An Integrated Mobile Application for Automated Detection of Plant Leaf Diseases and Pest Infestations

This research study intends to develop a comprehensive mobile application poised to transform agricultural practices by enabling accurate identification of leaf, disease, and pest damage in plants. By analyzing the capabilities of modern smartphone technology, the application will leverage the device’s camera for users to capture images of affected plants. Anchored by an extensive database, featuring a diverse collection of crop and plant images, meticulously curated to showcase various symptoms induced by pests, our application ensures thorough analysis. Advanced algorithms for object detection, image classification, and pattern recognition will be employed to provide precise identification and analysis of user inputs. Additionally, pattern matching algorithms will enable the detection of exact matches within existing patterns. By providing farmers and agricultural professionals with a powerful tool for prompt and reliable identification of plant issues, our application aims to significantly enhance pest management and crop preservation efforts. Our proposed method includes the utilization of 11 features calculated with the Gray Level Co-occurrence Matrix (GLCM) for precise detection, augmenting the accuracy of our system. Additionally, we integrate hybrid Convolutional Autoencoder (CAE) and Convolutional Neural Network (CNN) models such as ResNetV2 and CNNIR-OWELM, enabling precise identification of plant damage and pest infestation, thereby enhancing the effectiveness of our solution in aiding farmers and agricultural professionals.