Traffic Light Detection via Power Line Phase Synchronization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing traffic light detection systems struggle to accurately detect distant traffic lights due to their small image size, which makes it difficult to determine circularity and thus fails to provide reliable detection.
Innovation Solution
A traffic light detecting device that captures images of the surroundings and uses phase information of the electric power system to extract synchronized pixels with luminance variations matching the alternating current cycle, allowing for accurate detection of traffic lights regardless of size or shape.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If circularity-based detection is used to identify traffic lights, then detection accuracy is improved for nearby traffic lights, but detection capability deteriorates for distant traffic lights with small image sizes
Solution Approach 1:
The invention changes the detection parameter from geometric circularity to luminance variation phase. By analyzing whether the luminance variation of pixels matches the phase of the electric power system (e.g., 100Hz for 50Hz AC), the system can detect traffic lights regardless of their size in the image. This parameter transformation allows distant traffic lights with small image sizes to be detected effectively.
2Reliability
If image size is too small to determine circularity, then detection reliability deteriorates, but the traffic light may still be present and detectable through other means
Solution Approach 1:
The invention replaces the mechanical/geometric measurement approach (circularity calculation based on pixel coordinates) with a temporal/electrical approach (phase analysis of luminance variation). By substituting spatial analysis with temporal analysis synchronized to the electric power cycle, the system recovers detection capability without relying on shape information that is lost in small images.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables high-accuracy detection of distant traffic lights by utilizing phase information of the electric power system, even when the image size is too small to determine circularity, thus overcoming the limitations of existing systems.
Implementation Method 1
detects phase information of an electric power system used in an area around the vehicle including the traffic light from a cycle of luminance variation in the series of multiple images
Implementation Method 2
extracts from the images a synchronized pixel with a luminance which varies in synchronization with an alternating current cycle of electric power supplied to the traffic light by using phase information of the electric power system
Data Source
Figure 1
Figure 2
Figure 3(a)~3(b)
AI summary
A traffic light detecting device includes an image capturing unit (11) configured to repeatedly capture images in a travelling direction of a vehicle to obtain a series of multiple images (28) and a traffic light detecting unit (12) configured to detect traffic lights (32, 33) from the images (28). The traffic light detecting unit (12) detects phase information of an electric power system used in an area around the vehicle including the traffic lights from a cycle of a luminance variation in the series of multiple images and extracts from the images (28) a synchronized pixel with a luminance which varies in synchronization with an alternating current cycle of electric power supplied to the traffic light (32, 33) by using the phase information of the electric power system. The traffic light detecting device judges from the synchronized pixel whether or not the traffic light is present.