Drone Audio recognition based on Machine Learning Techniques

Assessing the Potential Impacts of Climate Change on Current Coastal Ecosystems—A Canadian Case Study

“…Historical satellite imagery between 2013 to 2022 was obtained from the United States Geological Survey (USGS) database, using the Landsat 8 Operational Land Imager (OLI)-Thermal Infrared Sensor (TIRS) Collection 2 Tier 1 in 30 m resolution. This dataset contains atmospherically corrected surface reflectance and land surface temperature, which comprise 5 visible and near-infrared (VNIR) bands and 2 shortwave infrared (SWIR) bands processed to orthorectified surface reflectance and one thermal infrared (TIR) band processed to orthorectified surface temperature [34]. Landsat 8 images are derived with the Land Surface Reflectance Code (LaSRC) algorithm, which utilizes the coastal aerosol band to perform aerosol inversion tests, employs auxiliary climate data from the Moderate Resolution Imaging Spectroradiometer (MODIS), and features a unique radiative transfer In recent years, climate-change-related natural hazards, such as temperature and precipitation variability, floods, storm surge, coastal erosion, sea level rise, and salinity intrusion, have significantly impacted PEI's natural ecosystems, economy, and society.…”

“…ML can be used to classify remotely sensed images efficiently and precisely because its capabilities include the ability to handle high-dimensional data and map classes with very complicated characteristics [27]. The application of ML in environmental studies has been explored widely in the past, including in studies of land use and land cover change [28,29], climate change [30][31][32], and drone-based applications for change detection [33,34]. Various ML algorithms that are popularly used to deal with complex problems in ecosystem research include the random forest (RF) algorithm, the support vector machine (SVM) algorithm, artificial neural networks (ANNs), decision trees (DTs), the K-nearest neighbors (KNN) algorithm, principal component analysis (PCA), and clustering algorithms (CAs), as well as other approaches as reviewed by Bansal et al [35] and Ray [36].…”