Abstract :

The Extraction Of Primary Urban Road Networks From High-resolution Satellite Imagery Plays A Vital Role In Urban Planning, Traffic Management, And Autonomous Navigation Systems. Conventional Approaches, Which Depend On Manual Mapping And Traditional Image Processing Techniques, Are Often Laborintensive, Time-consuming, And Susceptible To Inaccuracies. To Address These Limitations, This Paper Proposes A Machine Learning-based Framework For The Automated Extraction Of Urban Roads From Satellite Images. The Proposed Approach Utilizes Advanced Deep Learning Architectures, Including U-Net And DeepLabV3+, For Semantic Segmentation To Accurately Identify Road Structures. Preprocessing Techniques Such As Contrast Enhancement And Data Augmentation Are Applied To Improve The Robustness And Generalization Capability Of The Models. Furthermore, Post-processing Methods, Including Morphological Operations And Graph-based Refinements, Are Employed To Enhance The Continuity And Accuracy Of The Extracted Road Networks. The Performance Of The Proposed System Is Evaluated Using Benchmark Datasets, Demonstrating High Accuracy And Efficiency In Road Detection Tasks. Experimental Results Indicate That The Proposed Method Significantly Outperforms Traditional Techniques In Terms Of Precision And Reliability. The Generated Road Maps Can Be Effectively Integrated Into Geographic Information Systems (GIS), Supporting Real-world Applications Such As Smart City Development And Intelligent Transportation Systems.

Published:

16-4-2026

Issue:

Vol. 26 No. 4 (2026)

Page Nos:

2210 - 2218

Section:

Articles

License:

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

How to Cite