Rice Leaf Disease Detection Using ResNet50

Although rice is a staple food worldwide, it is susceptible to various diseases that can significantly decrease the crop's quality and yield. A swift and precise diagnosis is essential for effective treatment. The earlier traditional approaches were time-consuming and needed to classify all diseases better. Novel technologies like computer vision and deep learning are being investigated to detect and identify crop diseases. In this paper, ResN et50 is employed, a type of Convolutional Neural Network used to classify Bacterial blight, Blast, Brown spot, and Tungro diseases. The Rice Leaf Disease Detection (RLDD) system is tested using a dataset of high-resolution images of healthy and diseased rice leaves of varying severity. RLDD is a cutting-edge solution that accurately identifies and categorizes rice leaf diseases. Hence, this method performs well when pre-processing methodologies are used to increase image quality and data augmentation to diversify datasets. The comprehensive evaluation of the proposed model considers several metrics, such as Fl-score, accuracy, precision, and recall, with outstanding results of nearly 100
^%
accuracy across four classes of rice diseases.