Camera Inspection Overlay Images for ML Anomaly Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Advanced quality assurance techniques for camera-based inspection systems face challenges in ease of configuration and monitoring due to high throughput and frequent changeovers in manufacturing, requiring simplified hardware configurations and software-based solutions to visualize complex algorithm outputs effectively.
Innovation Solution
A video processing pipeline generates quality assurance metrics using containerized image analysis tools, combining them with image data to create enhanced images that include color and transparency information, allowing for easier monitoring and analysis, and utilizing machine learning models like convolutional neural networks for anomaly detection and dimension modification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If complex machine learning models are used for quality assurance inspection, then measurement precision and anomaly detection capability are improved, but device complexity and difficulty of monitoring increase
Solution Approach 1:
The patent introduces overlay images as an intermediary visual layer that mediates between the complex machine learning inspection process and the user. These overlay images display quality assurance metrics, anomaly locations, and inspection results in an intuitive visual format, allowing users to monitor complex ML model outputs without needing to understand the underlying algorithmic complexity.
Solution Approach 2:
The patent replaces traditional mechanical or manual inspection methods with machine learning-based automated inspection. Containerized image analysis tools and ML models automatically perform anomaly detection, dimension measurement, and quality assessment, substituting manual visual inspection with intelligent automated systems that provide precise measurements while reducing human labor.
2Measurement precision
If hardware triggers are used for camera-based inspection, then measurement precision and timing accuracy are improved, but device complexity and ease of configuration worsen
Solution Approach 1:
The patent replaces physical hardware triggers with software-based triggering mechanisms. The containerized image analysis tools and video processing pipeline use software-controlled timing and synchronization to capture and analyze images at appropriate moments, eliminating the need for external hardware trigger devices while maintaining timing accuracy through programmable control.
Solution Approach 2:
The patent creates a universal software-based inspection system where the same containerized tools and processing pipeline can handle multiple inspection scenarios without requiring specific hardware configurations. The system adapts to different inspection needs through software configuration rather than hardware changes, making it easier to deploy and configure across different applications.
3Measurement precision
If multiple camera modules are deployed for comprehensive inspection, then measurement precision and coverage are improved, but device complexity and cost increase
Solution Approach 1:
The patent employs multiple camera modules with different focal distances (wide-angle and telephoto) that serve universal inspection purposes. The same containerized image analysis tools process images from all camera types, and the system selectively uses appropriate cameras based on inspection needs, providing comprehensive coverage while maintaining a unified software platform that simplifies configuration and operation.
Data Source
AI summary
A video processing pipeline receives data derived from a feed of images of a plurality of objects passing in front of an inspection camera module forming part of a quality assurance inspection system. Quality assurance metrics for the object are generated by one or more containerized image analysis inspection tools forming part of the video processing pipeline using the received data for each object. Overlay images are later generated that characterize the quality assurance metrics. These overlay images are combined with the corresponding image of the object to generate an enhanced image of each of the objects. These enhanced images are provided to a consuming application or process for quality assurance analysis.