Print Product Image Alignment Using Region-Specific Non-Rigid Registration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing alignment techniques for inspecting print products, particularly those with pre-printed images, suffer from decreased accuracy due to varying update tendencies of control points between pre-printed and printed images, leading to incorrect correction of control points near defects and reduced alignment precision.
Innovation Solution
The method involves generating an alignment image through projection transformation of the inspection target image with a reference image, followed by separate non-rigid registration in local regions of pre-printed and print image areas, ensuring accurate alignment by treating each image type independently.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If non-rigid registration is performed using control points updated based on both pre-printed image and print image, then alignment can be performed on the entire image, but control points near defects are incorrectly corrected and alignment accuracy decreases
Solution Approach 1:
The patent divides the image into distinct pre-printed image region and print image region, and performs non-rigid registration separately for each region. This segmentation prevents control points in the pre-printed region from being incorrectly influenced by defects in the print image region, thereby maintaining alignment accuracy while avoiding erroneous corrections.
Solution Approach 2:
The patent applies different registration strategies to different regions: the pre-printed image region uses control points updated based on pre-printed image characteristics, while the print image region uses control points updated based on print image characteristics. This local differentiation ensures that each region is aligned according to its specific properties, preventing cross-contamination of registration errors.
2Manufacturing precision
If control points are updated based on both pre-printed image and print image, then comprehensive alignment can be achieved, but varying update tendencies between image types cause approximate lines to mismatch and alignment accuracy to decrease
Solution Approach 1:
The patent segments the control point update process into two independent parts: one for the pre-printed image region and another for the print image region. Each region has its own control points updated based on its specific image characteristics, eliminating the complexity of handling varying update tendencies across different image types.
Solution Approach 2:
The patent implements local quality by using different update rules for control points in different regions. The pre-printed image region uses update rules optimized for pre-printed images, while the print image region uses rules optimized for printed images, thereby maintaining high alignment precision without the complexity of unified update strategies.
3Ease of operation
If non-rigid registration is performed without separating pre-printed and print image regions, then the alignment process is simpler, but control points near defects are shifted to unexpected positions and alignment accuracy decreases
Solution Approach 1:
The patent segments the alignment process into separate operations for pre-printed and print image regions. Although this increases operational complexity slightly, it prevents control points near defects from being incorrectly shifted, thereby maintaining high alignment accuracy that outweighs the moderate increase in process complexity.
Solution Approach 2:
The patent applies local quality by treating different regions with different registration characteristics. This ensures that control points in the print image region are not incorrectly corrected by pre-printed image characteristics, maintaining alignment accuracy while using a systematic approach that manages complexity.
Data Source
AI summary
An image processing apparatus includes: an image obtaining unit configured to read a print product in which a print image is printed on a pre-printed sheet and generate an inspection target image of the print product; a first alignment unit configured to generate an alignment image by aligning the inspection target image as a whole with a reference image by means of projection transformation, the reference image indicating a correct image of the inspection target image; and a second alignment unit configured to perform alignment by means of non-rigid registration in each of local regions of the alignment image. The second alignment unit uses at least one of a print image region surrounding the print image and a pre-printed image region surrounding a pre-printed image printed on the pre-printed sheet before printing of the print image in the alignment image as the local regions.