mailto:uumlib@uum.edu.my 24x7 Service; AnyTime; AnyWhere

Improving the identification and classification of Malaysian medical leaf images using ensemble method

Sainin, Mohd Shamrie and Ahmad, Faudziah and Alfred, Rayner (2016) Improving the identification and classification of Malaysian medical leaf images using ensemble method. In: International Conference on ICT for Transformation 2016, 05-07 April 2016, Center for postgraduate UMS Sabah Malaysia.. (Unpublished)

[thumbnail of ICT4T2016 186 191.pdf] PDF
Restricted to Registered users only

Download (610kB) | Request a copy

Abstract

Malaysia has abundant natural resources especially plants which can be used for medicinal or herbal purposes.However, there is less research to preserve the knowledge of these resources to be utilized by the community in identifying useful medicinal plants using computing tools.This paper presents the implementation of digital opportunities for Malaysian medicinal plants via leaf image identification and classification. Of late, experts in traditional medicine and herbs have become few and the younger generation are mostly unknowledgeable about the medicinal and herbal properties of the plants.Therefore, this work is important in assisting the community (rural and urban) to identify and possibly share the knowledge of Malaysian medicinal plants with the future generation.The focus of this paper is to prepare the identification phase before the actual system is developed.Thus, the implementation of such a system is vital in order to enable the community to preserve these important resources.

Item Type: Conference or Workshop Item (Paper)
Additional Information: Organized by: Universiti Utara Malaysia
Uncontrolled Keywords: Malaysian medicinal plant, classification, ensemble learning
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions: School of Computing
Depositing User: Mrs. Norazmilah Yaakub
Date Deposited: 28 Feb 2018 01:57
Last Modified: 28 Feb 2018 01:57
URI: https://repo.uum.edu.my/id/eprint/23497

Actions (login required)

View Item View Item