Variable-Data Printing Inspection via Static Area Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing digital printing inspection systems struggle to automatically detect printing defects in variable-data printing, particularly in areas with unique serial numbers, track and trace elements, and variable barcodes, as they cannot effectively differentiate between static and variable graphics, leading to inefficiencies in defect detection.
Innovation Solution
The system divides a page into non-overlapping areas with unique identifiers, generating rasterized static and variable images for each area, allowing for efficient comparison of captured images with pre-stored defect-free images to detect printing defects such as spots, smears, and color shifts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a reference image is created for every printed instance to enable accurate defect detection in variable-data areas, then measurement precision is improved, but storage requirements and bandwidth requirements become prohibitive
Solution Approach 1:
The page is divided into multiple non-overlapping areas, with each area containing either static graphics only or variable graphics. This segmentation allows the system to store reference images only for static areas and use OCR/barcode analysis for variable areas, dramatically reducing storage requirements while maintaining defect detection capability.
Solution Approach 2:
The patent extracts variable data areas from the overall page image and handles them separately using OCR and barcode quality analysis algorithms. This extraction eliminates the need to store reference images for variable areas, as these can be verified through algorithmic analysis rather than image comparison.
2Ease of operation
If OCR techniques or barcode quality analysis algorithms are used for variable text or barcode areas, then ease of operation is improved, but manufacturing precision deteriorates as these algorithms cannot detect printing defects such as spots, smears, and color shift
Solution Approach 1:
Different inspection methods are applied to different areas of the page based on their characteristics. Static areas use image comparison for comprehensive defect detection (spots, smears, color shifts), while variable areas use OCR and barcode analysis appropriate to their content type. This local quality approach ensures each area is inspected by the most suitable method.
3Productivity
If a complete unique image is stored for every instance of a page to enable inspection at printing speed, then productivity is improved, but storage requirements and bandwidth requirements become prohibitive
Solution Approach 1:
The system performs preliminary processing by dividing the page into static and variable areas before printing. Reference images for static areas are prepared in advance and stored, while variable areas are marked for algorithmic inspection. This preliminary action enables rapid inspection during printing without requiring storage of complete unique images for each instance.
Data Source
Figure 1
Figure 2
Figure 3A
AI summary
A method and system are presented for performing automatic inspection on the results of a print job that prints variable data. The print job is to generate multiple printed instances of a page containing variable graphics, and the inspection is to detect any printing defects on a printed instance of the page. In one embodiment, the method includes dividing the page into multiple non-overlapping areas using a division map prepared for the page. The non-overlapping areas include one or more static areas that include only static graphics and one or more areas that include variable graphics. The method includes capturing an image of the printed instance and, for each of the non-overlapping areas, retrieving an indication indicating whether or not the area includes any variable graphics, retrieving a rasterized image of an instant of the area from a memory location according to the indication, and comparing the rasterized image with the captured image of the area to detect any printing defects in the area of the printed instance.