Digital Twin Optical Quality Control With Domain-Adapted Image Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing optical quality control methods for production installations, especially in automated industrial settings, require significant engineering effort and are not adaptable to changing conditions, making them inefficient for autonomous production systems.
Innovation Solution
A computer-implemented method using a digital twin and domain adaptation via a generative adversarial network to compare real product images with synthetically rendered images, allowing for adaptive quality control by transferring images from a real domain to an artificial domain, enabling simpler image comparison and reduced engineering outlay.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional rule-based optical quality control methods are used, then quality control can be established, but significant engineering effort and complexity are required
Solution Approach 1:
The patent creates a digital twin (virtual copy) of the production installation and uses rendering to generate synthetic images that replicate real product appearances. This copying approach eliminates the need for complex rule-based quality control systems by comparing real images against rendered reference images, significantly reducing engineering effort while maintaining reliable quality assessment.
Solution Approach 2:
The patent replaces traditional mechanical/optical measurement and rule-based analysis systems with a computational imaging approach. Instead of using complex physical measurement devices and manual rule configuration, the system uses image rendering and domain adaptation algorithms to automatically assess product quality, reducing device complexity and engineering requirements.
2Reliability
If traditional optical quality control methods are used, then quality assessment is possible, but adaptability to changing conditions is poor
Solution Approach 1:
The patent implements a dynamic system where the digital twin and rendering engine can adapt to changing production conditions. When product designs or manufacturing processes change, the digital twin is updated accordingly, and new rendered reference images are generated automatically. This dynamic adaptation capability allows the quality control system to remain reliable under varying conditions without requiring manual reconfiguration of complex rules.
Solution Approach 2:
The patent changes the fundamental parameters of the quality control approach by transitioning from rule-based threshold checking to rendered image comparison. By adjusting rendering parameters to match actual camera settings (intrinsic and extrinsic parameters) and using domain adaptation to align image domains, the system achieves high adaptability to different products and conditions while maintaining consistent quality assessment reliability.
3Reliability
If rule-based quality control with manual parameter adjustment is used, then quality rules can be defined, but the process is time-consuming and not scalable
Solution Approach 1:
The patent performs preliminary actions by pre-rendering reference images for all expected product variations and storing them in the digital twin. This preparation work is done in advance, so when actual quality control is needed, the system can immediately compare real images against pre-prepared references without requiring time-consuming manual parameter adjustment or rule definition, dramatically reducing commissioning time while maintaining reliable quality enforcement.
Solution Approach 2:
The patent implements self-service quality control where the rendering engine automatically generates appropriate reference images based on product CAD data and camera parameters without human intervention. The system self-adjusts to new products by automatically updating the digital twin and generating new reference images, eliminating the need for engineers to manually configure quality rules and significantly improving productivity and scalability.
Data Source
AI summary
For optical quality control, a product image of a production installation captured by an image capture device for given intrinsic and extrinsic parameters is used and digital twin data of a digital twin of the production installation are used, to render a synthetic simulation image based on the digital twin data, wherein the rendered synthetic simulation image is based on the same intrinsic and extrinsic parameters as during product image capture, to transfer the product image from a real domain into an artificial domain by a trained domain adaptation and in the process to generate a synthetic product image from the product image with domain transfer parameters obtained by the training, to compare the synthetic product image with the synthetic simulation image by a comparison operator, and to output a comparison result which qualitatively assesses the product.