Pothole road detection and identification based on transfer learning

Yuying Mao; Wenzhe Su; Haotian Chen

doi:10.1117/12.3029139

6 May 2024 Pothole road detection and identification based on transfer learning

Yuying Mao, Wenzhe Su, Haotian Chen

Proceedings Volume 13107, Fourth International Conference on Sensors and Information Technology (ICSI 2024); 131073R (2024) https://doi.org/10.1117/12.3029139
Event: Fourth International Conference on Sensors and Information Technology (ICSI 2024), 2024, Xiamen, China

Abstract

Building a powerful country in transportation is a major strategic decision made based on national conditions, focusing on the overall situation and facing the future. The identification of potholes has important applications in the field of reducing accident rates as well as geological exploration and autonomous driving. In this paper, based on 301 road pictures, the training samples of the model are increased by data augmentation method. After normalization and standardization of the augmented training data, the random weighted sampling method is adopted as the training input of the model. In this paper, the VGG16 model is adopted as the classification model of road images, and the model parameters of lmageNet 1000 data set are taken as the feature extraction layer of the model in the way of transfer learning, and the model is trained on the modified classification layer. Two indexes, F1-score and Kappa coefficient, were used to evaluate the model. The model F1-score fluctuates around 0.5 and is higher than 0.4, and the Kappa coefficient is close to 0.6 in most cases. This shows that the model trained in this paper has high generalization ability and classification accuracy.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Yuying Mao, Wenzhe Su, and Haotian Chen "Pothole road detection and identification based on transfer learning", Proc. SPIE 13107, Fourth International Conference on Sensors and Information Technology (ICSI 2024), 131073R (6 May 2024); https://doi.org/10.1117/12.3029139

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
6 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Data modeling

Roads

Education and training

Machine learning

Deep learning

Performance modeling

RGB color model

Show All Keywords

Keywords/Phrases

Search In:

Publication Years