SSIM-Based Video Activity Detection for Manufacturing Workflow Monitoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing template-based matching algorithms for monitoring manufacturing operations are inefficient in handling new operations without pre-existing templates and fail to accurately assess individual performance variations, especially in scenarios involving human operators with different skill sets and patterns.
Innovation Solution
A method and system utilizing the Structural Similarity (SSIM) index for activity detection in manufacturing workflows, which involves extracting video frames, measuring SSIM indices, identifying local maxima, and applying machine learning to determine activities and assess performance, allowing for real-time monitoring and comparison across different operators.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If template based matching algorithm is used for monitoring manufacturing operations, then system can monitor known operations with available templates, but it fails to monitor completely new production operations where templates are unavailable and requires huge computation for object matching
Solution Approach 1:
The patent changes the fundamental parameter of operation monitoring from template-based pattern matching to SSIM index-based change detection. Instead of comparing video frames against pre-stored templates, the system computes SSIM indices between consecutive video frames to detect activity boundaries. This parameter change enables the system to handle any manufacturing operation without requiring pre-existing templates, thus resolving the contradiction between monitoring reliability and adaptability to new operations.
2Reliability
If template based matching algorithm is used, then known operations can be monitored, but huge set of image repository with all variants is required which decreases system efficiency due to large computation
Solution Approach 1:
The patent extracts only the essential information needed for activity detection by computing SSIM indices between consecutive video frames, rather than storing and comparing entire video frames or maintaining a huge image repository. This extraction approach reduces the data processing burden significantly while maintaining accurate activity detection, thus resolving the contradiction between detection reliability and system efficiency.
3Reliability
If template based matching algorithm is used, then general operations can be monitored, but fails to provide performance assessment of each individual operator comparing with others simultaneously
Solution Approach 1:
The patent segments the video stream into individual activity instances for each operator by detecting activity boundaries through SSIM index analysis. This segmentation enables the system to isolate and analyze each operator's performance separately, capturing detailed information about individual operations, timing, and efficiency. The system can then compare performance metrics across different operators, resolving the contradiction between general monitoring capability and individual performance assessment.
Data Source
AI summary
The present invention discloses a method and a system for monitoring manufacturing operation workflow using Structural Similarity (SSIM) index based activity detection. The method comprising receiving video data corresponding to a manufacturing operation activity, extracting a plurality of video frames from the video data, measuring SSIM index for each video frame of the plurality of video frames with respect to next consecutive video frame of the plurality of video frames, comparing the SSIM index of the each video frame with the SSIM index of next consecutive video frame of the plurality of video frames to identify one or more local maxima, and determining at least one manufacturing operation activity based on the one or more local maxima using machine learning technique.