Volume 11, Issue 22 (11-2023)
ifej 2023, 11(22): 110-120 |
Back to browse issues page
Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
rezaei mutlagh A, parsakhoo A, najafi A, mohamadi J. (2023). Specification of Forest Road Surface Potholes using UAVs Image Evaluation. ifej. 11(22), 110-120. doi:10.61186/ifej.11.22.110
URL: http://ifej.sanru.ac.ir/article-1-480-en.html
URL: http://ifej.sanru.ac.ir/article-1-480-en.html
Department of Forestry, Faculty of Forest Resources, Gorgan University of Agricultural Sciences and Natural Resources, Gorgan
Abstract: (575 Views)
Introduction and Objective
The surface layer of the road suffers various damages with the passage of time due to traffic and atmospheric factors, and their quality decreases. Therefore, quick and accurate extraction of layer-surface anomaly is very important for effective monitoring of road health status. In order to improve the efficiency of surface layer inspection, today drones are used as a useful tool to obtain reliable information in the field of road surface layer.
Material and Methods
This research was conducted with the aim of revealing the top layer damage of forest roads using UAV images and image processing techniques on a 3.6 km long road in Dr. Bahramnia educational and research forest, Golestan province. The images obtained from the UAV were prepared using photogrammetric processing calculations and orthomosaic image and digital model of the ground height, then the resulting images were used to identify and check the top layer pits using three supervised learning algorithms: nearest neighbor, K-nearest neighbor And the supporting machine was examined and evaluated.
Results
The results showed that the orthomosaic image obtained from photogrammetric calculations has high accuracy. Also, checking the accuracy of the algorithms used to classify and identify potholes showed that these algorithms have a good ability to identify the damage of the road surface layer. Nearest Neighbor, K-Nearest Neighbor, and Support Vector Machine estimated the top layer failures with accuracy of 92.04, 94.31, and 96.59 percent, respectively.
Conclusion
The support vector machine algorithm as a supervised learning algorithm with 96.59 percent accuracy had the highest classification accuracy compared to the other two algorithms and as a suitable method for classifying and identifying failures in this The research was introduced. The resulting UAV images and supervised learning algorithms can be used to identify the abnormality of forest roads including potholes.
The surface layer of the road suffers various damages with the passage of time due to traffic and atmospheric factors, and their quality decreases. Therefore, quick and accurate extraction of layer-surface anomaly is very important for effective monitoring of road health status. In order to improve the efficiency of surface layer inspection, today drones are used as a useful tool to obtain reliable information in the field of road surface layer.
Material and Methods
This research was conducted with the aim of revealing the top layer damage of forest roads using UAV images and image processing techniques on a 3.6 km long road in Dr. Bahramnia educational and research forest, Golestan province. The images obtained from the UAV were prepared using photogrammetric processing calculations and orthomosaic image and digital model of the ground height, then the resulting images were used to identify and check the top layer pits using three supervised learning algorithms: nearest neighbor, K-nearest neighbor And the supporting machine was examined and evaluated.
Results
The results showed that the orthomosaic image obtained from photogrammetric calculations has high accuracy. Also, checking the accuracy of the algorithms used to classify and identify potholes showed that these algorithms have a good ability to identify the damage of the road surface layer. Nearest Neighbor, K-Nearest Neighbor, and Support Vector Machine estimated the top layer failures with accuracy of 92.04, 94.31, and 96.59 percent, respectively.
Conclusion
The support vector machine algorithm as a supervised learning algorithm with 96.59 percent accuracy had the highest classification accuracy compared to the other two algorithms and as a suitable method for classifying and identifying failures in this The research was introduced. The resulting UAV images and supervised learning algorithms can be used to identify the abnormality of forest roads including potholes.
Type of Study: Research |
Subject:
Special
Received: 2022/10/18 | Accepted: 2022/12/13 | Published: 2024/02/3
Received: 2022/10/18 | Accepted: 2022/12/13 | Published: 2024/02/3
Send email to the article author
| Rights and permissions | |
 |
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |
