Invasive Species Compendium

Detailed coverage of invasive species threatening livelihoods and the environment worldwide

Abstract

Android application development for identifying maize infested with fall armyworms with Tamil Nadu Agricultural University integrated proposed pest management (TNAU IPM) capsules.

Abstract

Several pests and diseases wreak havoc on maize crops worldwide. Novel and rapid methods for detecting pests and diseases in real-time will make monitoring them and designing effective management measures easier. In the recent past, maize has been imperilled by fall armyworms (Spodoptera frugiperda), which have caused substantial yield losses in maize. This study aimed to create an Android mobile application via DCNN (Deep Convolutional Neural Network)-based AI pest detection system for maize producers. Everyone benefits from the deployment of these CNN models on mobile phones, especially farmers and agricultural extension professionals because it makes them more accessible. Automatic diagnosis of plant pest infestations from captured images through computer vision and artificial intelligence research is feasible for technological advancements. Therefore, early detection of maize fall armyworm (FAW) infestation and providing relevant recommendations in maize could result in intensified maize crop yields. An Android mobile application was created to identify fall armyworm infection in maize and included the recommendations given by Tamil Nadu Agricultural University proposed Integrated Pest Management (TNAU IPM) capsules in the mobile app on as to how to deal with such a problem. Digital and novel technology was chosen to address these issues in maize. Deep convolutional neural networks (DCNNs) and transfer learning have recently moved into the realm of just-in-time crop pest infestation detection, following their successful use in a variety of fields. The algorithm accurately detects FAW (S. frugiperda) infected areas on maize with 98.47% training accuracy and 93.47% validation accuracy.