Wearable Camera Hand Skeleton Detection for Operation Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current operation-correctness determination methods in manufacturing, which rely on image recognition, face challenges in reducing the learning load when the operator's hand or target object changes, requiring extensive relearning and increased complexity in nonconformity conditions or workpiece types.
Innovation Solution
An operation-correctness determining apparatus and method utilizing a wearable camera, hand skeleton detector, motion detector, and AI with image recognition capabilities to capture and analyze the operator's hand behavior and target object, allowing for reduced learning load by detecting pre-learned patterns in hand skeleton and behavior across variations in hand size and target object changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Difficulty of detecting and measuring
If image recognition technology is used to determine operation correctness, then the ability to detect and measure hand behavior and target object is improved, but the learning load increases significantly when operators or target objects change
Solution Approach 1:
The patent segments the hand into key anatomical points (wrist, finger joints, palm center) and tracks their coordinates independently. This segmentation allows the system to detect hand behavior through simple coordinate changes rather than complex image recognition, reducing learning load when operators change while maintaining detection accuracy.
Solution Approach 2:
The patent changes the detection parameters from complex image recognition to simple coordinate tracking of key hand points. By monitoring positional parameters (x, y coordinates) of wrist, finger joints, and palm center rather than recognizing hand images, the system reduces learning requirements while maintaining detection capability.
2Measurement precision
If extensive relearning is performed when target objects change, then the measurement precision of operation correctness is maintained, but the productivity decreases due to time loss
Solution Approach 1:
The patent performs preliminary detection of target object coordinates and hand key point coordinates before operation. By pre-establishing the positional relationship between hand points and target object, the system can determine operation correctness through simple coordinate comparison without requiring relearning when target objects change, thus maintaining precision while improving productivity.
Solution Approach 2:
The patent creates a coordinate-based model (copy) of the target object position and hand key point positions. This coordinate copy allows the system to compare actual operation against the pre-established coordinate model without needing to relearn the target object characteristics, enabling quick adaptation to different target objects while maintaining measurement precision.
3Adaptability or versatility
If complex image recognition models are used to handle various nonconformity conditions, then the adaptability to different workpiece types is improved, but the device complexity increases
Solution Approach 1:
The patent changes from complex image recognition parameters to simple coordinate parameters of hand key points and target object. By monitoring positional parameters rather than image features, the system achieves adaptability to different workpiece types through coordinate comparison while significantly reducing system complexity.
Solution Approach 2:
The patent discards the complex image recognition approach and recovers the simpler coordinate-based detection method. By abandoning the need for extensive image analysis and focusing on key point coordinate tracking, the system maintains versatility in handling various nonconformity conditions while reducing device complexity.
Data Source
AI summary
An operation-correctness determining apparatus capable of determining whether a predetermined operation has been performed correctly includes a wearable camera, a target object detector, a hand skeleton detector, a motion detector, and an operation-correctness determiner. The wearable camera captures a region larger than or substantially equal to an operator's field of view. The target object detector detects a target object within a captured region captured by the wearable camera. The hand skeleton detector detects an operator's hand skeleton within the captured region. The motion detector detects an operator's hand behavior from time series variation of the detected target object and from time series variation of the detected operator's hand skeleton. The operation-correctness determiner determines whether at least the detected target object and the detected operator's hand behavior substantially match a pre-learned target object and a pre-learned operator's hand behavior, and if so, determines that the predetermined operation has been performed correctly.