Secondary Detection System for Automated Optical Inspection Accuracy

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional automated optical inspection systems for electronic component pin solder joints have low performance and require extensive manual adjustments to improve first pass yield, lacking versatility due to high false abnormal detection rates.

Innovation Solution

A secondary detection system integrating automated optical inspection with a neural network, utilizing a receiving module, image capture, image query, pin solder joint image extraction, feature extraction via convolutional neural networks, and probability calculation using a siamese neural network to verify detection results, adjusting abnormal conditions to normal when probability values meet a threshold.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If conventional automated optical inspection is used to detect pin solder joints, then inspection coverage is achieved, but false abnormal detection rate increases and first pass yield decreases

Engineering Contradiction:

Improvedetection accuracyVSAvoidfalse abnormal detection

Core Design Contradiction:

ReliabilityVSObject-generated harmful factors

Solution Approach 1:

The patent introduces a secondary detection device as an intermediary between the automated optical inspection apparatus and the final judgment. This secondary device captures additional images from different angles, extracts features using convolutional neural networks, and calculates probability values to verify abnormal detections, thereby reducing false positives without compromising detection coverage

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The system implements a feedback mechanism where the secondary detection device verifies abnormal conditions detected by the automated optical inspection apparatus. Based on the calculated probability values from multiple image analyses, the system provides feedback to confirm or reject the abnormal detection, improving overall detection accuracy and reducing false alarms

Inventive Principle:

Principle #23Feedback

2Reliability

If engineers manually adjust qualification parameters to improve first pass yield, then detection accuracy may improve, but time and energy consumption increase significantly

Engineering Contradiction:

Improvedetection accuracyVSAvoidparameter adjustment time

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent implements a self-service mechanism where the secondary detection device automatically performs feature extraction, probability calculation, and verification of abnormal conditions without requiring engineer intervention. The system autonomously processes images through convolutional neural networks and makes judgment calls based on calculated probability values, eliminating manual parameter adjustment while maintaining high detection accuracy

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent replaces the manual mechanical process of parameter adjustment with an automated computational system. Instead of engineers manually tweaking qualification parameters, the system uses neural networks to automatically extract features and calculate probability values, substituting human mechanical adjustment with automated intelligent processing

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Productivity

If conventional automated optical inspection is used, then inspection speed is maintained, but versatility across different circuit boards is poor

Engineering Contradiction:

Improveinspection speedVSAvoidapplicability to different circuit boards

Core Design Contradiction:

ProductivityVSAdaptability or versatility

Solution Approach 1:

The patent enhances versatility by making the secondary detection device applicable to multiple different circuit board types and electronic components. The convolutional neural network-based feature extraction and probability calculation mechanisms can adapt to various solder joint configurations, allowing the same system to verify abnormalities across different board designs while maintaining inspection speed

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS11704788B2Secondary detection system for integrating automated optical inspection and neural network and method thereof

Publication Date: 2023.07.18 INVENTEC PUDONG TECH CORPOARTION

AI summary

A secondary detection system for integrating automated optical inspection and neural network and a method thereof are disclosed. In the secondary detection system, an automated optical inspection apparatus performs automated optical inspection for pin solder joints on circuit board, and when a detection result indicates abnormal condition, the secondary detection device calculates a detection image probability value based on the component image feature and the template image feature, and calculate pin solder joint image probability values based on the component pin solder joint image feature and the template pin solder joint image feature through siamese neural network, to obtain a minimum probability value among the detection image probability value and pin solder joint image probability values. The minimum probability value is used to determine whether to change the detection result, thereby providing accurate detection result of automated optical inspection and increasing a first pass yield.