Code Image Detection in Document Scanning via Region Prioritization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image reading systems face challenges in efficiently detecting and processing code images, such as QR codes, within documents, especially when the document is not in an upright position, leading to difficulties in determining the correct scanning region and prioritizing code images for information extraction.
Innovation Solution
An information processing device with a scanner and setting unit that detects code images within scanned regions, analyzes their orientation, and sets a scanning region based on a user-designated reference position, allowing for targeted scanning and decoding of code images without scanning the entire document surface.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the scanner scans the entire document surface to detect code images, then the detection coverage is complete, but the scanning time and processing load increase significantly
Solution Approach 1:
The patent divides the document scanning process into multiple passes or regions. Instead of scanning the entire document surface at once, the scanner focuses on specific regions containing code images detected in previous scanning passes. This segmentation allows the system to maintain complete detection coverage while significantly reducing the scanning time and processing load for subsequent passes.
2Reliability
If the scanner scans the entire document surface to ensure no code images are missed, then the detection reliability is high, but the productivity decreases
Solution Approach 1:
The patent performs preliminary scanning to detect code images and their orientations before conducting the main information extraction process. By identifying code images and determining their correct scanning regions in advance, the system ensures reliable detection while improving overall productivity. The preliminary action of detecting code image locations and orientations allows subsequent scanning to be focused and efficient.
3Ease of operation
If the system scans regions far from the reference position first, then the scanning follows a conventional pattern, but the time to detect and process the most important code images increases
Solution Approach 1:
The patent inverts the conventional scanning approach by prioritizing regions closer to the reference position over regions farther away. Instead of scanning from the outer edges or in a fixed sequential pattern, the system identifies and scans regions containing code images near the reference position first. This inversion allows the system to quickly extract the most important information while maintaining operational simplicity through automated region prioritization.
Data Source
AI summary
An information processing device includes: a scanner that performs scanning of a read image of a document and detects a code image included in a scanned region, which is a region where the scanning is completed; and a setting unit that sets, based on the code image detected by the scanner, a scanning region when the scanner scans an unscanned region, in which the scanning by the scanner has not been performed yet.