Vehicle Interior Detection for Left-Behind Articles in Storage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for detecting left-behind articles in vehicles, such as comparing images of a vehicle's interior when the occupant is getting on and off, fail to detect articles stored in spaces like glove boxes, as these images do not show the storage state of articles.
Innovation Solution
A detection system comprising an image acquirer, action determiner, article manager, and left-behind article determiner, which uses machine learning to analyze images of vehicle interiors and determine the presence of left-behind articles by tracking the movement of articles into and out of storage spaces, managing their existence status, and identifying left-behind articles based on this information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If image comparison method is used to detect left-behind articles, then the detection process is simple, but articles in storage spaces cannot be detected
Solution Approach 1:
The detection system segments the vehicle interior into multiple regions including storage spaces (glove box, door pockets, console). Each region is analyzed separately by the action determiner to detect article placement or removal actions, enabling comprehensive detection of left-behind articles in previously inaccessible storage compartments.
Solution Approach 2:
The patent introduces an article manager as an intermediary component that maintains article information (presence, position, state) based on detected actions. This intermediary layer bridges the gap between raw image data and final left-behind article determination, enabling reliable detection in storage spaces through structured article state tracking.
2Device complexity
If conventional image comparison is used, then the system structure is simple, but detection precision is insufficient for storage spaces
Solution Approach 1:
The system performs preliminary actions by detecting and recording article placement actions (storing articles in storage spaces) and removal actions (taking articles from storage spaces) before the occupant exits the vehicle. The article manager maintains this article information throughout the occupation period, enabling accurate determination of left-behind articles even in enclosed storage spaces not visible in final images.
Solution Approach 2:
The action determiner continuously monitors the vehicle interior and provides feedback to the article manager about article actions detected. This feedback mechanism updates the article information in real-time, allowing the system to track article presence and position accurately, thereby improving detection precision for storage spaces through continuous monitoring rather than single-point comparison.
Data Source
AI summary
A detection device includes: an image acquirer that acquires an image of an interior of a vehicle, including a predetermined space; an action determiner that determines whether or not any one of a first action of placing an article or storing the article in the predetermined space and a second action of taking the article or taking out the article from the predetermined space has been performed and which of the first and second actions has been performed when it is determined that any of the first and second actions has been performed, on the basis of the acquired image; an article manager that manages an existence status of the article on the basis of the determination result of the action determiner; and a left-behind article determiner that determines whether or not an article left behind exists in the predetermined space on the basis of the existence status of the article.