Deep Convolution Neural Network Based solution for Detecting Plant Diseases
M. Sunil Kumar, D. Ganesh, Anil V. Turukmane, Umamaheswararao Batta, Kutubuddin Sayyad Liyakat Kazi, Sayyadliyakat, Kekane Arjun, Maruti, V Natarajan, Ms Anantha, M Macha Babitha, Kumar, Mohammed Brahimi, Kamel Boukhalfa, Abdelouahab Moussaoui, Jia Shijie, Jia Peiyi, Hu Siping, Alvaro Fuentes, Saiqa Khan, Meera Narvekar, Mosin Hasan, Bhavesh Tanawala, Krina Patel, Chad Dechant, Ms Chauhan, Deepika, Mark Everingham, Martn Abadi, A Babitha, M Kumar, M, Gampala, M Veerraju, Sunil, C Kumar, E Sushama, Raj Irudaya, Natarajan, B Sreedhar, M Be, M Kumar, M Natarajan, D Macha Babitha
Abstract
More over 58 percentage of the population works in agriculture, which accounts for more than 20 percent of India's total GDP. This research concentrates on plant diseases because they pose a serious danger to small-scale farmers' livelihoods and food production. In conventional farming, skilled employees are used to visually inspect each row in order to identify plant diseases. This labor-intensive, time-consuming task is inherently fault because it is carried out by humans. The purpose of this study is to create an automated detection model for the three most common maize plant diseases Frequent Rust, Cercospora Spot, and Northern Leaf Blight by combining image recognition and deep learning methods (Faster R-CNN+ResNet50) to evaluate real-time photos. The suggested system effectively identified three maize diseases with a 93.5% accuracy rate.