Wafer Probe Mark Inspection Using Neural Network Automation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In integrated circuit manufacturing, manual probe mark inspections are prone to human errors, are time-consuming, and affect the quality and yield of wafers, leading to increased production costs and delayed time-to-market.
Innovation Solution
An automated test system incorporating a neural network subsystem and process control processor that analyzes probe marks on wafers using image recognition, reducing manual effort and enhancing productivity by automatically generating inspection results and updating records.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual probe mark inspection is performed by operators, then inspection can be conducted, but human errors are introduced and test quality is affected
Solution Approach 1:
The patent replaces the manual mechanical inspection process with an automated image recognition system using neural networks. The system captures images of probe marks and uses machine learning algorithms to automatically analyze and evaluate them, eliminating human operators from the direct inspection process and thereby removing human errors while maintaining high accuracy.
Solution Approach 2:
The system creates digital copies (images) of the probe marks on wafers and performs inspection on these copies rather than direct physical inspection. This allows for automated analysis without physical contact and enables multiple analyses of the same mark without degradation, improving both reliability and precision.
2Reliability
If manual probe mark inspection is performed, then inspection can be conducted, but the process is highly time consuming
Solution Approach 1:
The manual inspection process is replaced with an automated system that captures images and processes them through neural networks. This automation eliminates the time-consuming manual examination of each probe mark, significantly reducing test cycle time while maintaining or improving inspection quality through consistent algorithmic evaluation.
Solution Approach 2:
The automated system enables continuous inspection without the interruptions inherent in manual processes. The neural network can process multiple images in sequence without pause, and the system can operate continuously to inspect all wafers in a lot, thereby improving productivity while ensuring consistent quality assessment.
3Reliability
If manual probe mark inspection is performed, then inspection can be conducted, but operators and engineers are required which increases production cost
Solution Approach 1:
The patent replaces human operators and engineers with an automated image recognition system. While the initial investment in automation equipment is required, the system eliminates ongoing labor costs and reduces the need for highly skilled personnel, thereby reducing production costs in the long term while maintaining high inspection quality through consistent automated evaluation.
Solution Approach 2:
The system performs self-learning and self-improvement through the neural network's ability to learn from training data. The system automatically adjusts its inspection criteria and improves its accuracy over time without requiring continuous human intervention or expertise, reducing dependency on skilled personnel and associated costs.
4Extent of automation
If automatic analysis of probe marks is implemented, then customer service quality is improved, but system complexity increases
Solution Approach 1:
The automated system performs multiple functions: image capture, image processing, neural network analysis, result generation, and continuous learning. By consolidating these functions into a single integrated system, the patent achieves high automation while managing complexity through functional integration rather than separate components for each task.
Solution Approach 2:
The neural network acts as an intermediary between the raw image data and the final inspection results. It processes the complex image information and transforms it into meaningful quality assessments, thereby managing the complexity of automatic analysis while delivering high-quality automated inspection results.
Data Source
AI summary
A test system is provided, including an assessment subsystem, a neural network subsystem and a process control processor. The assessment subsystem receives a test image of a tested wafer from a probe apparatus. The process control processor controls, in response to the probe apparatus obtaining the test image, the assessment subsystem to perform an assessment operation to transmit the test image to the neural network subsystem in an automation mode. The neural network subsystem identifies an image specification of probe marks in the test image and generates an analyzed data of the test image to the assessment subsystem. The assessment subsystem further generates a first probe mark inspection result based on the analyzed data to the process control processor for generating a test result.