Tomato Leaf Disease Classification using Multiple Feature Extraction Techniques

Agriculture, with its allied sectors, is the largest source of livelihoods in India. Diseases in plants cause a substantial decrease in quality as well as quantity of crops or agricultural products. Detection of these diseases is the solution to prevent losses in the harvest and amount of agricultur...

Full description

Saved in:
Bibliographic Details
Published in:Wireless personal communications 2020-11, Vol.115 (1), p.633-651
Main Authors: Basavaiah, Jagadeesh, Arlene Anthony, Audre
Format: Article
Language:English
Subjects:
Accuracy
Algorithms
Classification
Classifiers
Communications Engineering
Computer Communication Networks
Computing time
Decision trees
Disease
Engineering
Feature extraction
Histograms
Networks
Plant diseases
Signal,Image and Speech Processing
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Agriculture, with its allied sectors, is the largest source of livelihoods in India. Diseases in plants cause a substantial decrease in quality as well as quantity of crops or agricultural products. Detection of these diseases is the solution to prevent losses in the harvest and amount of agricultural products. The main objective of the proposed method is to develop a technique to identify leaf disease in tomato plant by improving the classification accuracy and reducing computational time. The novelty of the work is fusion of multiple features in order to improve classification accuracy. Color histograms, Hu Moments, Haralick and Local Binary Pattern features are used for training and testing purpose. Rrandom forest and decision tree classification algorithms are uses for leaf disease classification. Based on the experiments conducted, it showed that the random forest classifier is more accurate than decision tree classifier. The classification accuracy is 90% for decision tree classifier and 94% for random forest classifier respectively.
ISSN:0929-6212
1572-834X
DOI:10.1007/s11277-020-07590-x