Noise Line Detection in Document Shadow Regions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image processing systems face challenges in detecting noise lines in scanned documents due to foreign substances like paper dust or glue on the imaging device, leading to noise lines in the image capture, which existing technologies struggle to correct with high precision.
Innovation Solution
An image processing apparatus and method that acquires input images, extracts edge pixels, detects document and shadow regions, identifies noise pixels connected in a predetermined number as noise lines, and corrects the document image based on this detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If noise line detection is performed in the entire image including document region, then detection coverage is improved, but detection precision deteriorates due to false positives from document edges and content
Solution Approach 1:
The patent divides the image into distinct regions: document region and shadow region. By segmenting the detection area, the system focuses noise line detection only on the shadow region where foreign substances appear as dark spots, eliminating false positives from document content while maintaining comprehensive detection coverage.
Solution Approach 2:
The patent extracts and isolates the shadow region from the complete image for specialized processing. By separating the shadow region detection from the document region, the system applies specific detection algorithms tailored to the characteristics of foreign substance appearances (dark spots on light background) without interference from document edges or content.
2Measurement precision
If detection threshold is lowered to identify faint noise lines, then detection sensitivity is improved, but false positive rate increases from document features
Solution Approach 1:
The patent applies different processing qualities to different regions: the shadow region receives specialized noise line detection with lowered thresholds appropriate for detecting dark spots on light backgrounds, while the document region is excluded from this specific detection process. This local quality approach allows high sensitivity detection without false positives from document features.
3Productivity
If shadow region detection range is expanded to cover entire periphery, then detection coverage is improved, but processing time increases
Solution Approach 1:
The patent performs preliminary detection of the document region using edge detection algorithms before identifying the shadow region. This preliminary action establishes boundaries that define the shadow region efficiently, allowing subsequent noise line detection to be confined to this predetermined area, thus maintaining both coverage and speed.
Data Source
AI summary
An image processing apparatus includes a processor for acquiring an input image of a document and a periphery of the document, extracting first edge pixels from the input image, detecting a document region based on the first edge pixels, detecting a shadow region from within a predetermined range outside of the document region, extracting noise pixels in the shadow region, detecting the noise pixels, a number of which connected with one another is equal to or more than a predetermined number, as a noise line, and correcting the document region of the input image based on a detection result of the noise line.