In-Cabin Seat Belt Detection Using Bounding Box Overlap

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing seat belt wearing detection systems are prone to inaccuracies due to the use of weight sensors and buckle sensors, which can be deceived by tools like fake buckles or improper wearing, leading to ineffective enforcement of seat belt usage.

Innovation Solution

A seat belt wearing determination method and apparatus that utilizes a camera to capture vehicle interior images, determining bounding boxes for the occupant's upper body and seat belt, and calculates the overlapping degree between these boxes to accurately assess seat belt wearing.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Device complexity

If weight sensors and buckle sensors are used to detect seat belt wearing, then the detection process is simple, but the measurement precision is poor due to being deceived by fake buckles or improper wearing

Engineering Contradiction:
Improvedetection process complexityVSAvoidseat belt wearing detection accuracy
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The patent replaces the mechanical sensor-based detection system (weight sensors and buckle sensors) with an optical detection system using cameras. The camera captures images of the occupant and seat belt, and image processing algorithms analyze the visual information to determine seat belt wearing status. This substitution eliminates the vulnerability to fake buckles and improper wearing that plagues the mechanical sensor system, while maintaining relative simplicity through automated image analysis.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Extent of automation

If sensors are used to detect seat belt status, then the system can automatically monitor occupants, but false positives occur when occupants use fake buckles or wear seat belts improperly

Engineering Contradiction:
Improveautomatic monitoring capabilityVSAvoiddetection reliability
Core Design Contradiction:
Extent of automationVSReliability

Solution Approach 1:

The patent implements a feedback mechanism where the image processing system continuously analyzes captured images to verify seat belt wearing status. The system compares the detected seat belt position and appearance against the occupant's body contours and expected wearing patterns. When discrepancies are detected (such as fake buckles or improper wearing), the system provides feedback to correct the detection, thereby maintaining high reliability while preserving automatic monitoring capabilities.

Inventive Principle:
Principle #23Feedback

3Measurement precision

If video information analysis is used to determine seat belt wearing, then the accuracy of detection is improved, but the device complexity increases due to image processing requirements

Engineering Contradiction:
Improveseat belt wearing detection accuracyVSAvoidimage processing system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent applies segmentation by dividing the image processing task into distinct functional modules: occupant detection module, seat belt detection module, and wearing status determination module. Each module processes specific aspects of the image independently, which simplifies the overall system architecture while maintaining high accuracy. The occupant detection module identifies body contours, the seat belt detection module locates the seat belt, and the determination module compares the two to assess wearing status.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP4310799B1Seat belt wearing determination apparatus
Publication Date: 2026.01.28 HYUNDAI MOBIS CO LTD
  • EP4310799B1 patent drawingFigure 1
  • EP4310799B1 patent drawingFigure 2
  • EP4310799B1 patent drawingFigure 3

AI summary

Introduced is a seat belt wearing determination apparatus including a camera configured to capture an interior of a vehicle, a controller configured to determine a first bounding box and a second bounding box from an image including a seated occupant captured by the camera and determine whether the seated occupant is wearing a seat belt based on the overlapping degree of the first and second bounding boxes, wherein the first bounding box corresponds to the upper body region of the seated occupant, and the second bounding box includes the seat belt detected from the captured image by the controller.