Automatic Diseases Detection and Classification in Maize Crop using Convolution Neural Network
J Huerta-Espino, R Singh, S Germn, B Mccallum, R Park, W Chen, S Bhardwaj, H Goyeau, S Weizheng, W Yachun, C Zhanliang, W Hongda, T Rumpf, A.-K Mahlein, U Steiner, E.-C Oerke, H.-W Dehne, L Plmer, A Kamilaris, F Prenafeta-Bold, K Ferentinos, A Kulkarni, A Patil, S Bashir, R Kaundal, A Kapoor, G Raghava, S Mohanty, D Hughes, M Salath, T Rumpf, A.-K Mahlein, U Steiner, E.-C Oerke, H.-W Dehne, L Plmer
Abstract
Maize is a major crop in Pakistan and it plays an important role in the economy of the country. However different types of plant diseases could affect both the quality and quantity of maize production. To cope with such issues, the majority of the farmers still depend on traditional methods, which are expensive, time-consuming, laborious, and not very effective. To address the issues, we proposed a Convolution Neural Network (CNN) based solution for the detection and classification of different types of maize diseases. We used a publicly available free dataset of 4000 images. The images were classified into four categories. The first three categories represent the Common rust, Cercospora leaf spot grey, and Northern leaf blightand diseases, while the last category represents normal leaves. To test, implement, and evaluate the performance of our proposed method, we used a MATLAB simulation environment. We also compared our results with two other solutions, available in the literature. Our solution achieved 96.53 % accuracy. From the results, we concluded that the proposed method could be used for the automatic detection and classification of different types of maize diseases.