Salient Feature Point Image Alignment for Semiconductor Wafer Inspection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image alignment methods for semiconductor manufacturing, such as those used in wafer level packaging, face limitations including inaccurate rotation estimation, inability to handle smaller image rotations, and high matching failure rates, which restrict the number of wafers that can be inspected per hour and reduce sensitivity to defects.
Innovation Solution
A system and method for image alignment that selects and aligns three or more salient feature points between test and reference images using a computer subsystem, applying a salient feature point detection method to generate and align images, and subsequently aligning the remaining image portions based on the aligned feature points.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If Fourier transform based alignment method is used, then computation speed is improved, but rotation estimation accuracy deteriorates
Solution Approach 1:
The patent segments the image alignment problem into two distinct phases: (1) coarse alignment using Fourier transform to handle large translations and estimate rotation, and (2) fine alignment using salient feature points to achieve precise sub-pixel accuracy. This segmentation allows each method to operate in its optimal performance range, resolving the contradiction between speed and precision.
Solution Approach 2:
The patent performs preliminary coarse alignment using Fourier transform before applying the more accurate but computationally intensive feature point matching. This preliminary action removes the bulk of misalignment, allowing the subsequent fine alignment to focus only on small residual errors, thereby achieving high precision without the full computational cost.
2Adaptability or versatility
If FFT-based whole image alignment is used, then large offsets can be measured, but matching failure rate increases
Solution Approach 1:
The patent divides the alignment process into coarse and fine stages. The coarse stage uses Fourier transform to handle large offsets, while the fine stage uses salient feature points to achieve reliable matching. This segmentation allows the system to maintain both the ability to handle large offsets and low matching failure rates.
Solution Approach 2:
The patent introduces salient feature points as an intermediary between coarse Fourier-based alignment and final precise alignment. These feature points serve as reliable markers that bridge the gap between approximate and precise positioning, reducing matching failures even when large initial offsets exist.
3Ease of operation
If simplified four-point offset measurement is used, then ease of operation is improved, but alignment accuracy deteriorates
Solution Approach 1:
The patent segments the alignment process into automated coarse alignment (maintaining simplicity) and automated fine alignment using salient feature points (improving precision). Both stages are automatically executed by the system, preserving ease of operation while achieving high accuracy through the combined approach.
Data Source
AI summary
Methods and systems for image alignment are provided. One method includes selecting three or more salient feature points for use in test image to reference image alignment by applying a selected salient feature point detection method to at least a reference image generated for the specimen. The method also includes detecting the three or more salient feature points in the test image and the reference image and aligning the detected three or more salient feature points in the test image to the detected three or more salient feature points in the reference image. The method further includes aligning remaining portions of the test image to remaining portions of the reference image based on results of the previous aligning step.