Nighttime Fog Detection via Color Temperature Thresholds
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current nighttime road fog detection systems struggle to accurately differentiate between fog and the surrounding environment in monochrome images captured by CCTV cameras at dawn, often incorrectly detecting the whitening phenomenon caused by vehicle headlights as fog, leading to degraded monitoring quality.
Innovation Solution
A nighttime road fog detection system that uses an image collecting unit to gather images from CCTV cameras, an image analysis unit to identify fog regions, a result display unit to extract and correct color information, and a mapping correction unit to remove masking regions caused by external light sources, employing a nighttime threshold value and environmental corrections to distinguish between fog and whitening phenomena.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional fog detection methods are used on nighttime CCTV images, then fog detection can be performed, but the system incorrectly detects whitening phenomena caused by headlights as fog, degrading monitoring quality
Solution Approach 1:
The patent applies parameter changes by analyzing color temperature characteristics of detected regions. The system calculates color temperature from RGB values and compares against threshold ranges to distinguish fog (higher color temperature) from headlight whitening (lower color temperature). This parameter-based differentiation resolves the contradiction by maintaining detection accuracy while eliminating false positives from external light sources.
Solution Approach 2:
The patent introduces color temperature as an intermediary parameter between the detected region and the classification decision. Rather than directly classifying regions as fog or non-fog, the system uses color temperature calculation as an intermediate step to mediate the detection process, enabling accurate differentiation between fog and headlight-induced whitening phenomena.
2Power
If monochrome images are used for nighttime monitoring, then power consumption is reduced and camera cost is lowered, but differentiation between fog and surrounding environment becomes difficult
Solution Approach 1:
The patent transforms the monochrome image data by calculating color temperature parameters from RGB values, effectively creating a new parameter space for analysis. This parameter transformation enables fog detection precision comparable to color cameras while maintaining the power and cost advantages of monochrome sensors.
3Adaptability or versatility
If daytime fog detection algorithms are applied to nighttime images, then fog detection functionality is achieved, but the system cannot distinguish fog from whitening phenomena caused by external light sources
Solution Approach 1:
The patent modifies the detection approach by introducing color temperature parameter analysis specifically adapted for nighttime conditions. The system calculates color temperature from RGB values and applies threshold-based classification to distinguish fog from headlight whitening, making the algorithm specifically adapted for nighttime monochrome image analysis while maintaining high precision.
Data Source
AI summary
Disclosed is a nighttime road fog detection system. More specifically, the present invention relates to a nighttime road fog detection system which rapidly and accurately detects the time of occurrence of nighttime fog that occurs at dawn before sunrise using open information of closed-circuit television (CCTV) cameras installed on a road and a detection method thereof. According to an embodiment of the present invention, in a mapping image obtained by masking a fog region identified by a fog detection program on an image signal, a region incorrectly determined due to a whitening phenomenon caused by an external light source generated at night is corrected, and thus information about the fog region is more accurately provided.