Vehicle Occupant Physique Estimation Under Skeletal Point Occlusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing occupant physique detection devices fail to effectively detect and estimate the physique of an occupant in a vehicle, as they are often seated in a seat, due to the posture of the camera, which often results in occlusion of the skeletal points of the shoulder portions or the waist portions, which are occluded by the forearm portion of the occupant, the thigh portions of the occupant, the thigh of the vehicle, the thigh of the vehicle, the thigh of the occupant, the thigh of the occupant, or baggage, or the like, which are obstacles, which are obstacles, which are not detected by the camera.
Innovation Solution
The occupant physique detection device includes a camera that detects the physique of an occupant in a vehicle, using a camera to capture an image and detect skeletal points with no obstacles, calculating the area of a polygon, and estimating the physique of the occupant.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If all four skeletal points (both shoulder portions and both waist portions) are used for physique detection, then the accuracy of physique determination is improved, but the detection fails when any of these points are occluded by obstacles such as forearms, hands, thighs, or baggage
Solution Approach 1:
The patent applies partial action by requiring only three out of four skeletal points to be detected for physique estimation, rather than all four points. This allows the system to function with incomplete data when occlusion occurs, maintaining detection availability while still providing reasonable physique assessment
Solution Approach 2:
The patent changes the parameter from requiring exactly four skeletal points to accepting three or more points. This parameter modification enables the system to adapt to occluded conditions while maintaining physique detection capability
2Area of stationary object
If the camera is positioned to capture the occupant in a seated posture, then the occupant's physique can be imaged, but occlusion occurs where skeletal points are blocked by body parts or baggage
Solution Approach 1:
The system accepts partial detection of skeletal points (three out of four) rather than requiring complete detection, enabling physique estimation to proceed even when some points are occluded by body parts or baggage in seated posture
Solution Approach 2:
The patent segments the skeletal point detection into independent evaluable units, where detection of three points is sufficient to proceed with physique estimation, rather than treating all four points as a single required set
Data Source
AI summary
An occupant physique detection device includes: a captured image acquiring unit that acquires a captured image, from a camera that images the occupant of a vehicle; a skeletal point detecting unit that detects, from the captured image, three skeletal points that have no obstacles between the camera and the skeletal points and can be used to estimate the physique of the occupant among five predetermined skeletal points including the skeletal points of both shoulder portions and the skeletal points of both waist portions of the occupant, and outputs the position coordinates of each of the three skeletal points in the captured image; an area calculating unit that calculates the area of a polygon having each skeletal point as a vertex, using the position coordinates of each of the skeletal points; and a physique estimating unit that estimates the physique of the occupant from the area of the polygon.