Disease Detection in Potato Plants Using Deep Learning Applied on Leaf Image

Abstract

The paper focuses on the detection of diseases in potato plants using deep learning applied to leaf images. It discusses the challenges faced by the agriculture sector in India and emphasizes the need for innovative solutions. The paper highlights the importance of potato cultivation in India, including its adaptability to various climates and its role as a crucial food source. Furthermore, the paper explores the use of convolutional neural networks (CNNs) for disease detection in potato plants. It describes the process of data collection, including the acquisition of a dataset containing both healthy and diseased potato leaf images. The paper discusses different methods of data collection, such as obtaining readymade data from third-party vendors or using a team of annotators to collect and classify images manually. The paper also explains the image acquisition process and the application of CNNs for feature extraction and classification. It discusses the use of convolution and pooling layers, as well as the activation function ReLU, in the CNN architecture. The paper presents the training and testing accuracy results of the proposed CNN model, demonstrating its effectiveness in accurately classifying healthy and diseased potato leaves. In conclusion, the paper highlights the potential of deep learning techniques, specifically CNNs, for disease detection in potato plants. It suggests that these techniques can contribute to improving the efficiency and productivity of the agricultural industry. The findings of the paper provide valuable insights for researchers and practitioners in the field of agricultural technology.