Method and device for surface defect detection of injection molded parts based on machine vision

Abstract

The invention discloses a method and device for detecting surface defects of injection molded parts based on machine vision. The method uses a three-dimensional scanner to scan the surface of the template workpiece and the workpiece to be tested respectively, and transforms the scanned three-dimensional point cloud data into several A two-dimensional depth image is processed by machine vision to judge defects and make statistics. The data obtained by the method of the invention has the advantages of high precision and intuitive display of defects; after obtaining the data, the specific shape and size of the defects can be obtained, which is convenient for summarizing the characteristics of the defects, improving the injection molding process and repairing the injection mold.

Description

technical field [0001] The invention relates to the field of real-time detection equipment for production line products, in particular to a surface defect detection device for injection molded parts based on machine vision, in particular to a surface defect detection device for injection molded parts based on three-dimensional scanning and depth map matching. Background technique [0002] Injection parts are mainly workpieces made of polyethylene or polypropylene mixed with various organic solvents. After being melted at high temperature, the plastic pellets are injected into the template, and then formed after being extruded and cooled. The molding process is affected by factors such as temperature, pressure and template, and the surface of the output injection molded parts may be lack of glue or excessive glue. Therefore, the injection molded parts usually need to be checked manually. [0003] Manual detection will cause visual fatigue of workers, subjective misjudgment,...

AI Technical Summary

Problems solved by technology

[0003] Manual detection will have disadvantages such as visual fatigue of workers, subjective misjudgment, time-consuming and high cost

However, it is difficult to obtain surface height information by using two-dimensional image detection of surface defects, and the accuracy is too low; the point cloud data obtained by purely using three-dimensional scanning is large, the processing time is long and the hardware requirements are high, it is difficult to realize real-time detection, and the high cost is not conducive to Marketing

[0004] At present, there are few defect detection technologies that use the combination of 3D and 2D images. For example, the article "Research on Injection Molded Disc and Battery Appearance Inspection Based on Machine Vision" simply matches the overall grayscale image without layering. The accuracy is low, and the defects detected at the same time cannot be quantified and displayed

In summary, there are few surface testing equipment for injection molded parts with high detection accuracy and fast speed on the market

Images