Stay Frequency Map for Automatic Bed Region Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In elderly care facilities, manually adjusting camera orientation and designating bed regions in monitoring systems is cumbersome, especially when room layouts change, requiring frequent recalibration.
Innovation Solution
An information processing device that detects moving objects, generates stay frequency maps from time-series data, and determines specific regions with high stay frequency, such as beds, using imaging from a ceiling-mounted camera, which automatically identifies and tracks frequent areas of activity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual adjustment of camera orientation and bed region designation is performed, then the watching system can correctly detect care-receiver behaviors, but the operation becomes very troublesome and time-consuming, especially in facilities with many rooms or when room layouts change
Solution Approach 1:
The system automatically designates the bed region by detecting the care-receiver's body position and determining the region where the care-receiver stays for a predetermined time or longer, eliminating the need for manual bed region designation by operators
Solution Approach 2:
The system performs automatic bed region designation before actual monitoring begins, using preliminary detection of care-receiver behavior patterns to establish monitoring parameters without requiring manual setup
2Reliability
If manual adjustment is performed for each room, then detection accuracy is maintained, but the time and effort required increases significantly when there are many rooms or when room layouts change
Solution Approach 1:
The system automatically adapts to room layout changes by detecting new care-receiver behavior patterns and recalculating bed regions without requiring manual intervention, maintaining detection reliability while eliminating setup time
Solution Approach 2:
The bed region designation is made dynamic and adaptable, allowing the system to automatically adjust to changes in room layout by detecting updated care-receiver behavior patterns over time
Data Source
AI summary
The present disclosure relates to an information processing device, an information processing method, a program, and a watching system that enable determination of a specific region, such as the region occupied by a bed, in accordance with a captured image.An information processing device as an aspect of the present disclosure includes: a moving object detection unit that detects a moving object in a captured image; a stay area information holding unit that holds stay area information indicating the position of the moving object detected in the captured image; a stay frequency map generation unit that extracts the stay area information corresponding to a predetermined time period from time-series stay area information being held, and generates a stay frequency map by combining the extracted stay area information; and a specific region determination unit that determines a specific region having a high stay frequency regarding the moving object, in accordance with the generated stay frequency map. The present disclosure can be applied to watching systems installed in elderly care facilities and the like, for example.