Smart Farming Plant Disease Recognition Using Model-based Statistical Features

Authors: Sameer Raut, Sarang Kulkarni, Harshada Thite, Renuka Dond, Prof. S V Athawale

Country: India

Full-text Research PDF File:   View   |   Download

Abstract: The recent focus of our research is to detect and categorize the plant disease in the agricultural domain, by implementing image processing techniques. We aspire to propose an inventive set of statistical texture features for classification of plant diseases images of leaves. The input images are taken by various mobile cameras and any good resolution cameras. The Scale-invariant feature transform (SIFT) features used as texture feature and it is invariant to scaling, rotation, noise and illumination. However, the exact mathematical model of Scale-invariant feature transform texture descriptor is too complex and takes high computing time in training and classification. The model-based arithmetical features are intended from Scale-invariant feature transform descriptor to represent the features of an image in a small number of dimensions. The major focus of our proposed feature is to reduce the computational cost of mobile devices. In our research, 10-Fold cross-validation with SVM classifiers is practical to show that our experiment has no data bias and exclude hypothetically derived principles.

Keywords: Tomato plant diseases, deep learning, Classification, Leaf Images, Convolutional Neural Network Classifier, Predication.

Paper Id: 393

Published On: 2018-12-12

Published In: Volume 6, Issue 6, November-December 2018

Cite This: Smart Farming Plant Disease Recognition Using Model-based Statistical Features - Sameer Raut, Sarang Kulkarni, Harshada Thite, Renuka Dond, Prof. S V Athawale - IJIRMPS Volume 6, Issue 6, November-December 2018.

Share this