Event Recorder Image Filtering by Driver Location History
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing event data recorders and driving assistance systems face issues in accurately determining dangerous situations, as they may record non-relevant images due to temporary vehicle acceleration and lack determination of driver state in unfamiliar locations, leading to inefficient collection of abnormal situation images.
Innovation Solution
An information processing method that obtains location information of a moving body during event detection, determines if the location is frequently visited by the driver based on action history and related-person information, and controls image recording accordingly to efficiently collect images of abnormal situations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If image information is stored whenever acceleration exceeds a predetermined threshold, then dangerous situations can be detected, but false recordings occur when temporary vehicle acceleration happens without actual danger
Solution Approach 1:
The system uses location information and action history as feedback to evaluate whether recorded images actually represent dangerous situations. By comparing current event data with historical patterns at the same location, the system determines whether to retain or delete recorded images, thereby reducing false recordings while maintaining reliable dangerous situation detection.
Solution Approach 2:
Location information serves as an intermediary between the acceleration sensor and the image storage system. Instead of directly storing images based solely on acceleration thresholds, the system uses location data to mediate the decision-making process, cross-referencing with historical action patterns to determine whether the acceleration event warrants image recording.
2Device complexity
If determination of dangerous state is only made at locations where biometric information varied in the past, then processing is simplified, but dangerous situations in unfamiliar locations are missed
Solution Approach 1:
The system performs preliminary analysis of location characteristics using action history data before making dangerous state determinations. By pre-processing location information and comparing it with historical patterns, the system prepares decision-making frameworks in advance, enabling it to handle both familiar and unfamiliar locations efficiently without excessive processing complexity during real-time operation.
Solution Approach 2:
The system dynamically adjusts the parameters for dangerous state determination based on location familiarity. For unfamiliar locations, it uses broader criteria including environmental factors and deviation from normal driving patterns, while for familiar locations it relies on established biometric patterns, thereby maintaining reliability across different scenarios without excessive complexity.
3Quantity of substance
If photographic images are recorded at all event locations, then comprehensive data is collected, but storage efficiency decreases due to recording of non-abnormal situations
Solution Approach 1:
The system extracts and analyzes location information and action history patterns to identify which recorded images represent abnormal situations. By separating relevant from irrelevant recordings through this extraction process, the system maintains comprehensive data collection for analysis purposes while enabling efficient storage by identifying and prioritizing only the abnormal situations that require retention.
Solution Approach 2:
The system discards image data that is determined to represent normal situations based on location and action history analysis, while recovering and retaining images that represent abnormal situations. This selective discarding and recovering process maintains the completeness needed for comprehensive analysis while significantly improving storage efficiency by eliminating redundant normal situation recordings.
Data Source
AI summary
An information processing method obtains location information indicating a location of a moving body at a time of detection of occurrence of a predetermined event that triggers recording of a photographic image acquired by photography using a photographic apparatus installed on the moving body. Then, the method determines whether or not the location indicated by the location information is a location where the driver is estimated to have visited a plurality of times, based on at least one of action history information indicating action history of a driver of the moving body and related-person information regarding a related person of the driver. Finally, the method controls to record the photographic image, according to the determination of whether or not the location indicated by the location information is a location where the driver is estimated to have visited a plurality of times.