On-Board Driving Diagnosis Using Target Recognition Updates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing driving diagnosis methods require significant image data transmission to a server outside the vehicle, burdening communication circuits.
Innovation Solution
An information processing device that acquires and processes travel information and target recognition data on-board, updating target recognition states and diagnosing events requiring attention, reducing the need for image data transmission.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If image data is transmitted from the vehicle to the server for driving diagnosis, then accurate driving behavior analysis is achieved, but communication circuit burden increases significantly
Solution Approach 1:
The patent extracts only the essential information needed for driving diagnosis (target recognition results and travel information) from the complete image data, transmitting only this extracted information to the server while leaving the full images at the vehicle. This resolves the contradiction by maintaining diagnosis accuracy through target state information while dramatically reducing communication data volume by excluding redundant pixel data.
Solution Approach 2:
The patent creates simplified copies of the essential diagnostic information (target presence/absence and travel parameters) rather than transmitting the original complete image data. These informational copies contain sufficient detail for accurate driving behavior analysis while occupying minimal communication bandwidth, thus resolving the contradiction between diagnosis precision and data transmission volume.
2Loss of information
If complete image data is transmitted for driving diagnosis, then comprehensive analysis is possible, but communication bandwidth is consumed
Solution Approach 1:
The system extracts only the diagnostically relevant information (target recognition outcomes and travel information) from the complete image data, transmitting this extracted subset to the server. This maintains information completeness for diagnosis purposes while minimizing communication energy consumption by transmitting only essential data elements.
Solution Approach 2:
The patent segments the complete image data into essential diagnostic components (target recognition results) and non-essential components (full pixel information). Only the essential segment is transmitted to the server, preserving diagnostic information completeness while reducing communication energy usage by eliminating transmission of redundant segments.
3Measurement precision
If target recognition information is updated using temporal section data, then recognition accuracy is improved, but processing complexity increases
Solution Approach 1:
The patent performs preliminary processing of target recognition information by updating recognition results with temporal section data before transmission. This preliminary action improves recognition accuracy by incorporating temporal context while managing processing complexity through efficient update operations that leverage existing recognition results rather than reprocessing complete images.
Solution Approach 2:
The system implements feedback by using previously recognized target information to update current recognition results within temporal sections. This feedback mechanism improves recognition accuracy by incorporating historical context while controlling processing complexity through iterative updates rather than complete reanalysis, transmitting only the updated essential information.
Data Source
AI summary
An information processing device includes: an acquisition unit that acquires travel information related to travel of a vehicle and information as to whether a target has been recognized from a captured image captured by a camera mounted on the vehicle; a preprocessing unit that updates, when information indicating that the target has been recognized is included in the information acquired during a certain section including a certain time, information indicating that the target is not recognized at the certain time with the information indicating that the target has been recognized; and a diagnosis unit that diagnoses that an event requiring attention has occurred at each time when there is the information indicating that the target has been recognized and a detection condition based on a state of the target and the travel information is satisfied.