Image Processing Apparatus for Print Defect Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing systems fail to accurately detect protrusions of print products from the imaging range, leading to erroneous defect detection, and lack a method to differentiate between intended protrusions due to paper size or conveyance displacement and actual abnormalities.
Innovation Solution
An image processing apparatus that acquires print paper size, sets thresholds for protruding portions based on paper size, detects protrusions, and identifies conveyance displacement abnormalities by determining permissible flags and lengths or areas for protrusions within the imaging range.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the protruding region is set as a non-inspection region to prevent erroneous detection, then false defect detection is reduced, but the ability to detect actual conveyance displacement abnormalities is lost
Solution Approach 1:
The patent applies local quality by differentiating the treatment of different regions within the image. Instead of uniformly excluding all protruding regions from inspection, the system selectively applies non-inspection status only to specific protruding regions that are determined to be normal variations based on paper size and conveyance displacement thresholds. This allows the system to maintain high defect detection accuracy in inspected regions while avoiding false positives in legitimately protruding regions.
Solution Approach 2:
The patent utilizes parameter changes by dynamically adjusting the protrusion threshold based on paper size and conveyance displacement characteristics. The system calculates permissible protrusion distances as parameters that change according to the specific printing conditions, allowing the inspection system to adapt its sensitivity and accurately distinguish between normal protrusions and actual defects or abnormalities.
2Area of stationary object
If comparison is made by extrapolation of pixel value to protruding region, then complete coverage inspection is achieved, but erroneous defect detection occurs in protruding regions
Solution Approach 1:
The patent applies local quality by implementing region-specific inspection strategies. The system identifies protruding regions and applies different processing rules to them compared to normal regions. Specifically, protruding regions are marked as non-inspection regions when they exceed permissible thresholds, preventing erroneous defect detection while maintaining thorough inspection coverage in valid areas.
Solution Approach 2:
The patent segments the image into different regions based on protrusion characteristics. By dividing the image into inspected regions and non-inspected protruding regions, the system can apply appropriate processing to each segment, ensuring comprehensive coverage where needed while avoiding false positives in protruding areas.
3Device complexity
If a fixed non-inspection region is set for all print products, then processing is simplified, but adaptability to different paper sizes and conveyance conditions is reduced
Solution Approach 1:
The patent implements dynamics by making the non-inspection region boundaries dynamic rather than fixed. The protrusion thresholds are calculated based on paper size and conveyance displacement characteristics, allowing the inspection system to automatically adapt to different printing conditions. This dynamic approach maintains relatively simple processing logic while achieving high adaptability across various paper sizes and conveyance scenarios.
Data Source
AI summary
An image processing apparatus includes a print setting acquisition unit configured to acquire a print paper size, a threshold setting unit configured to set, based on the print paper size, a threshold for a protruding portion of a print product, the protruding portion protruding from a reading range, an image acquisition unit configured to acquire a scanned image of the print product, a detection unit configured to detect the protruding portion from the scanned image, and a conveyance displacement abnormality detection unit configured to detect conveyance displacement abnormality from the threshold and a result of the detection.