Multi-feed Detection Using Image Analysis and Edge Continuity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional sheet conveyance devices face challenges in detecting multi-feed, especially when sheets are unaligned, and require dedicated sensors, leading to increased costs and inefficiencies.
Innovation Solution
An information processing apparatus that uses image processing to detect multi-feed by evaluating edge continuity, calculating document area size, and comparing it with user-specified sizes, allowing for multi-feed detection regardless of sheet alignment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If dedicated sensors are used for multi-feed detection, then detection reliability is improved, but device cost and complexity increase
Solution Approach 1:
The scanner unit, originally designed for document imaging, is made multi-functional by enabling it to perform both image capture and multi-feed detection. The image processing unit analyzes scanner data to detect multi-feed conditions, eliminating the need for dedicated sensors while maintaining detection reliability through image-based analysis of document edges and overlaps.
Solution Approach 2:
The patent replaces the mechanical/optical sensor-based detection system with an image processing-based system. Instead of using dedicated sensors to detect document thickness or position, the system uses digital image analysis to identify multi-feed conditions by examining edge continuity, document areas, and overlap patterns in the scanned images.
2Measurement precision
If conventional multi-feed detection methods are used, then detection capability is improved for aligned sheets, but detection accuracy deteriorates for unaligned sheets
Solution Approach 1:
The patent transitions from one-dimensional thickness measurement (using sensors that measure document height) to two-dimensional image analysis. By analyzing the scanner image data, the system can detect document edges, calculate document areas, and identify overlaps in both horizontal and vertical dimensions, enabling accurate multi-feed detection regardless of sheet alignment or inclination.
Solution Approach 2:
The system creates a digital copy of the document through scanning and uses this copy for analysis. The image processing unit analyzes the scanned image to detect multi-feed conditions by examining edge continuity, document areas, and overlap patterns, allowing detection of unaligned sheets that would be missed by traditional thickness-based sensor methods.
3Device complexity
If image processing is used for multi-feed detection, then device cost is reduced by eliminating dedicated sensors, but detection precision may deteriorate
Solution Approach 1:
The image processing analysis is segmented into multiple independent detection methods: edge continuity analysis, document area calculation, and overlap detection. By dividing the detection task into these separate analytical components, the system achieves high detection precision through multiple verification points while using only the existing scanner and image processing unit.
Solution Approach 2:
The system uses feedback from the scanner unit to provide real-time image data to the image processing unit. The multi-feed determination unit receives both the scanned image data and original data, continuously analyzes them for multi-feed conditions, and can trigger feedback mechanisms (such as stopping the document feed or alerting the user) when multi-feed is detected, ensuring high detection precision through iterative verification.
Data Source
AI summary
An information processing apparatus includes a reading unit, an edge extraction unit, a feature point search unit, a document area size calculation unit, a user operation unit, and a multi-feed determination unit. The reading unit optically reads a document to generate image data. The edge extraction unit extracts an edge of the document from the image data. The feature point search unit searches edge data of the edge extracted by the edge extraction unit for a feature point. The document area size calculation unit calculates a document area size from the feature point found by the feature point search unit. The user operation unit is used by a user to specify a document size. The multi-feed determination unit compares the document area size calculated by the document area size calculation unit with the user-specified document size so as to determine whether multi-feed of the document has occurred.