Differential Image Overlapping for Defect Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Scanning Probe Microscopes, particularly Atomic Force Microscopes, face challenges in detecting small defects on flat surfaces like wafers due to limitations in the optical vision system's ability to distinguish between defects and noise, leading to incomplete defect review.
Innovation Solution
An image acquiring method that involves acquiring images of different regions on a target surface, subtracting one image from another to create a differential image, and then overlapping this differential image multiple times to enhance defect visibility, utilizing a CCD or CMOS image sensor and precision stages for accurate movement.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the optical vision system is used to observe defects on a flat surface, then the measurement process can be performed, but small defects with narrow width cannot be distinguished from background noise
Solution Approach 1:
The patent applies preliminary action by acquiring multiple images of the same region before performing differential processing. By capturing a series of images (first image series and second image series) of overlapping regions, the system prepares the necessary data in advance to enable subsequent differential image processing that enhances defect visibility while suppressing background noise.
Solution Approach 2:
The patent utilizes parameter changes by varying the acquisition conditions between two sets of images. The first and second image series are acquired under different conditions (such as different focal positions or illumination states), and by taking the differential between these series, the system transforms the parameter differences into enhanced defect contrast while eliminating static background components.
2Measurement precision
If multiple images are overlapped to achieve averaging effect, then defect visibility improves, but the processing time and complexity increase
Solution Approach 1:
The patent extracts only the differential information between two image series rather than processing all individual images separately. By calculating the differential image that represents only the changes between the first and second image series, the system achieves defect enhancement with significantly reduced processing complexity and time compared to traditional multiple-image overlapping methods.
3Measurement precision
If the probe is moved to each defect position for detailed imaging, then defect analysis can be performed, but the review process becomes time-consuming for multiple defects
Solution Approach 1:
The patent segments the wafer surface into multiple overlapping imaging regions that are acquired in parallel or sequence. By dividing the large-area wafer into manageable regions with overlaps, the system can process multiple regions simultaneously and identify defects across the entire surface without requiring sequential probing of each defect location, thereby improving overall review efficiency.
Data Source
AI summary
An image acquiring method for acquiring an image using a measurement apparatus including an image acquiring means which acquires an image of a surface of a target to be measured in the unit of predetermined size pixels and a moving means capable of moving the target to be measured, the image acquiring method includes: acquiring an image of a first region from the surface of the target to be measured through the image acquiring means; acquiring an image of a second region, which is different from the first region, by moving the target to be measured, through the moving means; acquiring a differential image by subtracting, from either the image of the first region or the image of the second region, the other image; and overlapping the differential image multiple times.