Transportation cost management is necessary for entrepreneurs in industry and business. One way to do this is to report the daily mileage numbers read by the employees of a company, but still encounter errors in human mileage reading, resulting in incorrect information received and difficulties in planning effective revenue management, as well as increased workload and complications for employees in checking mileage information. Therefore, the objective of this research was to create a machine-learning model for detecting and reading the mileage numbers of both digital and analog odometer displays of freight vehicles. The model has two processes; 1) To identify the position of the odometer in the speedometer image and remove unrelated backgrounds from the image, and 2) To read the mileage figure from the isolated odometer image. Both processes use object detection with the Faster-RCNN. To detect the position of the odometer , isolate and then read the mileage correctly, 220 test images were used and 187 of these images were correctly identified (85% accuracy). The accuracy of object detection on the analog odometer was 98.53% and the accuracy of object detection on the digital odometer was 97.37%. The accuracy of classification of the analog odometer was 83.82% with 85.53% accuracy of classification of the digital odometer. The results of the study demonstrated satisfactory performance that meets the requirements needed for real-life applications in the transportation and logistics industry.
Keywords: Object Detection, Mileage Reading, Faster-RCNN, Freight Vehicles
Bulan, O., Kozitsky, V., Ramesh, P., & Shreve, M. (2017). Segmentation- and Annotation-Free License Plate Recognition with Deep Localization and Failure Identification. IEEE Transactions on Intelligent Transportation Systems, 18(9), 2351-2363.
Chauhan, A., Rastogi, A., Gaur, A., Singh, A., & Gupta, S. (2020). Traffic Sign Detection Using Deep Learning. International Journal of Engineering Applied Sciences and Technology, 5, 355-358.
Dadras, E. A., & Huang, S. (2022). Artificial-Neural-Network-Based Surrogate Models for Structural Health Monitoring of Civil Structures: A Literature Review. Buildings, 12(12), 1-28.
Girshick, R.B., Donahue, J., Darrell, T., & Malik, J. (2014). Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation. 2014 IEEE Conference on Computer Vision and Pattern Recognition, 2014, 580-587.
Girshick, R. B. (2015). Fast R-CNN. Proceedings of the IEEE International Conference on Computer Vision (ICCV), December 2015 (pp. 1440–1448), Beijing.
LeCun, Y. (1989). Generalization and Network Design Strategies. In R. Pfeifer, Z. Schreter, Fogelman, F., & L. Steels, (Eds.), Connectionism in Perspective, (pp. 132-134). Zurich, Switzerland: Elsevier.
Liu, L., Ouyang, W., Wang, X., Fieguth, P., Chen, J., Liu, X., & Pietikäinen, M. (2020). Deep Learning for Generic Object Detection: A Survey. Retrieved from https://doi.org/10.1007/s11263-019-01247-4
Masood, S.Z., Shu, G., Dehghan, A., & Ortiz, E. (2017). License Plate Detection and Recognition Using Deeply Learned Convolutional Neural Networks. Winter Park, FL: Computer Vision Lab, Sighthound.
McCulloch, W. S., & Pitts, W. A. (1943). logical calculus of the ideas immanent in nervous activity. Bulletin of Mathematical Biophysics, 5, 115–133.
Nagaoka, Y., Miyazaki, T., Sugaya, Y., & Omachi, S. (2017). Text Detection by Faster R-CNN with Multiple Region Proposal Networks. Proceedings14th IAPR International Conference on Document Analysis and Recognition (ICDAR), 9-15 November 2017 (pp. 15-20). Kyoto, Japan.
Ravirathinam, P., & Patawari, A. (2020). Automatic License Plate Recognition for Indian Roads Using Faster-RCNN. Retrieved from https://www.researchgate.net/publication/338116738_Automatic_
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.