Large-Format Scanner Document Edge Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting document width and position in large-format scanners face challenges such as low accuracy due to dirt, brightness variations, and similar colors between the reflector and document, leading to incorrect edge detection and limited format recognition.
Innovation Solution
A method involving pre-processing to reduce optical and reflector interference, followed by rough and fine edge detection using statistical variables and edge detection methods, allowing for accurate detection even with minimal shadows and small brightness differences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a white reflector is used to match common document backgrounds and handle transparent originals, then compatibility and transparency handling are improved, but the ability to distinguish document edges from reflector edges is worsened
Solution Approach 1:
The patent applies preliminary action by performing pre-processing of image information before document arrival to reduce interference from soiling, and by using rough detection to identify transition areas before fine detection of document edges. This staged approach allows the system to maintain a white reflector for compatibility while achieving accurate edge detection through preparatory processing steps.
Solution Approach 2:
The patent segments the detection process into distinct stages: pre-processing to reduce interference, rough detection to identify transition areas, and fine detection to precisely locate document edges. This segmentation allows the system to handle the challenge of white reflector-document similarity by breaking down the complex detection task into manageable steps with different objectives.
2Device complexity
If threshold value method is used for width detection, then simplicity of implementation is improved, but reliability of detection is worsened due to dirt and brightness variations
Solution Approach 1:
The patent applies preliminary action by implementing a pre-processing step that reduces interference from soiling of optics and reflector before the main detection process. This preliminary cleaning of the image information allows subsequent simple threshold-based detection to remain reliable despite the presence of dirt and brightness variations.
Solution Approach 2:
The patent applies local quality by focusing detection efforts on specific transition areas between reflector and document rather than analyzing the entire image uniformly. By identifying and concentrating on these critical transition zones, the system achieves reliable detection even with simple threshold methods, as the local characteristics in transition areas provide sufficient contrast information.
3Measurement precision
If edge-oriented method is used for segmentation, then detection capability under minimal shadow conditions is improved, but susceptibility to false edges from disturbances is worsened
Solution Approach 1:
The patent applies preliminary action by pre-processing image information to reduce interference from soiling before edge detection. This preliminary step removes or reduces false edges caused by dirt on optics and reflector, allowing the edge-oriented method to operate on cleaner image data with fewer false positives.
Solution Approach 2:
The patent segments the detection process into rough detection of transition areas followed by fine detection of document edges. This segmentation allows the system to first identify general transition zones where document edges are likely to be found, then apply more precise edge detection only in these restricted areas, reducing the impact of false edges from disturbances elsewhere in the image.
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
Enables reliable detection of document width and position with minimal interference from dirt and brightness variations, effectively handling documents with dog-ears or torn corners.
Implementation Method 1
the scanner system having image capture elements, for example CI sensors, for recording the image information, one arranged in front of it Optics and having a reflector arranged opposite the image acquisition elements
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
The invention relates to a method for identifying the width and position of documents (4) in a large-format scanner system (1) using recorded image information, wherein the scanner system (1) has image capture elements (2) for recording the image information, an optical system (3) arranged upstream thereof and a reflector (5) arranged opposite the image capture elements (2), having the following steps: 51 the recorded image information is preprocessed in order to reduce faults as a result of soiling of the optical system (3) and the reflector (5), 52 horizontal and/or vertical transition regions (12, 13) from the reflector (5) to the document (4) are subjected to coarse identification, and 53 horizontal and/or vertical document edges (10, 11) within the transition regions (12, 13) detected by the coarse identification as per step S2 are subjected to fine identification.