Open AccessPerformance Evaluation of VGG models in Detection of Wheat Rust
Author(s) -
Rajwinder Singh,
Rahul Rana,
Simranjit Singh
Publication year - 2018
Publication title -
asian journal of computer science and technology
Language(s) - English
Resource type - Journals
eISSN - 2583-7907
pISSN - 2249-0701
DOI - 10.51983/ajcst-2018.7.3.1892
Subject(s) - rust (programming language) , convolutional neural network , usability , computer science , agricultural engineering , crop , agriculture , stripe rust , artificial intelligence , agronomy , engineering , human–computer interaction , geography , plant disease resistance , biochemistry , chemistry , archaeology , gene , biology , programming language
The agricultural sector is the backbone of Indian economy and social development but due to lack of awareness towards crop management, a large number of crops get wasted each year. Automated Systems are required for this purpose. This paper tries to highlight the efficiency of two existing models of deep learning, VGG16 and VGG19 for proper detection of wheat rust disease in the infected wheat crop. These two models use convolutional neural networks for image classification and which can be used to design an intelligent system which can easily detect wheat rust in crop images. This paper basically presents the comparative analysis of the accuracy and efficiency along with usability to select the best model for systems that can be used for crop safety.