| Summary: | Motorcycle accidents have been rapidly growing throughout the years in many countries. Due to various social and economic factors, this type of vehicle is becoming increasingly popular. Over the past years, automated mechanisms to inspect traffic violations such as radars and surveillance cameras are being used ever more. This paper’s goals are the study and implementation of some methods for automatic detection of motorcycles on public roads. Traffic images captured by cameras were used. For feature extraction of images, the algorithms SURF, HAAR, HOG and LBP were used as descriptors. For image classification, Multilayer Perceptron, Support Vector Machines and Radial-Bases Function Networks were used as classifiers. Finally, the results are presented and discussed.
|
|---|