Posture Estimation Using Test Area Statistics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing posture estimation devices require significant data processing to specify the image area of a care receiver, leading to increased computational load and inability to estimate posture changes in real time.
Innovation Solution
A posture estimation device that acquires image data from multiple sensors, uses a transition model to estimate human body posture by calculating statistics in specific test areas of the image data, allowing for real-time estimation with reduced data processing.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If image processing is performed to specify the image area of the care receiver, then the posture estimation accuracy is improved, but the data processing amount increases and real-time estimation becomes impossible
Solution Approach 1:
The patent extracts only the necessary test areas from the full image data based on the transition model, rather than processing the entire image. By identifying and analyzing only the relevant regions where posture changes occur, the system achieves accurate posture estimation while significantly reducing the data processing amount and enabling real-time monitoring.
2Reliability
If the watching area is set above the bed and image capture is performed from lateral direction, then the rising behavior detection is improved, but the data processing amount increases due to image processing requirements
Solution Approach 1:
The patent divides the image data into multiple test areas corresponding to different body parts and posture states. By segmenting the analysis into specific regions of interest rather than processing the entire image, the system maintains reliable rising behavior detection while reducing computational complexity and processing requirements.
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
An image acquisition part (15) acquires image data from an image sensor (2, 3) provided in a room. A storage (6) stores information indicating a transition model of a human body posture and test information indicating a test area of the image data in each posture of the transition model. A posture estimator (5) calculates a statistic in the test area of the image data by referring to the information indicating the transition model and the test information, and estimates a current human body posture from a human body posture before one point of time based on the statistic.