PyTorch YOLOv3 Object Detection for Vehicle Identification
Mohith Rajendra1, Susobhit Panigrahi2, Rashmi Joyappa Kand3, Shreya Sridhar4
1Mohith Rajendra, Department of Software Developer, Bangalore (Karnataka), India.
2Susobhit Panigrahi, Department of Software Developer, Bangalore (Karnataka), India.
3Rashmi Joyappa, Department of Software Developer, Bangalore (Karnataka), India.
4Shreya Shridhar, Senior Business Analyst, Bangalore (Karnataka), India.
Manuscript received on 13 February 2020 | Revised Manuscript received on 20 February 2020 | Manuscript Published on 28 February 2020 | PP: 30-32 | Volume-8 Issue-5S February 2020 | Retrieval Number: E10070285S20/2020©BEIESP | DOI: 10.35940/ijrte.E1007.0285S20
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Detecting real-world vehicle objects captured from car-mounted cameras requires manual labelling of video images. Previous vehicle object detection papers such as the winners of the 2018 AI City Challenge [1] used a training set of over 4,500 hand labelled images. In this paper, we attempt to automate this task by applying transfer learning to a YOLOv3 model trained on Imagenet and then re-trained on a set of stock car images and a small subset of hand labelled images taken from front-mounted dashboard camera videos. The mean Average Precision (mAP) of the validation set is used to determine the effectiveness of model vehicle classification. There is a significant variance issue between the validation and training set because the video images are taken in 1) various weather and lighting conditions and 2) the stock images have different image perspectives. The experimental results demonstrate that the YOLOv3 model can reach an overall 16.07% mAP after 60 epochs of training and can identify classes of vehicles that had few training examples in the dataset.
Keywords: Object Detection, Image Processing, Pytorch, Yolov3, R-CNN, Fast R-CNN, Faster R-CNN, Deep Learning, Map, IOU.
Scope of the Article: Image Processing and Pattern Recognition