Document Image Defect Verification via Front-Back Correspondence
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image evaluation devices may incorrectly identify defects in digital images of financial documents, leading to unnecessary rescanning and increased costs, as they fail to accurately verify the presence of symmetric defects across both sides of a document.
Innovation Solution
The system compares defects identified in digital images of the front and rear sides of a document to determine if they correspond, using modules for defect detection and correspondence to output flags indicating the presence of defects that hinder processing, thereby reducing misidentification and improving image quality verification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If an image evaluation device identifies defects in a single image of a document side, then defect detection speed is improved, but measurement precision deteriorates due to incorrect identification of symmetric defects
Solution Approach 1:
The patent transitions from analyzing a single two-dimensional image to analyzing multiple images (front side and back side) of the document. By adding the dimension of multiple document sides, the system can cross-validate defect identifications and eliminate false positives caused by symmetric defect patterns, thereby improving measurement precision without significantly compromising detection speed.
2Device complexity
If an image evaluation device processes only one side of a document, then device complexity is reduced, but reliability deteriorates due to inability to verify symmetric defects
Solution Approach 1:
The system performs preliminary defect identification on the first document side, then uses the second document side image to verify symmetric defects. This preliminary action approach allows the system to quickly identify potential defects and then selectively verify them, improving reliability while maintaining reasonable processing complexity through targeted rather than exhaustive analysis.
3Measurement precision
If the image evaluation device requires manual verification of all identified defects, then measurement precision is improved, but productivity deteriorates due to increased processing time
Solution Approach 1:
The system performs self-verification by automatically comparing defect patterns across multiple document sides. Instead of requiring manual verification, the device uses its own multi-image analysis capability to self-validate defect identifications, eliminating the need for human intervention while maintaining high accuracy. This self-service approach preserves productivity by avoiding manual review bottlenecks.
Data Source
AI summary
In general, this disclosure describes techniques of reducing the possibility of an image evaluation device incorrectly identifying defects in images of documents. Some defects may be apparent in the images of both sides of a document. For instance, a tear in a document could be apparent in an image of a front side and an image of a rear side of the document. However, the image evaluation device could erroneously identify such a tear in an image of one side of the document. In this case, the tear would not be apparent in the image of the other side of the document. To reduce the possibility that the image evaluation device erroneously identifies such a defect, the image evaluation device may determine whether defects identified in the image of the front side of the document correspond to defects identified in the image of the rear side of the document.