Image Processing Apparatus Anomaly Detection via Characteristic Map Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image checking apparatuses face challenges in properly registering target images with reference images when registration markers are used, especially when similar objects near the registration marker exist, leading to improper anomaly detection.
Innovation Solution
An image processing apparatus that generates characteristic maps for both target and reference images using filter processes, derives a correction amount based on deviations between objects in these maps, and corrects the images to improve registration accuracy for anomaly detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a registration marker is used for registration between target image and reference image, then registration can be performed, but if an object similar to the registration marker exists near the registration marker, proper registration cannot be performed and anomaly detection fails
Solution Approach 1:
The patent extracts the registration function from the registration marker and implements it through characteristic maps generated by filter processes. Instead of relying on specific marker patterns that can be confused with similar objects, the system extracts geometric and structural characteristics from the images themselves to perform registration, thereby eliminating the vulnerability to similar object interference.
Solution Approach 2:
The patent introduces characteristic maps as an intermediary representation between the target image and reference image. These characteristic maps serve as a mediator that captures essential geometric features without relying on registration markers, enabling robust registration even when similar objects are present in the images.
2Ease of operation
If registration markers are used to align images, then image registration can be achieved, but the system becomes vulnerable to false registration when similar objects exist near markers
Solution Approach 1:
The patent changes the registration parameters from marker-based pattern matching to characteristic map-based geometric feature matching. By transforming the registration approach from relying on specific marker patterns to extracting general geometric characteristics through filter processes, the system achieves robust registration precision that is not affected by similar objects.
Data Source
AI summary
An image processing apparatus compares a target image and a reference image and thereby detects an anomaly in the target image, and includes an anomaly detecting unit. The anomaly detecting unit is configured to (a) generate a first characteristic map obtained by performing a filter process for the target image and a second characteristic map obtained by performing the filter process for the reference image, (b) derive a correction amount on the basis of a deviation between an object in the first characteristic map and an object in the second characteristic map, and (c) correct the target image and/or the reference image with the correction amount and thereafter compare the target image and the reference image and thereby detect an anomaly in the target image.