Traffic Sign Recognition via Valid Area Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing traffic sign recognition technologies face challenges in accurately recognizing light emitting traffic signs with pulse width modulation, as the periodic on/off operation of their light sources can lead to erroneous or non-recognition due to partial occlusion in camera images.
Innovation Solution
A traffic sign recognizing apparatus and method that extracts a valid area from images by excluding non-valid areas, calculates similarity using valid area data, and updates comparison data to improve recognition performance, incorporating vehicle speed and direction information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a tracking algorithm is applied to recognize traffic signs using camera images, then recognition rate is improved, but erroneous recognition occurs when light sources in light emitting traffic signs operate with pulse width modulation
Solution Approach 1:
The patent segments the traffic sign image into multiple regions based on color information. By dividing the image into different color regions (e.g., red, yellow, green areas), the system can track each region separately and determine the overall traffic sign position and type more accurately, even when parts of the sign are obscured by periodic light source operation.
Solution Approach 2:
The patent applies different processing strategies to different regions of the traffic sign image. By identifying specific color regions and applying localized tracking and recognition algorithms to each region, the system maintains high recognition accuracy for light emitting traffic signs while handling the challenges posed by pulse width modulation light sources.
2Measurement precision
If the entire region of interest is used for similarity calculation, then comprehensive comparison is achieved, but computational complexity increases
Solution Approach 1:
The patent extracts only the relevant color regions from the entire region of interest for similarity calculation. Instead of processing the complete image, the system identifies and extracts specific color-based regions that contain traffic sign information, thereby reducing computational complexity while maintaining comprehensive comparison capability.
Solution Approach 2:
The patent applies partial action by focusing computational resources only on the necessary color regions rather than processing the entire image. This selective approach reduces the amount of data that needs to be processed for similarity calculation, decreasing computational complexity while still achieving accurate traffic sign recognition.
Data Source
AI summary
Disclosed are an apparatus for recognizing a traffic sign and an operating method thereof. The apparatus for recognizing a traffic sign according to an exemplary embodiment of the present invention includes: an image obtaining unit configured to obtain an image from a vehicle to a predetermined range; a region of interest designating unit configured to recognize a traffic sign within the image, and designate an area including the traffic sign as a region of interest; a valid area extracting unit configured to extract a valid area except for an area of a first color from the region of interest, and calculate valid area data; and a similarity calculating unit configured to calculate similarity of the traffic sign by using the valid area data.