Automated Training Image Generation for Digital Printer Error Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing digital printers face limitations in automatically identifying and tagging printing errors, leading to waste of sheet material and ink, increased maintenance, and higher costs due to inefficient error detection systems.
Innovation Solution
A computer-implemented process for automatically generating training images to train a machine learning system, which includes applying known printing errors to initial images, tagging them, and graphically manipulating these images with optical, structural, and material maps to create realistic rendered images that simulate printing on various materials, enabling the system to detect potential errors and malfunctions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If known printing errors are manually applied to images for training, then the machine learning system can be trained to detect errors, but the process is time-consuming and requires significant human intervention
Solution Approach 1:
The system automatically generates training images by applying printing errors to reference images without human intervention. The computer executes automated processes to select reference images, apply error patterns, and generate tagged training images, enabling the system to train itself efficiently
Solution Approach 2:
The system pre-generates a comprehensive database of training images with various printing errors before the machine learning training process begins. This preliminary preparation includes creating diverse error patterns and tagging them automatically, so the training can proceed efficiently without time-consuming manual intervention during the actual training phase
2Reliability
If printing errors are detected using traditional comparison methods, then the system can identify defects, but sheet material and printing inks are wasted due to delayed or inaccurate detection
Solution Approach 1:
The machine learning system continuously learns from training images and provides real-time feedback on printed material quality. The system compares detected printing errors against the trained models and immediately identifies defects, enabling rapid response that prevents further material and ink waste
Solution Approach 2:
The patent replaces traditional mechanical comparison methods with a machine learning-based automated detection system. This substitution enables faster, more accurate error identification that can operate in real-time during printing, preventing waste before it occurs rather than detecting errors after printing is complete
3Measurement precision
If a comprehensive error detection system is implemented, then printing errors can be accurately identified, but the system complexity and maintenance requirements increase
Solution Approach 1:
The machine learning system serves multiple functions: it detects various types of printing errors, continuously improves through automated learning from generated training images, and adapts to different error patterns. This multi-functionality consolidates what would otherwise require multiple separate systems into a single versatile platform, managing complexity while maintaining high detection accuracy
Data Source
AI summary
The present invention relates to a computer-implemented process for automatic generating a series of training images useful for training a machine learning software operating on a digital printer for detecting errors in digital prints. The process includes the steps of generating a series of initial digital images, applying at least one known printing error to each initial image, and automatically tagging at least one known printing error resulting in a series of images with a respective tagged error. It also forms an object of the present invention, a computerized system configured to perform the process for automatically generating training images, as well as a printing infrastructure comprising said computerized system.