Image Matching Using Contour Similarity for Security Inspection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In security inspection scenarios, such as subway stations and airports, the synchronization of X-ray images and appearance images of articles is often disrupted due to conveyor belt stops or blockages, leading to poor matching between the two image types.
Innovation Solution
A method and apparatus for matching images that involves extracting contours from X-ray and appearance images, determining their similarity using geometry-invariant moments, and setting a preset similarity threshold to ensure accurate matching, even when the security inspection equipment and image acquisition apparatus are not synchronized.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If time-based matching is used to associate X-ray images and appearance images, then the matching process is simple, but the matching accuracy deteriorates when conveyor belt stops or blockages occur
Solution Approach 1:
The patent replaces the mechanical time-based synchronization system with a content-based image matching system. Instead of relying on temporal correlation between X-ray and appearance images, the system extracts contours from both image types and performs geometric pattern matching. This substitution allows accurate association of images even when timing relationships are disrupted by conveyor belt stops or blockages.
Solution Approach 2:
The patent changes the matching parameter from temporal (time difference between image captures) to spatial/geometric (contour shape similarity). By transforming the matching criterion from time-based to geometry-based, the system becomes insensitive to conveyor belt interruptions while maintaining matching accuracy through contour comparison.
2Measurement precision
If contour-based matching with geometry-invariant moments is used, then image matching accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent extracts only the essential geometric features (contours) from the images, discarding irrelevant information such as texture, color, and internal details. By using geometry-invariant moments that capture only the essential shape characteristics, the system achieves accurate matching with reduced computational burden compared to processing complete images.
Solution Approach 2:
The patent applies local quality by focusing computational resources on the most discriminative feature - the external contour shape. The geometry-invariant moments capture local geometric properties that are sufficient for identification, allowing accurate matching without the need to process the entire image data, thus balancing accuracy and complexity.
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
The disclosure provides a novel method and apparatus for matching images. The method for matching images comprises: extracting contours of an X-ray image and an appearance image (S202); determining a similarity between the contours of the X-ray image and the appearance image (S204); and when the similarity between the contours of the X-ray image and the appearance image is larger than or equal to a preset similarity threshold, determining that the X-ray image and the appearance image are images of a same article (S206). The method and apparatus for matching images according to embodiments of the disclosure may improve the accuracy of matching images and thus improve security inspection efficiency.