Semiconductor Crack Detection via Image Gradient and Binarization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting cracks in semiconductor chips during manufacturing are inefficient and prone to errors, often requiring time-consuming conductivity measurements or complex image processing techniques that are difficult to automate and require numerous input values.
Innovation Solution
A detecting method and apparatus that utilizes image processing techniques, including image gradient analysis, binarization, filling closed regions, noise removal, extraction of long, narrow shapes, and removal of shapes connected to the edge portion, to accurately identify cracks and foreign objects on semiconductor chips, reducing the detection process to nine steps and minimizing user input.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional conductivity measurement methods are used to detect cracks, then detection accuracy may be maintained, but detection time increases significantly and automation becomes difficult
Solution Approach 1:
The patent replaces conventional conductivity measurement methods with optical image processing techniques. The image processor captures optical images of the semiconductor chip and uses image gradient analysis, binarization, and shape extraction algorithms to detect cracks, eliminating the need for time-consuming electrical conductivity measurements while maintaining detection accuracy and enabling automation
Solution Approach 2:
The patent creates a digital copy of the physical chip surface through image capture and processing. By analyzing the optical image copy through gradient calculation, binarization, and shape extraction, the system identifies crack patterns without physically contacting or electrically testing the chip, thus reducing detection time while preserving accuracy
2Measurement precision
If complex image processing techniques with numerous input values are used, then detection accuracy improves, but system complexity and ease of operation deteriorate
Solution Approach 1:
The patent segments the image processing task into distinct sequential steps: gradient calculation, binarization, closed region filling, noise removal, and shape extraction. Each step processes the image independently with clear input-output relationships, making the overall complex system easier to operate and control while maintaining high detection accuracy
Solution Approach 2:
The patent performs preliminary image processing actions in a fixed sequence before final crack detection. By pre-calculating gradients, pre-binarizing the image, and pre-filling closed regions, the system reduces the complexity of the main detection algorithm and simplifies user operation while preserving detection precision
3Reliability
If traditional crack detection methods are used, then comprehensive detection may be achieved, but the detection process becomes difficult to automate
Solution Approach 1:
The patent replaces manual or semi-automated detection methods with a fully automated image processing system. The computer-executable instructions automatically perform gradient calculation, binarization, region filling, noise removal, and shape extraction, achieving comprehensive crack detection without human intervention while maintaining reliability
Solution Approach 2:
The patent enables the detection system to automatically process and analyze images without external control. The image processor autonomously executes the detection algorithm, identifies crack shapes, and outputs results, making the system fully automatable while ensuring comprehensive and reliable detection
Data Source
AI summary
A detecting apparatus includes an image processor configured to detect a crack in a semiconductor chip through image processing of an input image of the semiconductor chip, and a result outputting unit configured to output results detected by the image processor. The image processor is configured to perform an image gradient analysis on the input image, to convert the input image into a binary image by image binarization, based on a result obtained by performing the image gradient, to fill a closed region in the binary image with the same color as a color representing an outline of a shape appearing in the binary image that forms the closed region, to extract a shape having a predetermined property from the binary image, and to detect the crack by removing a shape connected to an edge portion of the input image. The result outputting unit is configured to output a resulting image in which a crack portion is expressed in the input image.