glue strip. During the glue coating process, the robot drives The glue gun and the sensor move along the gluing trajectory. Due to the different camera positions and the occlusion of the glue gun, the images of the glue strips collected by multiple cameras are different. Especially when the sensor rotates at an angle, some cameras may not be able to capture the glue. Therefore, it is necessary to carry out the teaching process of the glue-coating sensor first, obtain the correct acquisition camera, and match the appropriate lighting light for different glue strips; the teaching process only needs to be carried out once, and the measurement process can be obtained The guidance information in the measurement process is used to detect the rubber strip in real time according to the teaching information. It can be seen that the accuracy of the teaching information is very critical.

[0003] The current teaching process requires operators to put in a lot of work, analyze a large number of collected images one by one, judge the rubber strip information in each image, select the camera at the best shooting position, and match the color of the light source to be used , which greatly affects the work efficiency. A calibration process takes several hours. At the same time, adjusting the brightness of the light source, camera exposure, gain and other parameters also requires repeated configuration and testing; this manual teaching process increases the operation on the one hand The workload of the personnel, on the other hand, the teaching information is affected by the operator's own subjective factors, and the best imaging effect cannot be guaranteed

Images