An automatic teaching method for glue strip detection

Abstract

The invention discloses an adhesive tape detection demonstration method which comprises the following steps: S1, acquiring an adhesive tape image by a camera, acquiring an image coordinate with a preset initial point, and setting a detection area by taking the coordinate as a starting point; S2, calculating m pairs of edge points and right edge points in the detection area; and calculating quality, and screening m pairs of edge points to obtain adhesive tape edge points: p left and p right; S3, updating the detection area in the gluing direction, and updating a reference value according to thecalculation result in the step S2; and S4, repeating the steps S2 and S3, and continuously storing the coordinate values of the p left point and the p right point, and stopping iteration until the total number of repetition times reaches a preset number of times, and taking a plurality of pairs of stored p left and p right coordinate values as final adhesive tape edges, thus completing demonstration of the adhesive tape contour. Through the adhesive tape detection demonstration method, miscellaneous points are effectively removed, edge points are screened, and reference information of each iteration is taken as a result of the previous iteration, and continuous iteration is performed, thus accurately extracting width information and position information of the adhesive tape.

Description

technical field [0001] The invention relates to the technical field of teaching, in particular to an automatic teaching method for rubber strip detection. Background technique [0002] With the development of industrial automation, the gluing process is widely used in various fields of industrial production, and the detection of gluing quality has gradually realized automation and intelligence. Using the principle of machine vision, the image processing method of the strip image is adopted. It is an effective gluing detection method. The existing robot gluing detection method has many steps and complicated calculations. At present, the gluing detection process is mainly divided into two steps: the teaching process and the measurement process. The teaching process is to use the relevant prior information to measure the position and width of the rubber strip, and use it as the teaching information in the measurement process; the measurement process is to use the teaching infor...

AI Technical Summary

Problems solved by technology

[0002] With the development of industrial automation, the gluing process is widely used in various fields of industrial production, and the inspection of gluing quality has gradually realized automation and intelligence. Using the principle of machine vision, the image processing method of the strip image is adopted. It is an effective gluing detection method. The existing robot gluing detection method has many steps and complex calculations. At present, the gluing detection process is mainly divided into two steps: the teaching process and the measurement process. The teaching process is to use the relevant prior information to measure the position and width of the rubber strip, and use it as the teaching information in the measurement process; the measurement process is to use the teaching information to detect the position and width information of the rubber strip; currently, the teaching The main method of the process is: click the skeleton point at the center of the rubber strip. The skeleton point of the rubber strip is the point on the center line of the rubber strip. This method performs curve fitting according to the skeleton point to obtain a smooth curve, and then according to the curve Points on the surface, generate a model for detecting the edge of the rubber strip, and then use the information such as the width, position, and gray level of the rubber strip to detect the edge of the rubber strip. This method has high requirements for the accuracy of manual clicks on the skeleton points. The above is that it is necessary to point at the center of the rubber strip to achieve the width detection accuracy. When the position of the skeleton point deviates from the center of the rubber strip, it will affect the angle of the edge detection model, resulting in an error between the detected width of the rubber strip and the actual width of the rubber strip. It is also possible No glue breakage of the glue strip can be detected

Images