Emergency Vehicle Light Detection Using Color and Flash Frequency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing ADAS systems struggle to reliably distinguish emergency vehicles from standard vehicles due to variability in vehicle types, lighting conditions, and occlusions, leading to inaccurate detection and potential obstruction of emergency vehicle passages.
Innovation Solution
An image processing method using colorimetric segmentation, tracking, and frequency analysis to detect emergency vehicles by identifying flashing lights, incorporating post-segmentation filtering and classification to reduce false positives, and calculating an overall confidence index for reliable detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing ADAS systems use standard vehicle detection methods, then all vehicles can be detected, but priority vehicles cannot be reliably distinguished from standard vehicles
Solution Approach 1:
The patent applies colorimetric segmentation to detect the characteristic flashing colors of emergency vehicle lights. The system segments images based on color information to identify red, yellow, blue and other colored flashing lights that distinguish priority vehicles from standard vehicles, resolving the contradiction between general detection and specific differentiation.
Solution Approach 2:
The patent utilizes frequency analysis to detect the periodic flashing pattern of emergency lights. By analyzing the temporal frequency of light intensity variations, the system can reliably identify priority vehicles with flashing beacons, solving the problem of distinguishing vehicle types while maintaining broad detection capability.
2Reliability
If the system detects all luminous areas, then flashing lights can be identified, but false positives from other lights increase
Solution Approach 1:
The patent changes detection parameters by combining colorimetric segmentation with frequency analysis. Instead of detecting all luminous areas, the system specifically identifies lights that exhibit both characteristic colors and periodic flashing patterns, significantly reducing false positives from stationary lights or non-flashing sources while maintaining high reliability.
Solution Approach 2:
The system employs a confidence index calculation that integrates multiple detection criteria (color segmentation results, frequency analysis outcomes, and tracking consistency). This feedback mechanism allows the system to filter detections and reduce false positives by requiring multiple criteria to be satisfied simultaneously.
3Measurement precision
If the system uses multiple detection criteria, then detection accuracy improves, but processing time increases
Solution Approach 1:
The patent implements preliminary colorimetric segmentation to pre-identify candidate luminous areas before performing more computationally intensive frequency analysis. This preliminary filtering reduces the number of candidates requiring full multi-criteria evaluation, thereby maintaining high detection precision while reducing overall processing time.
Solution Approach 2:
The detection process is segmented into distinct stages: colorimetric segmentation to identify candidate areas, frequency analysis to verify flashing patterns, and confidence index calculation to confirm detections. This segmentation allows the system to apply different processing levels to different candidates, optimizing the balance between precision and processing time.
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
The invention relates to a method for processing a video stream of images captured by a colour camera (2) and used by a computer (3) on board a motor vehicle (1) in order to detect a priority vehicle (4), the method comprising the following steps: - a step (100) of acquiring an image sequence; for each image of the image sequence: - a step (200) of colour segmentation by thresholding allowing the detection of coloured light areas; - a step (300) of tracking each segmented light area; - a step (400) of colour classification of each segmented light area; - a step (500) of frequency analysis of each segmented light area allowing a flashing nature of the area to be determined; - a step (700) of calculating an overall reliability index of each image of the sequence of images, allowing a light area to be identified as a rotating light.