Vehicle Driving-Action Information Provision With Contextual Scene Images
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies struggle to easily provide scene information shared among traffic participants, particularly in identifying and providing details of near-miss events, which are scattered and difficult to contextualize in time and location.
Innovation Solution
An information provision device and method that utilizes a detection system to identify desirable or undesirable driving actions, extracts relevant images, and provides content corresponding to these actions, including questions without requiring answers, using trained models to enhance accuracy and reduce process load.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If scene information is shared among traffic participants, then driving safety and awareness are improved, but information becomes scattered and difficult to identify and provide in context
Solution Approach 1:
The patent segments scene information into structured components including location data, time stamps, image data, and action types (desirable/undesirable). This segmentation allows scattered information to be organized into identifiable units that can be easily retrieved and presented in context, resolving the contradiction between information sharing and information accessibility.
Solution Approach 2:
The patent introduces an information provision device as an intermediary that collects, processes, and distributes scene information to traffic participants. This mediator organizes scattered information from multiple sources and delivers it in a structured, contextualized format, enabling safe driving awareness while maintaining information accessibility.
2Loss of information
If detailed scene information is provided to all traffic participants, then awareness of near-miss events is improved, but system complexity and information processing load increase
Solution Approach 1:
The patent extracts only the essential and relevant scene information components needed for safety awareness, such as location, time, image data, and action type classification. By extracting only necessary information rather than processing all possible data, the system reduces processing complexity while maintaining effective near-miss event information delivery.
Solution Approach 2:
The patent changes the parameter of information representation by categorizing actions into desirable and undesirable types, and by structuring data with specific parameters (location, time, image, action type). This parameterization simplifies information processing and reduces system complexity while preserving critical near-miss event details.
3Ease of operation
If scene information is provided without contextualization, then information delivery is simplified, but ability to identify specific actions and events in specific places and times is reduced
Solution Approach 1:
The patent applies preliminary action by pre-structuring scene information with contextual parameters (location, time, image data, action type) before delivery. This preliminary organization enables both easy information provision and precise event identification, as the context is already embedded in the structured data format ready for distribution.
Data Source
AI summary
The information provision device includes a detection part, a determination part, an extraction part, and a provision part. The determination part determines whether driving of the vehicle performed by a driver is a desirable action or an undesirable action with respect to a predetermined criterion on the basis of one or both of the behavior of the vehicle detected by the detection part and an image captured by an imaging part that images a surrounding situation of the vehicle. The extraction part extracts a first image in a first predetermined time before a first time at which the desirable action has been performed or a second image in a second predetermined time before a second time at 10 which the undesirable action has been performed out of images captured by the imaging part. The provision part provides content corresponding to a type of the action using the extracted image.