Inkjet Print Inspection Using Random Dot Pattern Training
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing inkjet printers face challenges in efficiently increasing inspection accuracy for various characters and symbols, particularly in changing printing targets, due to the difficulty in individually learning fixed form characters.
Innovation Solution
Incorporating a machine learning function with a neural network that includes a random test printing function unit to print dot arrangement patterns based on various characters and an evaluation function training unit to optimize the neural network's evaluation function for improved inspection accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If machine learning is used to improve inspection accuracy, then inspection accuracy is improved, but the complexity of the inspection system increases
Solution Approach 1:
The system performs preliminary actions by pre-training the neural network with synthetic dot arrangement patterns before actual inspection. This preliminary training phase allows the machine learning model to learn basic character recognition patterns without requiring complex real-world examples, thereby improving inspection accuracy while controlling system complexity.
Solution Approach 2:
The invention uses copying by generating synthetic dot arrangement patterns that replicate the structure and characteristics of actual printed characters. These copied patterns serve as training data for the neural network, enabling the system to learn from simplified representations rather than requiring complex real character datasets, thus improving accuracy while maintaining manageable system complexity.
2Measurement precision
If fixed form characters are learned, then inspection of specific characters is accurate, but adaptability to different characters and printing targets is reduced
Solution Approach 1:
The neural network is designed with universal functionality to handle multiple character types and printing targets. By training the network with diverse dot arrangement patterns representing various characters during the preliminary action phase, the system achieves adaptability across different character forms while maintaining inspection accuracy through the unified neural network architecture.
Solution Approach 2:
The inspection system employs dynamic adaptability where the neural network can adjust its evaluation function based on the specific printing target being inspected. This dynamic capability allows the system to transition from learning fixed form characters to adapting to different characters and printing targets, balancing inspection accuracy with versatility through the machine learning framework.
3Measurement precision
If all characters are individually learned, then comprehensive inspection coverage is achieved, but the time and resources required for learning increase significantly
Solution Approach 1:
The system applies partial action by focusing the preliminary training phase on learning the fundamental dot arrangement patterns and structural characteristics of characters rather than attempting to memorize every possible character variation. This partial learning approach achieves comprehensive inspection coverage through general pattern recognition while significantly reducing the time and computational resources required compared to individual character learning.
Solution Approach 2:
The invention utilizes parameter changes in the neural network's evaluation function to achieve comprehensive character inspection coverage. By adjusting the training parameters and dot arrangement pattern variations during preliminary action, the system learns to recognize diverse character forms through parameter optimization rather than exhaustive individual learning, thereby achieving broad coverage with reduced learning time and resource consumption.
Data Source
AI summary
Provided is a novel inkjet printer system that can use machine learning to improve the inspection performance of a printing inspection device. The inkjet printer system includes, in the machine learning, a random test printing function unit (301) that has a function that prints a random dot arrangement that is based on various characters printed in the printing area of a printing target, and an evaluation function training function unit (302) that has a function that optimizes an evaluation function for inspecting a printed image on the basis of a captured image of the results of the printing.