Classification of Citrus (Rutaceae) by Using Image Processing

Mohd Tabri, Najwa Bari’ah (2019) Classification of Citrus (Rutaceae) by Using Image Processing. [Undergraduate Final Year Project Report] (Submitted)

Full text not available from this repository.

Abstract

Image processing has been increasingly used for agricultural applications for crop management especially in identification of crop status, quantity and quality. The aim of this study are to identify the classification of four selected Citrus species which are Citrus microcarpa (calamondin), Citrus aurantifolia (common lime), Citrus hystrix (kaffir lime) and Citrus maxima (pomelo). This research will be conducted by using digital image processing approach based on the morphological features of leaf with the combination of gray level co-occurrence matrix (GLCM), Prewitt and Canny algorithm and training classification models by using support vector machine (SVM). A machine learning algorithms, SVM have been used to build species identification models. The study present how to classify selected Citrus genus species with similar leaf shapes based on leaf images by using digital image vision machine classification. Though the developed system is not intended to replace human taxonomists, it may provide a rapid and easily accessible technique to identify plants with acceptable accuracy. The image pixels of the Citrus was classified using the difference in the leaf features of the plant species. SVM models achieved satisfactory results demonstrating its usefulness in identification and classification tasks by occupied 93% of the highest overall accuracy based on the combination of GLCM and Canny and algorithm features where C. maxima, C. aurantifolia and C. microcarpa obtained 100% accuracy while C. hystrix obtained 80% of accuracy. This study provided a rapid and easily accessible technique to identify plants which are beneficial by using digital image classification.

Item Type: Undergraduate Final Year Project Report
Faculty: Faculty of Earth Sciences
Depositing User: Users 9761 not found.
Date Deposited: 11 May 2021 05:18
Last Modified: 11 May 2021 05:18
URI: http://umkeprints.umk.edu.my/id/eprint/11059
Repository Statistic: View Download Statistic

Actions (login required)

View Item View Item