Open Access Open Access  Restricted Access Subscription Access

A Comparison of YOLO Models for License Plate Detection


Affiliations
1 Student Researcher (Computer Science), at American International School 1106, Jawaharlal Nehru Road, Tharamani, Chennai, Tamil Nadu 600113, India

   Subscribe/Renew Journal


License plate detection and recognition (LPDR) has become an increasingly popular field of study under image recognition. The ability to autonomously create a bounding box around a license plate using machine learning models for image recognition is an important component of intelligent transportation systems. The YOLO model has become a popular CNN model for image detection because of its speed and accuracy. This work focuses on the use of the YOLO model for object localization as applied to license plate detection. The research compares the accuracy and speed between the YOLOv3 and YOLOv4 models along with various configuration parameters and finds that the YOLOv4 model is most accurate while the YOLOv3 with reduced image resolution is the fastest in prediction time.

Keywords

Image recognition, LPDR, YOLO Model.
User
Subscription Login to verify subscription
Notifications
Font Size

  • S. Pal and S. K. Behera, “Traffic sign recognition for self driving vehicles using matlab and tensorflow,” Indian J. Comput. Sci., vol. 5, no. 6, Nov.-Dec. 2020, doi: 10.17010/ijcs/2020/v5/i6/157502
  • O. Russakovsky et al., "ImageNet large scale visual recognition challenge." 2015. [Online]. Available: arXiv: 1409.0575 [cs.CV]
  • P. Dwivedi. “YOLOv5 compared to faster RCNN. Who Wins?” Towards data science.com. (accessed : May 7, 2021). https://towardsdatascience.com/yolov5compared-to-faster-rcnn-who-wins-a771cd6c9fb4. .
  • J. Redmon, S. Divvala, R. Girshick, and A. Farhadi. "You only look once: Unified, real-time object detection," in 2016 IEEE Conf. Comput. Vision Pattern Recognition (CVPR), 2016, pp. 779-788. [Online]. A v a i l a b l e : h t t p s : / / w w w . c v foundation.org/openaccess/content_cvpr_2016/papers/ Redmon_You_Only_Look_CVPR_2016_paper.pdf
  • G. Karimi. “Introduction to Yolo algorithm for object detection, ” Section.io. https://www.section.io/engineeringeducation/introduction-to-yolo-algorithm-for-objectdetection/ (accessed May 7, 2021).
  • “Car License Plate Detection, Version 1”, Kaggle, Sep. 2020. https://www.kaggle.com/andrewmvd/carplatedetection/activity (accessed May 10, 2021).
  • ] D. Radečić. “How to detect license plates with Python and YOLO.” Towards Data Science.com. https://towardsdatascience.com/how-to-detect-licenseplateswith-python-and-yolo-8842aa6d25f7 (accessed May 6, 2021).
  • P. Canuma. “Image pre-processing.” Medium.com. https://prince-canuma.medium.com/image-preprocessingc1aec0be3edf (accessed May 9, 2021).
  • J. Solawetz. “YOLOv4 - Ten tactics to build a better model.” robloflow. https://blog.roboflow.com/yolov4tactics/ (accessed May 12, 2021).
  • A. Rosebrock. “Intersection over Union (IOU) for Object Detection.”PyImageSearch.com , https://www.pyimagesearch.com/2016/11/07/intersectio n-over-union-iou-for-object-detection/ (accessed May 18, 2021).
  • R. Rajkumar, A. Kaushal, and A. Saha, “Accelerating machine learning research using transfer learning,” Indian J. Comput. Sci., vol. 3, no. 2, Mar.-Apr. 2018, doi: 10.17010/ijcs/2018/v3/i2/123212
  • T. C. Arlen. “Understanding the MAP evaluation metric for object detection.” Medium.com. https://medium.com/@timothycarlen/understandingthemap-evaluation-metric-for-object-detectiona07fe6962cf3 (accessed March 7, 2021).
  • J. Hui, “Map (Mean Average Precision) for Object Detection.” Medium.com. [Online]. Available: https://jonathan-hui.medium.com/map-mean-averageprecisionfor-object-detection-45c121a31173 (accessed May 14, 2021).
  • J. Nelson, “When to use contrast as a preprocessing step,” roboflow.com. https://blog.roboflow.com/whentouse-contrast-as-a-preprocessing-step/ (accessed May 19, 2021).

Abstract Views: 226

PDF Views: 0




  • A Comparison of YOLO Models for License Plate Detection

Abstract Views: 226  |  PDF Views: 0

Authors

Arjan Chakravarthy
Student Researcher (Computer Science), at American International School 1106, Jawaharlal Nehru Road, Tharamani, Chennai, Tamil Nadu 600113, India

Abstract


License plate detection and recognition (LPDR) has become an increasingly popular field of study under image recognition. The ability to autonomously create a bounding box around a license plate using machine learning models for image recognition is an important component of intelligent transportation systems. The YOLO model has become a popular CNN model for image detection because of its speed and accuracy. This work focuses on the use of the YOLO model for object localization as applied to license plate detection. The research compares the accuracy and speed between the YOLOv3 and YOLOv4 models along with various configuration parameters and finds that the YOLOv4 model is most accurate while the YOLOv3 with reduced image resolution is the fastest in prediction time.

Keywords


Image recognition, LPDR, YOLO Model.

References





DOI: https://doi.org/10.17010/ijcs%2F2021%2Fv6%2Fi6%2F167643