Street Marking Color Detection with Red-Clear Camera Thresholding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing street marking detection systems face challenges in distinguishing between different colors, especially under varying lighting conditions, particularly with red-clear cameras that lack green and blue channel information, making it difficult to accurately identify yellow and white lane markings.
Innovation Solution
A method and image processing system that determine the color of street markings by comparing a color score to threshold values, using a combination of color information from pixels, even with limited color data from red-clear cameras, and adapting threshold values based on statistical distributions of color scores from multiple frames to compensate for illumination changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If traditional color balance methods are used with red-clear cameras, then the device complexity is reduced, but the measurement precision of color detection deteriorates due to lack of green and blue channel information
Solution Approach 1:
The patent changes the parameter representation from raw RGB color values to a transformed color space using luminance (Y) and color difference components (R-Y, Y-B). This transformation allows red-clear cameras to achieve accurate color detection by utilizing the luminance information from all pixels and color difference information from the red and clear pixels, effectively compensating for the missing green and blue channel data through mathematical parameter transformation.
Solution Approach 2:
The patent introduces an intermediary color space transformation as a mediator between the red-clear camera's limited color information and the required accurate color detection. By transforming the limited color data into a new representation that emphasizes luminance and color differences, the system bridges the gap between the camera's hardware limitations and the application's color detection requirements.
2Measurement precision
If conventional RGB cameras are used, then the color detection capability is improved, but the device complexity and cost increase compared to red-clear cameras
Solution Approach 1:
The patent extracts and utilizes only the essential color information needed for street marking detection from the red-clear camera's limited data. By focusing on luminance and color difference components rather than requiring full RGB information, the system achieves effective color detection with a simplified camera system, extracting maximum utility from the available red and clear pixel data.
3Reliability
If color balance adjustment is applied to compensate for lighting conditions, then the reliability of color detection under varying illumination is improved, but the processing time and computational complexity increase
Solution Approach 1:
The patent employs parameter transformation to luminance and color difference components that are inherently more robust to illumination variations. By representing colors in this transformed space, the system achieves reliable color detection under varying lighting conditions without requiring complex, time-consuming iterative color balance adjustments, as the transformation itself provides illumination invariance.
4Measurement precision
If full RGB color information is captured, then the color classification accuracy is improved, but the loss of information from clear pixels is increased when using red-clear cameras
Solution Approach 1:
The patent transforms the color information representation to maximize the utilization of available data from red-clear cameras. By changing from a direct RGB representation (which would leave clear pixels underutilized) to a luminance-color difference representation, the system extracts full value from the clear pixels' luminance information and combines it with color difference information from red pixels, achieving accurate color classification without information loss.
Data Source
Figure 1
Figure 2~3
Figure 4~5
AI summary
The present invention relates to a method and an image processing system for determining the color of a street marking, by: capturing an image of a street, detecting a street marking as a set of pixels provided by the image, wherein the pixels include at least two different pieces of color information, determining a color score for the street marking by comparing said at least two different pieces of color information, and determining the color of the street marking by comparing the color score to at least one threshold value.