Iranian Vehicle Images Dataset for Object Detection Algorithm

Providing a dataset with a suitable volume and high accuracy for training deep neural networks is considered to be one of the basic requirements in that a suitable dataset in terms of the number and quality of images and labeling accuracy can have a great impact on the output accuracy of the trained network. The dataset presented in this article contains ۳۰۰۰ images downloaded from online Iranian car sales companies, including Divar and Bama sites, which are manually labeled in three classes: car, truck, and bus. The labels are in the form of ۵۷۶۵ bounding boxes, which characterize the vehicles in the image with high accuracy, ultimately resulting in a unique dataset that is made available for public use.The YOLOv۸s algorithm, trained on this dataset, achieves an impressive final precision of ۹۱.۷% for validation images. The Mean Average Precision (mAP) at a ۵۰% threshold is recorded at ۹۲.۶%. This precision is considered suitable for city vehicle detection networks. Notably, when comparing the YOLOv۸s algorithm trained with this dataset to YOLOv۸s trained with the COCO dataset, there is a remarkable ۱۰% increase in mAP at ۵۰% and an approximately ۲۲% improvement in the mAP range of ۵۰% to ۹۵%.

Keywords:

Dataset , Object Detection , YoloV۸s , Vehicle Dataset , Deep Neural Network