An Intelligent Image Processing and Convolutional Neural Network Framework for Plant Disease Identification

Early and accurate detection of plant diseases is essential for improving crop productivity and ensuring food security. This work presents an intelligent plant disease classification system using image processing techniques and deep learning based on Convolutional Neural Networks (CNNs). The proposed system automatically classifies plant diseases from leaf images using the publicly available PlantVillage dataset, which contains 54,305 images covering 38 disease and healthy classes across 14 plant species. The dataset includes a wide range of disease types such as bacterial, viral, fungal (oomycete), mite-induced diseases, as well as healthy leaf samples. Initially, input leaf images undergo preprocessing steps to enhance image quality and remove noise. Feature learning and classification are then performed using a CNN model, enabling effective extraction of discriminative visual patterns associated with different plant diseases. The trained model outputs the predicted disease class for each input image. Performance evaluation is carried out using standard metrics, including Accuracy, Precision, Recall, and F1-score, to assess the robustness and reliability of the proposed approach. Furthermore, the classification results are transmitted to the ThingSpeak IoT channel for real-time data storage and visualization, enabling remote monitoring and analysis of plant health conditions. The experimental results demonstrate that the proposed deep learning–based framework achieves high classification performance, making it suitable for practical agricultural disease monitoring applications.

Keywords— Plant disease classification, Convolutional neural network, PlantVillage dataset, Image processing, ThingSpeak IoT visualization