Printed Article Inspection Using Reduced-Resolution Image Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for inspecting printed objects, such as banknotes, are either time-consuming due to high computational requirements for full-resolution image analysis or miss defects when using lower resolutions, making it difficult to achieve a balance between resource efficiency and accuracy in real-time quality control.
Innovation Solution
A method involving the creation of reduced digital images by lowering resolution, color depth, and number of color channels, followed by a rough check against a reduced reference image, with only potentially faulty items undergoing a detailed full-resolution comparison.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If full-resolution digital images are used for inspection, then measurement precision and defect detection capability are improved, but computing power requirements and processing time increase significantly
Solution Approach 1:
The inspection process is segmented into two distinct stages: a rough check using reduced-resolution images for quick filtering, and a detailed check using full-resolution images only for items that failed the rough check. This segmentation allows the system to maintain high defect detection capability while significantly reducing overall computing power requirements by limiting full-resolution processing to a small subset of items.
Solution Approach 2:
Instead of applying full-resolution inspection to all items (excessive action), the system applies reduced-resolution inspection to all items and reserves full-resolution inspection only for items that require it (partial action). This approach maintains measurement precision where needed while avoiding unnecessary computational overhead for items that can be adequately inspected at lower resolution.
2Measurement precision
If full-resolution digital images are used for inspection, then measurement precision is improved, but productivity decreases due to slower processing speed
Solution Approach 1:
The inspection workflow is divided into a fast rough check phase using reduced images and a slower detailed check phase using full-resolution images. This segmentation enables the system to process items at high speed during the rough check and only apply the time-consuming detailed check to items that actually require it, thereby maintaining both measurement precision and overall productivity.
Solution Approach 2:
The system implements a periodic inspection strategy where reduced-resolution imaging is performed continuously at high speed, and full-resolution imaging is periodically applied only to items that fail the reduced-resolution check. This periodic use of high-resolution processing maintains defect detection capability while maximizing processing throughput.
3Productivity
If reduced-resolution images are used for inspection, then computing power consumption and processing time are reduced, but measurement precision and defect detection capability deteriorate
Solution Approach 1:
The system segments the inspection process into two resolution levels: reduced-resolution for initial screening and full-resolution for detailed examination. This ensures that measurement precision is maintained for items that require it while achieving high processing efficiency for the majority of items through reduced-resolution processing.
Solution Approach 2:
Instead of using reduced-resolution images for all inspections (which would lose precision), the system uses reduced-resolution images partially for initial screening and applies full-resolution images when needed. This partial use of high-resolution imaging maintains defect detection capability while achieving the processing efficiency benefits of reduced-resolution imaging for most items.
4Loss of time
If reduced digital images are created for rough check, then loss of time and computational resources is reduced, but loss of information may occur due to resolution reduction
Solution Approach 1:
The inspection system segments image processing into two stages: reduced-resolution imaging for rapid initial assessment and full-resolution imaging for detailed analysis. This segmentation minimizes information loss by ensuring that full-resolution images are available when needed, while accepting reduced information content only for the preliminary screening stage where lower detail requirements are acceptable.
Solution Approach 2:
The system performs preliminary inspection using reduced-resolution images to quickly identify items that require further examination. This preliminary action filters out items that do not require detailed inspection, thereby preventing unnecessary loss of time and resources while ensuring that information loss in the reduced images does not compromise the final inspection quality for items that proceed to detailed checking.
Data Source
Figure 1~3
Figure 2
AI summary
The invention relates to a method for inspecting printed objects (1), in particular printed matter such as banknotes, wherein a) a digital image (2) is created of the respective object (1) to be inspected and assigned to the respective object (1), b) the digital image (2) is reduced with respect to its resolution, in particular its color depth, the number of its color channels and its spatial resolution, and a reduced digital image (3) is created and assigned to the respective object (1), characterized in that c) as part of a preliminary inspection, a comparison of the reduced digital image (3) with at least one reduced reference image (5) is carried out, d) in the case of a match exceeding a predetermined first threshold value (S1), the reduced digital image (3) is recognized as belonging to a defect-free object (1) and the inspection of the respective object (1) is terminated.e) if the reduced digital image (3) does not match the reduced reference image (5), the item is classified as potentially defective; f) if an item (1) is classified as potentially defective, the item is subjected to a detailed inspection, whereby the digital image (2) assigned to the item (1) is used and this digital image (2) is compared with at least one reference image (4), whereby the item (1) is classified as "defect-free" if the match exceeds a second threshold, and as "defective" otherwise; g) that the items (1) to be examined are transported before and/or during their recording relative to the recording unit (11), in particular by a transport device (22), preferably a conveying device.h) that the items (1) are separated after their initial inspection with regard to the test result, in particular by a first switch (23a), and the items (1) identified as potentially defective are transferred to an intermediate storage area (24), and i) that the items (1) located in the intermediate storage area (24) are fed to a second switch (23b) after the final inspection and separated according to their classification.