Method and system for needle tip alignment degree detection based on deep learning

Abstract

The invention relates to the technical field of image processing, and discloses a deep learning-based needle tip alignment degree detection method and system, comprising: acquiring needle tip top view data of multiple connector samples, labeling the data and dividing a data set, and counting needle tip prior information ; Determine the standard needle tip feature of the needle tip, construct a standard first relative coordinate system, and determine the relative standard distance between the center position of the needle tip and the origin; obtain the needle tip top view data of the connector to be tested, and obtain the needle tip ROI area of ​​the connector to be tested; Extract the sub-pixel contour edge information of the needle tip and the needle tip feature, and construct a second relative coordinate system in the same way as the above steps to determine the relative distance between the center position of the needle tip and the origin; compare the extracted needle tip feature with the standard needle tip feature and relative distance with The corresponding relative standard distances are compared, and according to the comparison results, it is judged whether the alignment degree of the connector to be tested is qualified. The invention improves the accuracy of the detection of the alignment degree, and is suitable for connectors.

Description

technical field [0001] The present invention relates to the technical field of image processing, and in particular to a method and system for detecting needle tip alignment based on deep learning. Background technique [0002] In the strategic layout of the Industrial Internet, industrial defects are a very popular problem, especially the problem of workpiece accuracy detection. For high-precision instruments and products, millimeter-level accuracy is required. [0003] In the workpiece accuracy detection, the alignment detection is mainly to detect the verticality, horizontal straightness, coplanarity, gap, needle width, needle length and other indicators of the needles of various IC chips, electronic connectors and other electronic components. . However, the traditional alignment detection is based on the idea of ​​pattern matching to match all needle tips, so as to judge the distance measurement between the needle tip and the needle tip. For the connector, there are miss...

AI Technical Summary

Problems solved by technology

[0004] The present invention aims to solve the problem of poor accuracy in the existing detection of needle tip alignment, and proposes a method and system for detecting needle tip alignment based on deep learning

Images