Worker Posture Analysis for Multi-Operator Efficiency Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies for analyzing work efficiency statuses of workers in factories face challenges in precision when dealing with multiple workers, as they require extensive setting operations for each worker, becoming burdensome for administrators.
Innovation Solution
A work analyzing device and method that uses joint position estimation and posture match determination based on video recordings, employing machine learning to estimate joint positions and determine posture matches with reference postures, reducing the need for individual worker settings.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If image pattern extraction and fixed portion matching are used to analyze worker efficiency, then analysis can be performed with simple technology, but analysis precision deteriorates when multiple workers with different holding styles are analyzed
Solution Approach 1:
The system changes from analyzing fixed image patterns to analyzing dynamic posture parameters. By extracting joint positions and comparing posture sequences, the system adapts to different workers' holding styles while maintaining analysis precision. The posture parameters (joint coordinates, body orientation) are changed to be worker-independent yet task-specific.
Solution Approach 2:
The patent replaces the mechanical image pattern matching system with a posture-based recognition system. Instead of matching fixed visual patterns that vary by worker, the system uses joint position estimation and posture sequence comparison, which are invariant to individual holding styles but specific to the work task.
2Measurement precision
If individual setting operations are performed for each worker to improve analysis precision, then analysis precision is improved, but administrative burden increases
Solution Approach 1:
The posture-based analysis system serves multiple workers simultaneously with a single configuration. By focusing on task-specific posture sequences rather than worker-specific image patterns, the system achieves universality across different operators while maintaining precision for each individual's work style.
Solution Approach 2:
The system performs automatic posture extraction and comparison without requiring administrator intervention for each worker. The joint position estimation and posture match determination are executed automatically, eliminating the need for manual setting operations while maintaining high analysis precision.
3Productivity
If the number of analysis targets is increased to improve productivity, then overall productivity is improved, but the burden on administrators to perform settings increases
Solution Approach 1:
The system performs preliminary posture model creation by automatically extracting joint positions and determining posture sequences from video data. This preliminary action establishes a worker-independent posture reference that can be applied to multiple workers without additional settings, enabling scalable analysis.
Solution Approach 2:
The system automatically handles the scaling to multiple workers through self-service posture extraction and comparison. Each worker's video data is processed automatically to extract joint positions and compare against the posture model, eliminating the need for administrators to perform repetitive setting operations as the number of analysis targets increases.
Data Source
AI summary
In order to enable, even when an increased number of workers become analysis targets, efficient analysis on work efficiency statuses of workers without putting a greater burden on an administrator to perform setting operations for the analysis, analytical information is produced by performing a joint position estimation based on respective frames of video recordings to estimate joint positions of a worker, performing a posture match determination based on results of the joint position estimation to determine whether or not to achieve a posture match between a posture of the worker in each frame of the video recordings and a predetermined reference body posture, and measuring, based on results of the posture match determination, a working time and a presence time of the worker used as analytical information on a work efficiency status of the worker.