Document Edge Detection via On-Line Pixel Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing technologies inaccurately detect document edges when content with a gray level similar to the document backing surface is near the edge, leading to potential erasure of document content during edge correction.
Innovation Solution
An image processing apparatus and method that includes a first edge pixel detector, a straight line detector, a classifying module, a second edge pixel detector, and a determining module to accurately classify and identify document edges by distinguishing between on-line and non-on-line edge pixels, determining whether non-on-line edge pixels represent document edges based on connection via second edge pixels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If simple edge detection methods are used to identify document edges for correction, then processing speed is improved, but measurement precision deteriorates causing erroneous detection of content as document edges
Solution Approach 1:
The patent segments the edge detection process into multiple stages: first edge pixel detection, straight line detection, classification into on-line and non-on-line edge pixels, and second edge pixel detection. This multi-stage segmentation allows the system to maintain high processing speed while achieving accurate edge detection by handling different aspects of edge detection in separate processing steps.
Solution Approach 2:
The patent introduces straight line detection as an intermediary step between simple edge pixel detection and final edge determination. This intermediary mechanism helps distinguish actual document edges from content elements by checking whether detected edge pixels align with straight lines, thereby improving measurement precision without significantly increasing processing complexity.
2Manufacturing precision
If aggressive edge correction is applied to remove detected edge regions, then document edge quality is improved, but object-generated harmful factors worsen due to accidental erasure of content
Solution Approach 1:
The patent performs preliminary classification of edge pixels into on-line and non-on-line categories before applying correction. By identifying and classifying edge pixels in advance based on their relationship to detected straight lines, the system can apply correction only to appropriate regions, improving edge quality while preventing accidental erasure of content elements that are not actual document edges.
Solution Approach 2:
The patent uses feedback from straight line detection results to guide the correction process. The classification of edge pixels based on their alignment with detected straight lines provides feedback that informs which regions should be corrected and which should be preserved, thereby achieving high correction quality without harming content.
Data Source
AI summary
Provided are an image processing apparatus, image processing method and a computer-readable, non-transitory medium that can accurately detect document edges from a readout image. The image processing apparatus includes a first edge pixel detector for detecting a plurality of first edge pixels from an input image, a straight line detector for detecting a straight line from the first edge pixels, a classifying module for classifying the first edge pixels into on-line edge pixels and non-on-line edge pixels, a second edge pixel detector for detecting second edge pixels located between two of the on-line edge pixels, when there is any one of the non-on-line edge pixels between two of the on-line edge pixels, and a determining module for determining whether the non-on-line edge pixel represents a document edge, based on whether the two on-line edge pixels are connected by the second edge pixels via the non-on-line edge pixel.