Wafer Backside Brightfield Image Anomaly Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting wafer backside brightfield image anomalies are inefficient and prone to missing hidden anomalies, as they rely on manual screening by engineers.
Innovation Solution
A method involving histogram equalization of wafer backside brightfield images, followed by statistical analysis of gray histograms and calculation of abnormal pixel scores, to automatically identify and highlight anomalous images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual screening by engineers is used to detect wafer backside brightfield image anomalies, then the method is simple to implement, but the detection efficiency is low and hidden anomalies are prone to be missed
Solution Approach 1:
The patent replaces the manual mechanical screening process with an automated image processing system. The system uses histogram equalization to enhance image contrast, followed by gray histogram analysis to calculate abnormal pixel scores, thereby substituting human visual inspection with computational algorithms that can efficiently detect anomalies including hidden ones that are invisible to the naked eye.
Solution Approach 2:
The detection system performs self-service by automatically processing wafer backside brightfield images through histogram equalization and gray histogram analysis. The system independently calculates abnormal pixel scores and identifies anomalies without requiring human intervention, thereby improving detection efficiency while maintaining simplicity through automated workflows.
2Measurement precision
If manual screening by engineers is used to detect wafer backside brightfield image anomalies, then the method requires minimal equipment, but the detection precision is low and hidden anomalies are missed
Solution Approach 1:
The patent changes key parameters of the image data through histogram equalization to enhance contrast and make anomalies more detectable. By transforming the pixel intensity distribution and analyzing the gray histogram, the system achieves higher detection precision for hidden anomalies that are invisible in original images, while the parameter transformation process remains computationally manageable.
Solution Approach 2:
The patent introduces gray histogram analysis as an intermediary between the raw image data and the final anomaly detection. This intermediary step calculates the distribution of pixel intensities and computes abnormal pixel scores, serving as a mediator that enhances the detection capability without requiring complex direct image analysis, thereby improving precision while controlling system complexity.
Data Source
AI summary
The present application provides a method for automatically detecting a wafer backside brightfield image anomaly, at least comprising: processing wafer backside brightfield images by means of histogram equalization, so as to obtain processed images; compiling statistics for a gray histogram of the processed images; calculating the number of abnormal pixels in each of the images; and providing a threshold, and highlighting the image with a score less than the threshold. In the present application, the wafer backside brightfield images are analyzed by means of image preprocessing and a specific calculation method, so as to quickly and automatically detect an abnormal wafer backside image.