Document Edge Detection for Damaged Paper Using Feature Distribution
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image processing techniques struggle to accurately detect document edges in damaged documents with features like waviness, folding, or curl, leading to errors in skew correction and registration alignment.
Innovation Solution
An image processing apparatus with a first edge point group detection unit, edge point feature amount calculation unit, distribution calculation unit, edge point selection unit, and edge straight line calculation unit is employed to detect and correct edge points, even in damaged documents, by calculating feature amounts and selecting edge points corresponding to maximum distributions to approximate an edge straight line.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If edge points from damaged portions (waviness, folding, curl) are used for linear equation calculation, then more edge points are available for calculation, but edge detection accuracy deteriorates due to errors in the edge straight line
Solution Approach 1:
The patent extracts and removes edge points from damaged portions (waviness, folding, curl) from the set of edge points used for linear equation calculation. By identifying and excluding these problematic edge points that deviate from the true document edge, the system maintains high edge detection accuracy while still utilizing sufficient edge points from undamaged regions for robust calculation.
Solution Approach 2:
The patent applies different treatment to different regions of the document edge. Edge points in damaged portions are identified and excluded, while edge points in intact regions are retained for calculation. This local differentiation ensures that only high-quality edge points contribute to the edge straight line determination, resolving the contradiction between quantity and accuracy.
2Productivity
If conventional edge detection techniques are used, then processing is simple and fast, but edge detection accuracy deteriorates for damaged documents
Solution Approach 1:
The patent performs preliminary classification of edge points to identify and separate edge points from damaged portions before conducting the linear equation calculation. This preliminary action of filtering out problematic edge points ensures that subsequent processing uses only high-quality data, achieving both accuracy for damaged documents and maintained processing efficiency through automated classification.
Data Source
AI summary
An image processing apparatus is to: detect a first edge point group comprising a set of edge points between a document background area and a document area in a read image of a document; calculate feature amounts from a plurality of edge points in the first edge point group and compose an edge point feature amount group; calculate a distribution of the feature amounts of the plurality of edge points from the edge point feature amount group and calculate a feature amount corresponding to a maximum distribution; select edge points corresponding to the maximum distribution from the first edge point group and select a second edge point group using a range of the feature amount corresponding to the maximum distribution; and calculate an edge straight line approximating the second edge point group.