AI Surface Inspection Using Light Indicia for Large Assembly Areas
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing machine vision inspection systems struggle with large-scale industrial inspections due to high computational costs and inefficiencies, particularly when inspecting large or complex surfaces, and are inadequate for detecting small anomalies in high-volume manufacturing processes.
Innovation Solution
A system utilizing a laser projector to strategically project light indicia onto a surface, combined with a machine learning model and CNN algorithms, allows for efficient inspection by focusing analysis on a defined area of interest, reducing computational complexity and eliminating reliance on CAD data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If machine vision inspection using cameras and sensors is implemented to inspect whether a component has been properly assembled, then inspection coverage can be increased, but computational cost and processing time increase significantly for large surfaces
Solution Approach 1:
The patent divides the large inspection surface into multiple smaller tiles or regions. Each tile is inspected independently by the CNN model, which processes only the local area containing the light indicia rather than the entire large surface. This segmentation reduces the computational burden while maintaining comprehensive inspection coverage across the full area.
Solution Approach 2:
The patent extracts only the relevant portion of the image containing the light indicia and the component of interest for CNN processing. By cropping and focusing analysis on specific regions rather than processing complete large-scale images, the system reduces computational complexity while maintaining inspection effectiveness.
2Measurement precision
If CNN algorithms are trained to detect anomalies in images of parts under inspection, then inspection accuracy improves, but computation cost limits the ability to process larger images of very large objects
Solution Approach 1:
The large inspection surface is divided into smaller tiles, and the CNN model processes only the relevant tile containing the light indicia and potential anomalies. This segmentation enables high-accuracy anomaly detection while reducing computation cost by avoiding processing of entire large images.
Solution Approach 2:
The system applies high-computation CNN analysis only to local regions containing light indicia where anomalies are most likely to occur, rather than uniformly processing entire large surfaces. This localized approach maintains detection accuracy while reducing overall computational burden.
3Productivity
If periodic human inspection is used to achieve modest improvements in quality production, then implementation simplicity is maintained, but efficiency and accuracy remain inadequate to meet modern quality standards
Solution Approach 1:
The patent replaces human inspection with an automated system combining laser projectors, cameras, and CNN-based AI analysis. This substitution dramatically improves inspection efficiency and consistency while the use of light indicia and localized processing keeps the system complexity manageable.
Solution Approach 2:
The system transforms the inspection approach by projecting light indicia onto the surface and using these optical markers to guide CNN analysis. This parameter change from direct surface inspection to indicia-based inspection improves efficiency while maintaining manageable system complexity.
4Reliability
If higher percentages of production are inspected to meet quality standards, then quality control improves, but human inspection becomes obsolete due to inefficiency
Solution Approach 1:
The patent replaces human inspection with automated CNN-based analysis that can process images at high speed. This substitution enables inspection of higher percentages of production while maintaining or improving quality control reliability, as the AI system operates continuously without fatigue or error.
Solution Approach 2:
The automated inspection system operates continuously without interruption, enabling high-volume inspection throughput. The system can process production items in real-time or near-real-time, maintaining quality control reliability while achieving productivity levels unattainable with human inspection.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enhances inspection accuracy and efficiency by simplifying code analysis and reducing computational demands, enabling precise detection of assembly distortions and defects on large surfaces without the need for extensive training data.
Implementation Method 1
a light source for projecting light indicia onto the component assembled to the workpiece
Implementation Method 2
An imager includes an image sensor system for imaging the workpiece and signaling a current image of the workpiece to the controller
Data Source
AI summary
A system for identifying accurate assembly of a component to a workpiece is disclosed. The system includes a light source for projecting light indicia onto the component assembled to the workpiece. A controller includes an artificial intelligence (AI) element defining a machine learning model that establishes a convoluted neural network trained by stored images of light indicia projected onto the component assembled to the workpiece. An imager includes an image sensor system for imaging the workpiece and signaling a current image of the workpiece to the controller. The machine learning model directs inspection of the workpiece to the light indicia imaged by said imager. The AI element determines disposition of the component disposed upon the workpiece through the neural network identifying distortions of the light indicia in the current image.