Please use this identifier to cite or link to this item: http://hdl.handle.net/10603/439955
Title: Hybrid Features Based Plant Leaf Disease Detection And Classification With BPSO Based Feature Selection And Machine Learning Techniques
Researcher: ASHUTOSH KUMAR SINGH
Guide(s): BHARTI CHOURASIA
Keywords: Engineering
Engineering and Technology
Engineering Electrical and Electronic
University: Sarvepalli Radhakrishnan University
Completed Date: 2022
Abstract: ABSTRACT: This research work is divided into three phases. The first phase intended to detect three newlinerice disease namely Brown-spot, Bacterial Leaf blight and Leaf smut using machine newlinelearning techniques with image processing. The color moments are extracted for color newlinefeatures while the Gabor Wavelet and Harris Corner methods are used for texture newlinefeatures extraction of PlantVillage Dataset images for rice plant leaf disease detection. newlineThe binary particle swarm optimization (BPSO) is then applied for the feature selection newlinefrom the extracted features. Finally various classifiers are used for the classification of newlineextracted features. newlineThe second phase seeks to show the technological relevance of the use of machine newlinelearning in the identification of diseases in tomato, potato and rice plants. An automatic newlineplant leaf disease detection framework is developed using K-means segmentation, color newlineand texture features. The color moments are extracted for color features while the newlineGLCM, HoG, and LBP are used for texture features extraction of PlantVillage Dataset newlineimages for leaf disease detection. Finally Random Forest Classifier and Support Vector newlineMachine are used for the classification of extracted feature to obtain the simulation newlineresults. newlineThe third phase follows two methodologies and their simulation outcomes are newlinecompared for performance evaluation. In the first part, data augmentation is performed newlineon the PlantVillage dataset images (for apple, corn, potato, tomato and rice plants) and their deep features are extracted using Convolutional Neural Network (CNN). These newlinefeatures are classified by Bayesian Optimized Support Vector machine classifier and newlineresult is attained in terms of precision, sensitivity, f-score and accuracy. The second newlinepart of methodology starts with the pre-processing of dataset images and their color and newlinetexture features are extracted by color moments, GLCM and Histogram of Oriented newlineGradient (HoG). Here, the three types of features, i.e. color, texture and deep features newlineare combined to form hybrid features. Th
Pagination: 
URI: http://hdl.handle.net/10603/439955
Appears in Departments:ELECTRONICS AND COMMUNICATION ENGINEERING

Files in This Item:
File Description SizeFormat 
02_prelim pages.pdfAttached File2.56 MBAdobe PDFView/Open
03_content.pdf354.48 kBAdobe PDFView/Open
04_abstract.pdf105.47 kBAdobe PDFView/Open
05_chapter 1.pdf342.55 kBAdobe PDFView/Open
06_chapter 2.pdf1.13 MBAdobe PDFView/Open
07_chapter 3.pdf1.08 MBAdobe PDFView/Open
08_chapter 4.pdf1.69 MBAdobe PDFView/Open
09_chapter 5.pdf2.11 MBAdobe PDFView/Open
10_chapter 6.pdf218.57 kBAdobe PDFView/Open
11_annexures.pdf6.96 MBAdobe PDFView/Open
80_recommendation.pdf433.2 kBAdobe PDFView/Open
Show full item record


Items in Shodhganga are licensed under Creative Commons Licence Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0).

Altmetric Badge: