MangroveLens: A Smart Solution for Mangrove Species Identification Through MobileNetV3 Network Architecture and Biodiversity Monitoring

Abstract

Mangroves, as vital coastal ecosystems, play a significant role in maintaining biodiversity and providing essential ecosystem services. However, difficulties with species identification and biodiversity monitoring impede their conservation. In this work, we provide MangroveLens, a smart solution for biodiversity monitoring and mangrove species identification that makes use of MobileNetV3. We suggest utilizing mobile device images to recognize mangrove species using a deep learning-based method. MobileNetV3, renowned for its accuracy and efficiency, is used because it works well in contexts with limited resources. We exhibit the effectiveness of MangroveLens by conducting extensive experiments on real-world mangrove datasets, demonstrating its capacity to precisely identify different species of mangroves and monitor their biodiversity. Streamline biodiversity monitoring with an easy-to-use interface, real-time data documentation, and a cloud-based setup for viable analysis. Our findings suggest that MangroveLens provides a useful and effective tool for policymakers, academics, and conservationists to monitor and manage mangrove ecosystems, enabling focused conservation efforts and improving our knowledge of these important habitats.