METHOD FOR CALCULATING ROAD SURFACES DAMAGE ON BORDER SECTIONS USING COMPUTER VISION
Authors:
- Dmytro HykaDHDmytro Hyka
- Ольга ПавловаОПОльга Павлова
- Mykola LysyiMLMykola Lysyi
- Микола ЛисийМЛМикола Лисий
Open Access
- 1 January 2025
- journal article
- Published by Borys Grinchenko Kyiv University in Cybersecurity: Education, Science, Technique
- Vol. 3 (31) , 483-498
Abstract
Within the framework of the conducted research, an analysis and comparison of modern approaches and technological solutions aimed at assessing the condition of transport routes, road surfaces and roads of border areas was carried out. The aim of the work is to improve and optimize the processes related to monitoring, diagnostics and planning of restoration works to increase the efficiency of the relevant services. The international experience of determining the technical condition of roads, methods for calculating quality indicators was analyzed, and the use of innovative technologies that ensure increased accuracy of assessment and automation of data processing was also investigated. An own concept of a system for detecting road damage, calculating consumables and volumes of work at the preparatory stage was proposed. The system is based on the application of artificial intelligence and computer vision technologies using methods for analyzing images and video materials obtained using a video recorder. A specialized neural network training algorithm based on the modern YOLOv8 model has been implemented, which provides automatic detection of defects on the road surface and their quantitative assessment for further determination of the extent of damage, required resources and boundaries of deformation areas. A system for organizing and storing results in local or cloud databases has been proposed, which contributes to effective management of information flows. The developed approach has broad prospects for application in assessing the state of border zones of any state for the purpose of defense, analytical monitoring and quick decision-making. In addition, the improved methodology can be used to automate monitoring of the state of rock, dirt and gravel roads, which will significantly increase the efficiency and quality of repair work in rural, mountainous and border areas.Keywords
This publication has 12 references indexed in Scilit:
- Efficient Optimized YOLOv8 Model with Extended VisionSensors, 2024
- SOD-YOLOv8—Enhancing YOLOv8 for Small Object Detection in Aerial Imagery and Traffic ScenesSensors, 2024
- POT-YOLO: Real-Time Road Potholes Detection Using Edge Segmentation-Based Yolo V8 NetworkIEEE Sensors Journal, 2024
- YOLOv8-PD: an improved road damage detection algorithm based on YOLOv8n modelScientific Reports, 2024
- RDD-YOLO: Road Damage Detection Algorithm Based on Improved You Only Look Once Version 8Applied Sciences, 2024
- Pothole detection for autonomous vehicles using deep learning: a robust and efficient solutionFrontiers in Built Environment, 2024
- Pothole detection in the woods: a deep learning approach for forest road surface monitoring with dashcamsInternational Journal of Forest Engineering, 2023
- YOLOv8-Based Visual Detection of Road Hazards: Potholes, Sewer Covers, and Manholes2023
- Development of a new method for assessing condition of forest road surfaceJournal of the Croatian Association of Civil Engineers, 2020
- Deep learning in neural networks: An overviewNeural Networks, 2015