Image Classification for Potato Plant Leaf Disease Detection using Deep Learning

• Published in: 2023 International Conference on Sustainable Computing and Smart Systems (ICSCSS) pp. 154 - 158
• Main Authors: Durai, S., Sujithra, T., Iqbal, M. Mohamed
• Format: Conference Proceeding
• Language: English
• Published: IEEE 14-06-2023
• Subjects: Convolution Neural Network; Deep Learning; image segmentation; Machine learning; Pooling; ReLU; ResNet50
• DOI: 10.1109/ICSCSS57650.2023.10169446

Identifying potato leaf diseases at an early stage is a difficult task due to the variability in crop species, crop disease symptoms, and environmental factors. To overcome this challenge, machine learning techniques have been developed. However, current models are limited to specific regions and cannot detect diseases in various crop species. This research proposes a multi-level deep learning model to recognize potato leaf diseases. The model uses a unique convolutional neural network to detect early blight and late blight potato infections from leaf images after extracting potato leaves from plant images using ResNet50 image segmentation. The model is trained and tested using a potato leaf disease dataset, achieving 99.75 percent accuracy. Furthermore, it outperforms state-of-the-art models in terms of accuracy and computational cost.