Image Alignment via Quadratic Homography for Variable Speed Inspection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Image alignment accuracy is reduced due to distortion in images captured by an image sensor row when the inspection target is not moved at a fixed speed, leading to errors in comparison and failure detection.
Innovation Solution
An image processing method that extracts feature points from both images, associates them, and estimates a conversion formula based on a quadratic equation to accurately align images, even when the inspection target's speed varies, using a homography matrix with variables that account for uniform accelerated motion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If feature point extraction and homography matrix estimation are used for image alignment, then alignment can be performed between sample image and inspection target image, but alignment accuracy is reduced when the inspection target is not moved at a fixed speed during image capture
Solution Approach 1:
The patent changes the parameters of the conversion formula from a standard homography matrix to a quadratic equation-based formula that incorporates acceleration terms. This allows the alignment to adapt to variable speed motion by accounting for uniform accelerated motion in the quadratic equation, thereby maintaining alignment accuracy even when the inspection target's speed varies during image capture.
2Ease of manufacture
If a standard homography matrix is used for coordinate conversion, then the conversion process is simple, but it cannot accurately handle distortion caused by variable speed motion of the inspection target
Solution Approach 1:
The patent transitions from a static homography matrix to a dynamic conversion formula based on quadratic equations that can adapt to changing motion conditions. The quadratic equation incorporates time and acceleration variables, allowing the conversion to dynamically adjust to variable speed motion, thus improving alignment precision while maintaining reasonable computational complexity.
Data Source
AI summary
An image processing method causing an image processing device to execute a process including: obtaining a first image and a second image captured at different timings for an identical inspection target by passing through an imaging range of an image sensor row; extracting respective feature points of the first image and the second image; associating the feature points of the first image and the feature points of the second image with each other; estimating a conversion formula to convert the feature point of the second image to the feature point of the first image based on a restraint condition of a quadratic equation, in accordance with respective coordinates of three or more sets of the feature points associated between the first image and the second image; and converting the second image into a third image corresponding to the first image based on the estimated conversion formula.