Connector Image Inspection Using Dynamic Datum and Tolerance Lines
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image analysis methods for connector inspection are inadequate due to the inability to detect small tolerances visually, inefficiencies in manual and computer vision systems, and failure to adjust to connector depth after housing assembly, leading to potential pin bending and reliability issues.
Innovation Solution
A computer-implemented method using dynamic datum identification and deep learning algorithms to analyze connector images, including reference feature detection, pixel size calculation, tolerance line annotation, and convolutional neural network training for defect detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual inspection methods are used, then human visual analysis can be applied, but small tolerances cannot be detected and inspection efficiency is low
Solution Approach 1:
The patent replaces manual visual inspection with an automated computer vision system that uses image processing algorithms to detect connector features. The system captures images of connectors at different depths and uses automated analysis to measure pin positions and detect defects, eliminating the need for manual inspection while achieving both high precision tolerance detection and improved inspection efficiency.
Solution Approach 2:
The patent introduces an intermediate image processing system that acts as a mediator between the connector object and the inspection result. The system uses captured images as intermediaries to analyze connector features, allowing indirect observation and measurement of small tolerances through digital image analysis rather than direct manual measurement.
2Reliability
If golden image comparison is used, then existing connector images can be compared, but row shifts are not detected and false positives occur
Solution Approach 1:
The patent performs preliminary actions by capturing multiple images of the connector at different depths before analysis. By taking images at various focal planes, the system establishes a comprehensive set of reference data that accounts for potential row shifts and positioning variations, enabling more accurate defect detection without false positives from positional mismatches.
Solution Approach 2:
The patent applies dynamics by using adjustable focus and multiple image capture planes rather than a static single-image comparison. The system dynamically adjusts the focal plane to capture images at different depths, allowing it to accommodate row shifts and positioning variations while maintaining accurate defect detection through multi-plane image analysis.
3Measurement precision
If coordinate measuring machines are used, then precise measurements can be obtained, but inspection speed is very slow and not suitable for production
Solution Approach 1:
The patent replaces the mechanical coordinate measuring machine with an optical computer vision system. Instead of using physical probes to measure connector features, the system uses captured images and image processing algorithms to automatically measure pin positions and dimensions, achieving both high measurement precision and fast inspection speed suitable for production environments.
Solution Approach 2:
The patent creates digital copies of the connector through image capture rather than physical contact measurement. By creating optical copies (images) of the connector at different depths and analyzing these copies computationally, the system achieves precise measurements without the time-consuming mechanical probing process, significantly improving inspection speed while maintaining accuracy.
4Adaptability or versatility
If computer vision systems with fixed focus are used, then shallow connector features can be inspected, but depth adjustment is insufficient and pin bending cannot be detected after housing assembly
Solution Approach 1:
The patent applies dynamics by implementing a focus adjustment mechanism that allows the computer vision system to change its focal plane dynamically. The system can adjust focus to capture images at different depths within the connector housing, enabling inspection of features at various positions including those that may be bent or misaligned after assembly, thereby improving both adaptability and detection reliability.
Solution Approach 2:
The patent adds the depth dimension to the inspection process by capturing images at multiple focal planes rather than a single fixed plane. This multi-dimensional approach allows the system to inspect features throughout the depth of the connector housing, detecting pin bending and other defects that occur at different depths after housing assembly, significantly improving versatility and reliability.
Data Source
AI summary
A computer-implemented method includes receiving an image of an article of interest to be evaluated relative to one or more features of interest, identifying a reference feature of a known size in the received image, identifying two or more extremities of the reference feature in the received image and a number of pixels between the two or more extremities of the reference feature, calculating a pixel size for the selected image based on the reference feature size and the number of pixels between the two or more extremities of the reference feature, annotating the received image to include one or more tolerance lines for the one or more features of interest, and determining whether the one or more features of interest in the image comply with the one or more tolerance lines. A computer program product and computer system corresponding to the method are also disclosed.