Traffic light arrow shape recognition using HOG descriptors and SVM classifier
Zamani Md Sani1, Mohd Shahrul Hakimi Isa2, Hadhrami Abd Ghani3, Rosli Besar4.
Autonomous intelligent vehicles technologies research have expanded and more effort have been made to implement a better safety features in a transportation sector. The lack of visibility of the drivers to perceive the signal from the traffic light is one of the risks for safety.Hence, by introducing the detection and recognition of the traffic light arrow shape using image processing to deliver a clear signal to a drivers and eliminate their difficulties to perceive either the round or arrow shape of the traffic light, this risk can be reduced. The current research has focuses on the condition of the traffic light and not indicating the direction of the arrow. In this paper, an algorithm for traffic light recognition for arrow symbols at daytime has been proposed using digital image processing and machine learning. In the machine learning technique, the Support Vector Machine (SVM) classifier has been used for the learning process and conduct a classification process as well. The HOG descriptor was extracted from the feature extractionprocess for the purpose of the training and classification process. As a result, this algorithm has achieved 98.52% accuracy, 1.48% error, 98.52% precision and 98.68% of recall through the testing process.
Affiliation:
- Universiti Teknikal Malaysia Melaka, Malaysia
- Universiti Teknikal Malaysia Melaka, Malaysia
- Universiti Malaysia Kelantan, Malaysia
- Universiti Multimedia (MMU), Malaysia
Download this article (This article has been downloaded 50 time(s))