Document Edge Detection for Double Feed Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image reading apparatuses face challenges in accurately detecting double feeding of documents without increasing costs, as current methods rely on dedicated sensors or incomplete edge detection processing.
Innovation Solution
The apparatus employs two existing reading units to read both surfaces of a document, performing edge detection on the front and rear ends of both surfaces to determine double feeding, thereby eliminating the need for additional sensors and enhancing detection accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a dedicated double feed detection sensor using ultrasonic wave is provided, then double feeding detection capability is improved, but apparatus cost increases
Solution Approach 1:
The patent applies multi-functionality by enabling the existing reading units to serve dual purposes: both reading document images and detecting double feeding conditions through edge detection processing. Instead of adding a dedicated sensor, the system utilizes the reading units' capability to process image data for detection functions, thereby eliminating the need for additional hardware while maintaining reliable double feeding detection.
2Loss of time
If edge detection processing is performed in only a part of the front and rear end portions of the document, then processing time is reduced, but double feeding detection accuracy deteriorates
Solution Approach 1:
The patent applies local quality by performing edge detection processing specifically at the front end portion and rear end portion of the document, rather than uniformly across the entire document. This localized approach concentrates computational resources on the critical regions where double feeding is most likely to occur, maintaining high detection accuracy while minimizing processing time compared to full-document analysis.
3Measurement precision
If comprehensive edge detection processing is performed on both surfaces of the document, then double feeding detection accuracy is improved, but processing time increases
Solution Approach 1:
The patent applies segmentation by dividing the document into distinct regions (front end portion and rear end portion) and processing each region independently. This segmentation allows the system to perform comprehensive edge detection on both surfaces while managing processing time through targeted analysis of specific segments rather than treating the entire document as a single unit, thereby balancing accuracy and efficiency.
Data Source
AI summary
An image reading apparatus includes: a transport unit; a first reading unit which reads one surface of a document transported by the transport unit and generates first read data of a front end portion or a rear end portion of the one surface of the document; a second reading unit which reads the other surface of the document transported by the transport unit and generates second read data of a front end portion or a rear end portion of the other surface of the document; an edge detection unit which performs edge detection processing on each of the first read data and the second read data; and a determination unit which determines that there is double feeding of documents by the transport unit when the edge detection unit detects a plurality of edges from at least either the first read data or the second read data.