Convolutional Neural Network for Vehicle Detection in A Captured Image
Alia Abrougui, Mohamed Hayouni · 2022 International Wireless Communications and Mobile Computing (IWCMC) · 2022
Currently, vehicle detection is an important task in intelligent transportation systems. Indeed, it is very responsive in various intelligent applications such as traffic management and advanced driver assistance systems services. Computer vision can be influenced by several factors such as weather conditions, time of day, etc. Therefore, it is necessary to develop effective detection systems. In this work, we used YOLOv3 object detect algorithm with transfer learning to develop a vehicle detection model in a captured image. Firstly, we reset the size of the original anchors using k-means clustering. Then, we added preprocessing and data augmentation techniques to improve the accuracy and robustness of the algorithm. Finally, we added a script, for the total vehicle count. This model has been tested on two databases using the Google Colab Pro online tool. The proposed model has achieved good results performances.