ROAD EXTRACTION FROM HIGH-RESOLUTION SATELLITE IMAGES: CHALLENGES, SOLUTIONS, AND FUTURE DIRECTIONS

A C T Road extraction is a widespread use of remote sensing images; however, extracting roads precisely and effectively from satellite images is challenging.It has several uses in various industries, including city traffic control, updating geographic information systems, city planning, computerized mapping, transportation, military use, and emergency vehicles.This study addresses a variety of models put out in the literature for extracting roads from satellite images and related approaches.Three types of road extraction approaches exist fully automated road extraction approaches, semiautomatic road extraction approaches, and data-based road extraction approaches.The continuous data update for geographic information systems in real-time applications makes automatic road extraction challenging.Since semi-automated systems eventually involve manual input, automatic road obtaining from satellite images is a vital area of study.On the other hand, the effectiveness of data-based approaches in applying computer vision and road extraction has led to their widespread acceptance by the research community.Based on the fundamental approach used to address the issue of road extraction, the approaches at each taxonomy class are further categorized.The study highlights the difficulties in road extraction research and potential prospects for the findings.