Connector detection method and device based on machine vision

Abstract

The invention discloses a connector detection method and device based on machine vision, and the method comprises the steps: firstly obtaining an image sequence and the depth information of the image sequence, judging whether a to-be-detected connector exists or not according to the depth information, superposing a plurality of adjacent frames of backgrounds of the to-be-detected connector in time through employing an average background method, and averaging the superposed backgrounds as a background image, the method comprises the following steps: subtracting a background image from a to-be-detected image to obtain a difference image, then combining a watershed segmentation algorithm to obtain areas of a terminal and an end seat, calculating color characteristic values, transforming the difference image to a frequency domain by using discrete cosine, combining information embodied by high frequency and low frequency of the image in the frequency domain, extracting frequency domain characteristics of the image, and calculating the color characteristic values of the terminal and the end seat. And analyzing and judging whether the to-be-detected connector has the problems that the colors of the terminal and the end seat are not matched and the connector has copper wire exposure or not by integrating the characteristic values of the spatial domain and the frequency domain. According to the method, the detection efficiency can be improved, the false detection rate and the intelligent degree are reduced, manpower and time are greatly saved, and the method has very high practicability and economic value.

Description

technical field [0001] The invention relates to the field of visual detection, in particular to a method and device for detecting connectors based on machine vision. Background technique [0002] With the rapid development of the electrical and electronic industry, more and more connectors are used, and the types are becoming wider and wider. In order to prevent the defective products of these electronic connectors from flowing into the process, each connector needs to meet the process quality inspection requirements. Relying solely on manual quality inspection (slow speed and high labor costs) or random inspection of laboratory standard testing equipment is not enough to meet the quality control of the entire process of product quality in modern manufacturing and the high-speed requirements of online production testing. [0003] The existing machine vision-based connector defect detection method is greatly affected by changes in illumination and is sensitive to changes in e...

AI Technical Summary

Problems solved by technology

If relying solely on manual quality inspection (slow speed, high labor cost) or random inspection of laboratory standard testing equipment, it is not enough to meet the quality control of the entire process of product quality in modern manufacturing and the high-speed requirements of online production testing

[0003] Existing machine vision-based connector defect detection methods are greatly affected by changes in illumination and are sensitive to changes in environmental factors. The detection method has a high false detection rate, slow detection speed, and low reliability.

Images