Vehicle Safety Object Detection Using Navigation Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting safety critical objects in vehicles often overwhelm drivers with unnecessary safety confirmation information, as they process all detected objects without focusing on the most critical regions, leading to inefficient object recognition and potential missed critical alerts.
Innovation Solution
The method combines navigational data with image capture to derive specific regions of interest based on road infrastructure, such as quadrants, to prioritize analysis and recognition of safety critical objects, and uses visual focus determination to ensure drivers attend to these objects, generating targeted warnings.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all detected objects are processed for safety criticality, then comprehensive safety monitoring is achieved, but the driver is overwhelmed with unnecessary safety confirmation information
Solution Approach 1:
The patent segments the detection area into multiple regions based on road infrastructure information (e.g., road edges, intersections, curves). Only objects in these segmented critical regions are processed for safety criticality, filtering out objects in non-critical areas. This resolves the contradiction by maintaining comprehensive safety monitoring in critical zones while reducing unnecessary information in non-critical zones.
Solution Approach 2:
The patent applies different processing quality levels to different spatial regions. High-quality processing (full safety criticality analysis) is applied only to objects in derived critical regions, while low-quality processing or no processing is applied to objects outside these regions. This local differentiation maintains reliability where needed while improving ease of operation by reducing overall information load.
2Reliability
If all detected objects are analyzed for safety criticality, then no critical objects are missed, but the object recognition process becomes inefficient and time-consuming
Solution Approach 1:
The patent performs preliminary action by deriving critical regions from road infrastructure information before object detection and analysis. This pre-processing step establishes which regions require detailed object analysis, allowing the system to skip analysis in non-critical regions and thereby improving productivity while maintaining reliability in critical areas.
Solution Approach 2:
The detection space is segmented into critical and non-critical regions based on road infrastructure. Object analysis is performed only in segmented critical regions, reducing the total number of objects requiring analysis and improving processing efficiency while ensuring no critical objects are missed in the derived regions.
3Reliability
If the analysis covers the entire periphery of the vehicle, then all potential safety critical objects are detected, but the system processes unnecessary regions reducing overall efficiency
Solution Approach 1:
The system performs preliminary action by pre-defining critical regions based on road infrastructure information before the object detection process. This allows the system to limit analysis to only those regions where safety critical objects are likely to occur, reducing processing time while maintaining detection coverage for all potentially critical objects.
Solution Approach 2:
The patent applies high-quality detailed analysis only to locally defined critical regions derived from road infrastructure, rather than uniformly analyzing the entire periphery. This local quality approach maintains reliability for safety-critical detection while reducing the time spent processing the entire field of view.
Data Source
Figure 1~2
AI summary
The present invention relates to a method and a device for detecting safety critical objects to a vehicle using a navigation device. Safety critical objects are potential obstacles to the vehicle or the traffic environment signs which are critical for driving of the vehicle. The vehicle is equipped with the device and the method of the current invention and is further equipped with a navigation device and an image capturing device. The method of the present invention makes use of navigational data available from the navigation device and an image of the periphery of the vehicle, which is captured from the image capturing device for detecting the safety critical objects to the vehicle. The method of the present invention further estimates visual attention of a driver of the vehicle towards said detected safety critical objects and generates an appropriate warning to the driver depending on the estimated visual attention of the driver.