An efficient automatic traffic sign detection and recognition method for smartphones
| Autor: | Chun-Fei Hsu, Chi-Yi Tsai, Po-Cheng Shih |
|---|---|
| Rok vydání: | 2017 |
| Předmět: |
Maximally stable extremal regions
business.industry Computer science Feature vector Feature extraction 020206 networking & telecommunications Pattern recognition 02 engineering and technology Support vector machine Statistical classification Histogram 0202 electrical engineering electronic engineering information engineering Traffic sign recognition 020201 artificial intelligence & image processing Artificial intelligence business |
| Zdroj: | CISP-BMEI |
| Popis: | The usage of automatic driving assistance systems (ADAS) has become more and more popular in recent years. In the design of ADAS, traffic sign detection (TSD) and traffic sign recognition (TSR) are two important functions and have been widely studied in the literature. This paper addresses the design of a vision-based TSD and recognition (TSDR) system, which is computationally efficient and can run on a common smartphone with real time performance. To achieve this, a novel TSD algorithm is proposed based on the Maximally Stable Extremal Regions (MSER) algorithm to accurately detect all traffic sign candidates in real time. Then, the feature vector of each candidate region is extracted via the Histogram of Oriented Gradient (HOG) algorithm. To recognize the traffic sign, the proposed TSR algorithm is designed by combining a Linear Support Vector Machine (LSVM) classifier with a voting process to improve the traffic sign recognition rate in real-world environments. The proposed TSDR system had been implemented on iOS embedded platform and can operate at an average speed of 30 fps for processing 640 × 480 video streams. Moreover, experimental results show that the proposed system achieves about 96% precision rate for recognition on the German Traffic Sign Recognition Benchmark (GTSRB). Therefore, the proposed TSDR algorithm has a potential to be used in realistic products. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|