Scholarlink

Monitoring forests, grasslands, farms, empty areas, cities, and water is vital for studying global changes and urban planning. Information about Land cover and its variations is required to effectively manage natural resources, observe global environmental changes with their consequences. Detecting variations over time helps to assess the impact of human activities on the landscape, for disaster management, resource usage and sustainable development. The aim of the work involves developing a deep learning model for estimation of change detection in Land use and Land cover (LULC). The study area for this work is Nuzividu located in Andhra Pradesh, India. Experiments are done using freely available time series satellite data set of Resourcesat-1 taken from Bhuvan, NRSC Initially clustering is done using K-means clustering and Fuzzy C means (FCM) clustering algorithms. Also, Residual Neural Network (ResNet) architecture is used for the classification of different land cover types. Change detection is observed over a period of 10 years viz., 2009 to 2019.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Application of Deep Learning Algorithms for Land Use and Land Cover Change Detection

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Pallikonda Sarah Suhasini,
  • Talabattula Syam Vara Prasad,
  • Siriparapu Tatvik Shree Kartik,
  • Gochipathala Sandeep

摘要

Monitoring forests, grasslands, farms, empty areas, cities, and water is vital for studying global changes and urban planning. Information about Land cover and its variations is required to effectively manage natural resources, observe global environmental changes with their consequences. Detecting variations over time helps to assess the impact of human activities on the landscape, for disaster management, resource usage and sustainable development. The aim of the work involves developing a deep learning model for estimation of change detection in Land use and Land cover (LULC). The study area for this work is Nuzividu located in Andhra Pradesh, India. Experiments are done using freely available time series satellite data set of Resourcesat-1 taken from Bhuvan, NRSC Initially clustering is done using K-means clustering and Fuzzy C means (FCM) clustering algorithms. Also, Residual Neural Network (ResNet) architecture is used for the classification of different land cover types. Change detection is observed over a period of 10 years viz., 2009 to 2019.