Please use this identifier to cite or link to this item:
Title: Microscopic Image Based Classification Algorithms for few Herbal Plants Identification using Machine Learning Approaches
Researcher: Fataniya B.M.
Guide(s): Zaveri Tanish, Acharya Sanjeev
Keywords: Engineering and Technology,Engineering,Engineering Electrical and Electronic
herbal plants
image processing
University: Nirma University
Completed Date: 09/04/2019
Abstract: Identification of herbal plant is of great interest in image processing and computer vision. In newlineliterature, many methods are presented to identify the herbal plant like leaf-based newlineidentification, chemical-based evaluation, physical evaluation and biological evaluation. newlineIdentification of herbal plant is more difficult and challenging when it is presented in powder newlineform. newlineThis thesis presents microscopic image-based classification of a few herbal plants from its newlinepowder using various machine learning approaches. In this work, the cell characteristics of newlinethe herbal plants are studied. The dataset of the powder microscopic images of the three newlineherbal plant is created in our laboratory using a Lawrence and Mayo microscope. newlineIt is found from the literature survey that various object can be uniquely represented by shape newlineand texture based features which are further used for the object classification. In this thesis, newlineshape and texture feature based novel methods for classification of herbal plants (Liquorice, newlineRhubarb and Dhatura) are proposed. The effectiveness of shape and texture feature methods newlineare evaluated using different classifiers for classification. Three shape and five texture newlinefeatures are computed from the microscopic image dataset of the herbal plants. The newlineeffectiveness of the shape and texture based feature set and their combinations are newlineinvestigated using Support Vector Machine (SVM), K-Nearest Neighbour (KNN) and newlineEnsemble classifier. The effect of Speeded-Up Robust Features (SURF) is also analyzed newlineusing a different kernel of the SVM classifier. From the experiments highest classification newlineaccuracy of 99.9% is achieved when all the shape features with a combination of Gabor newlinewavelet features are applied to quadratic SVM classifier. Finally, from the proposed newlinealgorithm, it is observed that the combination of selected shape and texture features work newlinebetter for the classification of powder microscopic image of the herbal plant of Liquorice, newlineRhubarb and Dhatura. newline
Appears in Departments:Institute of Technology

Files in This Item:
File Description SizeFormat 
01_title.pdfAttached File76.66 kBAdobe PDFView/Open
02_certificate.pdf282.4 kBAdobe PDFView/Open
06_content.pdf71.48 kBAdobe PDFView/Open
07_list of tables.pdf64.73 kBAdobe PDFView/Open
08_list of figures.pdf65.68 kBAdobe PDFView/Open
09_abbreviations.pdf42.82 kBAdobe PDFView/Open
10_chapter1.pdf144.58 kBAdobe PDFView/Open
11_chapter2.pdf1.64 MBAdobe PDFView/Open
12_chapter3.pdf1.02 MBAdobe PDFView/Open
13_chapter4.pdf1.39 MBAdobe PDFView/Open
14_chapter5.pdf1.29 MBAdobe PDFView/Open
15_conclusion and future scope.pdf117.03 kBAdobe PDFView/Open
16_publication.pdf81.46 kBAdobe PDFView/Open
17_references.pdf130.62 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: