Local Image Rectification for Sheet-Fed Printing Inspection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing image inspection systems in printing machines struggle to compensate for non-linear local distortions, particularly in sheet-fed printing, where the trailing edge of print sheets vibrates and causes blurriness, leading to false positives in automated quality control, especially in areas without sufficient corners and edges.

Innovation Solution

The method involves recording and digitizing printed products, comparing them to a digital reference image, calculating and applying rectification factors, and inserting anchors in areas lacking edges to enable accurate compensation for non-linear distortions, allowing for the creation of a modified reference image that can be used for ongoing production.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If automated image inspection is implemented to improve quality control efficiency, then productivity increases, but false positives occur due to non-linear distortions from sheet vibration

Engineering Contradiction:

Improvequality control efficiencyVSAvoidinspection accuracy

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The patent applies parameter changes by transforming the reference image through geometric transformation parameters (rectification factors) that compensate for non-linear distortions. The system calculates local rectification factors for different image areas based on detected distortion patterns, thereby adjusting the reference image parameters to match the distorted captured images and eliminate false positives

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The patent replaces manual inspection with an automated image inspection system that uses computer-based image processing and analysis. This mechanical substitution enables continuous automated quality control while incorporating algorithms to detect and compensate for distortion patterns, maintaining reliability through intelligent software-based correction rather than human judgment

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If rectification factors are calculated for all image areas to compensate for distortions, then measurement precision improves, but device complexity increases

Engineering Contradiction:

Improvedistortion compensation accuracyVSAvoidimage processing complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent divides the image into multiple local areas and calculates rectification factors specifically for each area rather than applying a global transformation. This segmentation allows the system to handle non-linear distortions locally while reducing overall computational complexity by processing smaller, manageable image segments independently

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent applies local quality by using area-specific rectification factors tailored to the distortion characteristics of each local image region. Instead of uniform processing, the system adapts the rectification parameters to local distortion patterns, improving measurement precision in distorted areas while maintaining simplicity in areas with minimal distortion

Inventive Principle:

Principle #3Local quality

3Adaptability or versatility

If anchors are inserted in image areas without edges to enable rectification, then adaptability improves, but manufacturing precision of the reference image decreases

Engineering Contradiction:

Improvehandling of edge-less areasVSAvoidreference image accuracy

Core Design Contradiction:

Adaptability or versatilityVSManufacturing precision

Solution Approach 1:

The patent introduces anchors as intermediary reference elements in image areas that lack natural edges or features. These artificial markers serve as mediators to establish geometric relationships and enable calculation of rectification factors in previously unprocessable areas, extending the system's adaptability without compromising overall reference image quality through proper anchor placement and processing

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS11254119B2Image inspection method with local image rectification

Publication Date: 2022.02.22 HEIDELBERGER DRUCKMASCHINEN AG

AI summary

A method for image inspection on printed products in a printing material processing machine. Printed products are recorded and digitizing and the recorded images are compared with a digital reference image to find image areas with distorted regions. Suitable rectification factors are calculated and the digital reference image is rectified with suitable rectification factors for the distorted image areas. The modified digital reference image is then compared to images recorded during the production run. If deviations are found, the printed products are found defective and removed. The computer also identifies image areas in the reference image that do not have enough edges for calculating suitable rectification factors and inserts anchors into the image areas. The anchors are printed, recorded and digitized, as they become part of the recorded digital printed image. In that case, the computer calculates the local rectification factors by way of the anchors.