Deep Convolutional Neural Network with Artificial Bee Colony Optimization For Plant Disease Detection and Classification

Plant diseases pose severe risks to agricultural production and global food security. For prompt intervention and mitigation, early and precise disease detection is crucial. Due to the development of trustworthy computer vision and machine learning technologies, automated plant disease detection systems are currently considered a potential solution to this issue. In this study, we provide a unique Deep Convolutional Neural Networks (DCNN)-based method for diagnosing plant illnesses. Our DCNN model automatically extracts pertinent features from plant photos, making it possible to classify healthy and unhealthy plants accurately. The dataset used in this study comprises a sizable collection of images with labels that show different plant species affected by various diseases. To improve its generalization and robustness to changing environmental circumstances and imaging artifacts, the proposed model is trained utilizing a mix of data augmentation via Haralick Texture. The suggested method uses a deep CNN to extract high-level information from photos of plants, allowing it to recognize subtle patterns suggestive of various diseases. The CNN architecture has been extensively crafted and adjusted to achieve the best accuracy and efficiency. The CNN's hyperparameters are optimized using the ABCO (Artificial Bee Colony Optimization) approach to increase classification accuracy further. ABCO mimics honeybee foraging to explore the hyperparameter space and find the best parameters efficiently. This improves the DCNN-ABCO performance. The findings demonstrate high accuracy, precision, and recall across various illness classes, demonstrating the model's dependability for practical applications and its capacity to identify previously unidentified disease patterns.