Image leaf classification for plant diseases detection using grey wolf optimization technique
[ X ]
Tarih
2023
Yazarlar
Dergi Başlığı
Dergi ISSN
Cilt Başlığı
Yayıncı
Altınbaş Üniversitesi / Lisansüstü Eğitim Enstitüsü
Erişim Hakkı
info:eu-repo/semantics/closedAccess
Özet
Plant ailments have the potential to significantly affect agriculture and cause substantial
financial losses. They can affect crop yields and quality, leading to lower profits for farmers
and higher prices for consumers. In some cases, plant diseases can also lead to food shortages
and other economic and social consequences. Thus, it's critical to create efficient plans for
managing and avoiding plant diseases. The identification and diagnosis of plant illnesses
using learning techniques (ML and DL) can greatly reduce the harm and monetary losses
brought on by plant diseases.
In this thesis a method for plant disease detection is proposed using several approaches.
Three types of plant leaves are used in this thesis Peppers (two types), Potato (three types)
and Tomato (Nine types). image resizing, and data augmentation are used as a preprocessing. Three type of feature extraction Histogram of Gradient (HOG), Local Binary
Patterns (LBP) and Haralick feature are used. In order to choose the best features that
characterize, a Grey Wolf Optimization (GWO) is employed as a feature selection method.
Binary classification is carried by using Support Vector Machines (SVM) and K-Nearest
Neighbour (KNN) used multi classification in Machin learning while deep learning is usedThe proposed method had SVM algorithm for pepper plant achieved an accuracy of 93.14%
at 1800 sample size and 10 k-fold, while the KNN algorithm achieved an accuracy of 89.30%
for potato plant at 2202 sample size, 10 k-fold and accuracy of 95.18% for tomato plant at
12000 sample size ,10 k-fold. The CNN algorithm achieved an accuracy of 98.67%for
pepper plant, 99.85%for potato plant and 91.80%.
also for classification all three types of planet leaf.
Açıklama
Anahtar Kelimeler
Grey Wolf Optimization (GWO), K-Nearest Neighbour (KNN), Support Vector Machines (SVM), Local Binary Patterns (LBP), Histogram of Gradient (HOG), Convolutional Neural Network (CNN), Grey Level Co-Occurrence Matrix (GLCM)
Kaynak
WoS Q Değeri
Scopus Q Değeri
Cilt
Sayı
Künye
Jabbar, Amenah Nazar Jabbar. (2023). Image leaf classification for plant diseases detection using grey wolf optimization technique. (Yayınlanmamış yüksek lisans tezi). Altınbaş Üniversitesi, Lisansüstü Eğitim Enstitüsü, İstanbul.
