Document Processing Apparatus Image Difference Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing document comparison technologies incorrectly detect differences between images when their content is identical but positions or sizes are changed, leading to erroneous decisions.
Innovation Solution
A document processing apparatus that extracts image areas from documents, performs image recognition, compares information obtained from these areas, and decides on differences based on a decision rule, rather than just comparing images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If image comparison is performed to detect differences between documents, then visual differences can be identified, but false positives occur when image content is identical but positions or sizes are changed
Solution Approach 1:
The patent introduces image recognition technology as an intermediary between image comparison and difference decision. Instead of directly comparing images pixel-by-pixel, the system first performs image recognition to extract semantic information (objects, text, layout elements) from the images, then compares this recognized information to determine actual document differences. This intermediary step filters out false positives caused by positional or size changes while maintaining detection of genuine content differences.
Solution Approach 2:
The patent changes the comparison parameters from raw image data (pixels, coordinates, dimensions) to recognized content attributes (object types, text content, semantic relationships). By transforming the comparison basis from geometric parameters to semantic parameters, the system can ignore irrelevant changes in position and size while focusing on meaningful content differences.
2Productivity
If simple image comparison is used, then the process is fast and simple, but erroneous difference decisions are made when image positions or sizes change
Solution Approach 1:
The patent performs preliminary image recognition before the comparison operation. By pre-processing the images to extract and structure their semantic content (identifying objects, text regions, and their relationships), the system prepares the data in a form that enables rapid and accurate comparison. This preliminary action avoids the need for complex real-time analysis during comparison, maintaining productivity while improving reliability.
Data Source
AI summary
A document processing apparatus includes a processor configured to extract an image area, in which an image is disposed, from a plurality of documents to be processed, perform an image recognition process on the image area, compare information on the image of the image area for each document, the information being obtained by performing the image recognition process, and decide a difference for the documents based on a result of comparison of the information on the image and a decision rule for the documents.


