Machine Vision Lighting Setup for Joint Inspection Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In industrial product appearance checking and other judgment systems using machine learning, optimizing lighting and check algorithm parameters simultaneously is time-consuming and often results in suboptimal accuracy due to the need for manual adjustments and the complexity of multiple imaging and lighting parameters.
Innovation Solution
A method that uses a machine learning model to optimize both lighting and check algorithm parameters by capturing images under varying lighting conditions, comparing estimation results with label data to set the optimal lighting and algorithm parameters, thereby simplifying operations and improving accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual optimization of lighting and check algorithm is performed alternately and repeatedly, then detection performance can be achieved, but the process becomes very time-consuming
Solution Approach 1:
The patent combines lighting parameter optimization and check algorithm parameter optimization into a single unified process. The loss function simultaneously evaluates both lighting conditions and algorithm parameters, allowing them to be optimized together rather than alternately and repeatedly, thereby significantly reducing the time required while achieving the same detection performance.
Solution Approach 2:
The patent introduces a loss function as an intermediary mechanism that bridges lighting parameters and check algorithm parameters. This loss function evaluates both aspects together and guides the optimization process, enabling coordinated optimization without requiring separate manual adjustments for each component.
2Ease of operation
If lighting is regulated to be easily observed by the operator, then observability is improved, but optimal checking accuracy may not always be achieved
Solution Approach 1:
The patent enables the system to automatically determine optimal lighting conditions based on detection performance requirements rather than operator preference. The machine learning model self-adjusts lighting parameters to maximize detection accuracy, eliminating the need for operator intervention and ensuring optimal checking accuracy is achieved.
3Measurement precision
If multiple imaging and lighting parameters are changed and optimized simultaneously, then comprehensive optimization is achieved, but relatively long time is required
Solution Approach 1:
The patent implements continuous joint optimization of lighting and algorithm parameters through automated machine learning processes. Rather than performing discrete manual adjustments, the system continuously evaluates both parameters together using the loss function and automatically adjusts them, achieving comprehensive optimization without the time penalty of manual iteration.
4Ease of operation
If design parameters of lighting are included in machine learning check algorithm parameters, then direct optimization of both is enabled, but system complexity increases
Solution Approach 1:
The patent uses the loss function as an intermediary that simplifies the optimization process despite the increased complexity of having both lighting and algorithm parameters. The loss function provides a unified evaluation metric that guides the machine learning process, making the complex joint optimization manageable and straightforward to implement.
Data Source
AI summary
The present disclosure relates to a method, device, system and computer-program product for setting a lighting condition when an object is checked and a storage medium. The method includes that: the object is lighted by light sources capable of changing lighting parameters, and the object is captured by an image sensor in such lighting parameters to obtain captured images, wherein the object has known label data; and a part of or all of the captured images and the corresponding label data of the object are applied to learning of a machine learning model, and the lighting condition and the check algorithm parameters of the machine learning model is set simultaneously by optimizing both the lighting parameters and the check algorithm parameters, on the basis of a comparison result between an estimation result of the machine learning model and the label data. Therefore, operations are simplified.