Traffic Light Detection Using Pixel Response Values
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional traffic light detection methods face inefficiencies and reliability issues due to high computation loads from verifying false positives, such as trees, vehicle taillights, and logos, which are similar in color to traffic lights, leading to a need for improved filtering techniques.
Innovation Solution
A method involving pixel response value calculation using R, G, and B values to filter out non-traffic light pixels, followed by geometric analysis and verification using a traffic light classifier to reduce false positives and enhance detection efficiency and reliability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If color segmentation and template matching are used for traffic light detection, then detection capability is improved, but computation load increases due to verifying false positives
Solution Approach 1:
The patent applies preliminary action by performing pixel response value calculation and candidate pixel identification before full verification. The method calculates pixel response values using R, G, and B values to identify candidate pixels that are likely to belong to traffic lights, thereby filtering out non-traffic light pixels in advance and reducing the number of candidates requiring full verification
Solution Approach 2:
The patent changes parameters by introducing pixel response values as a new parameter for filtering. The pixel response value is calculated using a specific formula involving R, G, and B components with predetermined coefficients, creating a parameter that effectively distinguishes traffic light pixels from other objects based on their color characteristics
2Measurement precision
If conventional detection methods verify all color-similar objects, then detection accuracy is maintained, but false positives increase from objects like trees, vehicle taillights, and logos
Solution Approach 1:
The patent applies local quality by using different filtering criteria for different regions. The pixel response value calculation uses locally adapted coefficients that are optimized for traffic light color characteristics, allowing the system to distinguish traffic lights from other objects with similar colors by exploiting local color quality differences
Solution Approach 2:
The patent introduces pixel response values as an intermediary parameter between raw color data and final detection results. This intermediary calculation layer transforms R, G, and B values into a filtered representation that highlights traffic light characteristics while suppressing other objects, reducing false positives before final verification
3Quantity of substance
If pixel response value calculation with R, G, and B values is used to filter pixels, then non-traffic light segments are reduced, but computation load for pixel processing increases
Solution Approach 1:
The patent applies partial action by performing pixel response value calculation on all pixels but only proceeding with full verification for identified candidate pixels. This partial processing approach calculates the filtering parameter for all pixels to reduce the candidate set, then applies more intensive processing only to the reduced candidate population
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
A method and a system for traffic light detection are provided. The method may include: obtaining a color image; calculating pixel response values for pixels of the color image, respectively, where each of the pixel response values may be calculated using R, G, and B values of a corresponding pixel directly, such that pixel response values of red traffic light pixels are substantially distributed on a first side of a predetermined range and pixel response values of green traffic light pixels are substantially distributed on a second side of the predetermined range which is opposite to the first side; identifying pixels whose pixel response values are distributed on the first side or the second side as candidate pixels; identifying candidate blobs based on the candidate pixels; and verifying whether the candidate blobs are traffic lights. Efficiency and reliability may be improved.