Imaging Condition Control for Accurate Moving Workpiece Inspection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing inspection systems using artificial intelligence for defect recognition in moving workpieces face accuracy issues due to blurred images from slow shutter speeds and limited lighting conditions when generating trained models, which can lead to decreased recognition accuracy.
Innovation Solution
A control method that captures image data of workpieces under varying imaging conditions, including changing shutter speeds and positions, to improve the accuracy of the learning model by adjusting conditions until the recognition accuracy meets a predetermined value.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the shutter speed is increased to capture moving workpieces clearly, then image clarity improves, but the available image capturing locations are limited due to insufficient light
Solution Approach 1:
The system dynamically adjusts the shutter speed of the imaging apparatus based on the movement speed of workpieces and lighting conditions. By making the shutter speed variable rather than fixed, the system can optimize image clarity for moving workpieces while adapting to different capturing locations with varying light availability.
Solution Approach 2:
The system changes the imaging parameters (shutter speed) according to the specific conditions of each workpiece and capturing location. This allows the system to maintain image clarity across different scenarios by adjusting parameters rather than being constrained by a fixed setting.
2Device complexity
If image data is captured with fixed imaging conditions for generating trained models, then the model training process is simplified, but the recognition accuracy decreases due to blurred images from slow shutter speeds
Solution Approach 1:
Instead of using fixed imaging conditions for all training data, the system dynamically adjusts imaging parameters during data collection. This ensures that all training images are captured with optimal clarity regardless of workpiece movement speed, improving recognition accuracy without significantly complicating the training process.
Solution Approach 2:
The system performs preliminary adjustment of imaging conditions before capturing training data. By optimizing the shutter speed and other parameters in advance based on expected workpiece movement, the system ensures high-quality training images are captured without requiring complex post-processing or retraining.
3Adaptability or versatility
If the shutter speed is decreased to capture images under limited lighting conditions, then more capturing locations become available, but the workpiece images become blurred reducing recognition accuracy
Solution Approach 1:
The system changes the shutter speed parameter based on the specific lighting conditions at each capturing location. This allows the system to expand to more locations with varying light availability while maintaining image quality by adapting the shutter speed to each environment.
Solution Approach 2:
The imaging apparatus dynamically adjusts its shutter speed according to the lighting conditions and workpiece movement characteristics at each location. This dynamic adaptation enables the system to operate effectively across diverse capturing locations without sacrificing recognition accuracy.
Data Source
AI summary
A control method for controlling a system including an imaging apparatus and a processing apparatus having a learning model to which image data is input includes capturing a first workpiece using the imaging apparatus set to a first imaging condition, thereby obtaining first image data, performing machine learning on the learning model using the first image data as supervised data, obtaining second image data using the imaging apparatus set to the first imaging condition, inputting the second image data to the trained learning model and making an estimation regarding a second workpiece based on the second image data, in a case where an accuracy of the estimation is lower than a predetermined value, obtaining third image data using the imaging apparatus set to a second imaging condition different from the first imaging condition, and performing machine learning on the learning model using the third image data as the supervised data.