Hybrid Machine Vision Model for Automated Parameter Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional machine vision systems face challenges in efficiently and accurately training and maintaining machine vision jobs, with ML-based systems requiring significant computational resources and non-ML systems requiring extensive manual effort for setup and maintenance.
Innovation Solution
A hybrid machine vision model that uses a machine learning model to iteratively adjust machine vision jobs based on prediction values generated from training images, optimizing performance by adjusting parameters and execution orders of machine vision tools.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If machine learning is used to train machine vision systems, then accuracy and automation are improved, but computational intensity and hardware requirements increase
Solution Approach 1:
The patent segments the machine vision system into two distinct components: a training phase that uses machine learning to generate optimized parameter sets, and an execution phase that applies these pre-determined parameters without requiring machine learning computation. This segmentation allows the system to benefit from ML-based optimization during setup while avoiding continuous computational overhead during operation.
Solution Approach 2:
The system performs preliminary machine learning-based training and parameter optimization during the setup phase using training images. The optimized parameter sets are stored and then applied during runtime without requiring additional machine learning computation. This preliminary action transfers the computational burden from runtime to setup time.
2Power
If machine learning is avoided in machine vision systems, then computational load is reduced, but setup and maintenance difficulty increases
Solution Approach 1:
The system enables self-service by automatically generating optimized parameter sets through machine learning during the training phase, eliminating the need for manual setup and tuning by field engineers. The system self-optimizes its parameters based on training data, making it easier to deploy and maintain without requiring expert intervention.
3Adaptability or versatility
If extensive manual tuning is performed, then system adaptability is improved, but time consumption and complexity increase
Solution Approach 1:
The patent replaces manual mechanical tuning processes with an automated machine learning-based parameter optimization system. Instead of field engineers manually adjusting parameters, the system uses algorithms to automatically determine optimal parameter sets from training images, significantly reducing setup time and complexity while maintaining or improving adaptability.
Data Source
AI summary
Systems and methods for implementing a hybrid machine vision model to optimize performance of a machine vision job are disclosed herein. An example method includes: (a) receiving, at a machine vision job including one or more machine vision tools, a set of training images; (b) generating, by the machine vision tools, prediction values corresponding to the set of training images; (c) inputting the prediction values into a machine learning (ML) model configured to receive prediction values and output a change value corresponding to the machine vision job; (d) adjusting the machine vision job based on the change value to improve performance of the machine vision job; (e) iteratively performing steps (a)-(e) until the ML model determines that the prediction values satisfy a prediction threshold; and executing, on a machine vision camera, the machine vision job to analyze a run-time image of a target object and output an inspection result.