Document Scanner Skew Correction via Image-Sensor Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Document scanners face inefficiencies due to the need for manual operator intervention to select scanner features based on document characteristics, leading to potential errors from document skew, mischaracterization, and false detection of multi-feeds, especially when dealing with irregularly shaped documents and partial sensor coverage.
Innovation Solution
The method involves capturing image data and combining it with sensor data to automatically determine document characteristics, ignoring conflicting sensor data and accounting for document shift or skew, thereby optimizing scanner operation and reducing false detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If manual operator selection of scanner features is used, then operator control over document processing is improved, but operator workload and potential for human error increase
Solution Approach 1:
The scanner system performs self-characterization by automatically detecting document properties using sensors and image processing, eliminating the need for manual operator intervention to identify document characteristics while maintaining accurate document classification and processing optimization
2Reliability
If document detection sensors are always enabled, then detection of multi-feeds is improved, but false detection increases due to document skew and irregular shapes
Solution Approach 1:
The system merges data from multiple sources including document detection sensors, image capture devices, and image processing algorithms to create a comprehensive document characterization. This combination allows the system to cross-validate sensor readings against actual document images, eliminating false multi-feed detections caused by document skew or irregular shapes while maintaining reliable detection of actual multi-feeds
Solution Approach 2:
The system uses image processing results as feedback to validate and correct sensor data. By comparing sensor-detected document characteristics with actual image analysis, the system can identify and correct false detections, improving the reliability of multi-feed detection while reducing false positives from document skew or irregular geometries
3Measurement precision
If external image device is added to input tray, then document characterization capability is improved, but device complexity and cost increase
Solution Approach 1:
The scanner's existing imaging components (CCD or CIS sensors used for scanning) are repurposed to perform dual functions: capturing document images for scanning purposes and simultaneously characterizing document properties (size, shape, color, etc.). This eliminates the need for separate external image devices while maintaining comprehensive document characterization capabilities
Solution Approach 2:
The system combines the scanning imaging function with document characterization function into a single integrated process. The same image capture device and processing pipeline used for creating document scans are also used to extract document characteristics, eliminating redundant hardware and circuitry while achieving accurate document measurement and classification
4Loss of time
If imaging is performed before document enters transport, then early document analysis is improved, but document skew and shifting errors increase
Solution Approach 1:
The system performs preliminary document characterization using sensors positioned at the input tray before the document enters the transport mechanism. This early detection of document properties (size, shape, presence) allows for preliminary classification and setup optimization, while the actual detailed imaging and precise measurement are performed after the document is stabilized in the transport, avoiding skew-related errors
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach eliminates the need for manual sensor enablement, corrects for document skew, and reduces false detection, enhancing the scanning process by maximizing usable sensor data and ensuring accurate document processing.
Implementation Method 1
a charge coupled device (CCD) or complementary metal oxide semiconductor (CMOS) sensor to capture an image of the document
Implementation Method 2
Document detection sensors such as ultrasonic sensors can be used to detect multiple documents fed into the transport at the same time
Data Source
AI summary
A method for scanning documents in a document scanner (10) includes transporting (45) documents from an input tray (20) to an document detection sensor (135A-135E); detecting documents with the document detection sensor; capturing image data for a document with an image capture device; determining characteristics of the document based on the image data with an image processor; comparing characteristics of the document to document detection sensor data; and excluding document detection sensor data which conflicts with the image data.