Vehicle Seat Occupancy Classification via Image Object Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing seat occupancy detection systems in vehicles lack reliability, particularly in distinguishing between occupants and objects, and fail to accurately determine seat states under occlusion or uncertainty, which can impact safety features like airbag deployment.
Innovation Solution
A computerized method and system that combines image processing and object detection, using a seat-based classifier and object-based modules to determine seat occupancy states by analyzing images, detecting objects, and assigning probabilities, which includes bounding box classification, body keypoint analysis, and occlusion consideration to enhance accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If image processing and object detection are used to determine seat occupancy states, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The system divides the image processing task into multiple specialized modules: object detection module, body keypoint analysis module, seat-based classification module, and occlusion handling module. Each module focuses on a specific aspect of occupancy detection, improving overall precision while organizing complexity into manageable segments.
Solution Approach 2:
The system uses multiple parameters and features for classification including body keypoint coordinates, bounding box dimensions, confidence scores, and occlusion values. By analyzing multiple parameters simultaneously, the system achieves higher measurement precision in determining seat occupancy states.
2Reliability
If multiple detection methods (bounding box classification, body keypoint analysis, occlusion consideration) are combined, then reliability is improved, but device complexity increases
Solution Approach 1:
The system merges multiple detection methods including bounding box classification, body keypoint analysis, and occlusion consideration into a unified seat occupancy determination system. The fusion module combines results from all methods to produce a final occupancy state, improving reliability through multi-method verification.
Solution Approach 2:
The system incorporates feedback mechanisms where detection results from one method inform and refine the results of other methods. For example, body keypoint analysis results feed into seat-based classification, and occlusion detection feedback adjusts the confidence scores of object detection, creating a self-correcting system that improves reliability.
3Measurement precision
If detailed object analysis and probability determination are performed, then measurement precision is improved, but loss of time increases
Solution Approach 1:
The system performs preliminary object detection and classification before detailed analysis. The object detection module first identifies potential objects and their bounding boxes, then only performs detailed body keypoint analysis and probability determination on detected objects, reducing overall processing time while maintaining precision for relevant cases.
Solution Approach 2:
The system applies full detailed analysis (body keypoints, probability determination) only when necessary for ambiguous cases or when object detection confidence is below a threshold. For clear, high-confidence detections, the system uses simplified classification, reducing processing time while maintaining high measurement precision for critical cases.
Data Source
Figure 1~2
Figure 3A~3B
Figure 3C~4
AI summary
A computerized method of determining seat occupancy of a vehicle is presented. The method comprises obtaining an image of a vehicle cabin showing at least one seat of the vehicle, determining objects in the image and assigning objects to the at least one seat, determining probabilities for seat occupancy states of the at least one seat, and determining a seat occupancy state of the at least one seat based on the assigned objects and the probabilities.