IoT-Cloud Empowered Aerial Scene Classification for Unmanned Aerial Vehicles

Recent trends in communication technologies and unmanned aerial vehicles (UAVs) find its application several areas such as healthcare, surveillance, transportation, etc. Besides, the integration of Internet things (IoT) with cloud computing environment offers benefits for UAV communication. At same time, scene classification is one major research UAV-enabled MEC systems. In imagery, efficient image representation crucial purpose classification. The existing techniques generate mid-level features limited capabilities that often end up producing average results. Therefore, current work introduces a new DL-enabled model presented enables UAVs to capture images which are then transmitted further processing. Next, Capsule Network (CapsNet)-based feature extraction technique applied derive set useful vectors from image. It important have an appropriate hyperparameter tuning strategy, since manual parameter DL tend produce configuration errors. order achieve this determine hyperparameters CapsNet model, Shuffled Shepherd Optimization (SSO) algorithm implemented. Finally, Backpropagation Neural (BPNN) class labels images. performance SSO-CapsNet was validated against two openly-accessible datasets namely, UC Merced (UCM) Land Use dataset WHU-RS dataset. proposed outperformed state-of-the-art methods achieved maximum accuracy 0.983, precision 0.985, recall 0.982, F-score 0.983.