MangoLDB: A Dataset of Mango Leaves RGB, Binary and Gray-Scale Image
Hasna Hena1, Ahmed Al Marouf2, Rezwana Sultana3

1Most. Hasna Hena, B.Sc and M.Sc Degree with First Class Second Position in the Department of Information and Communication Engineering, from the University of Rajshahi, Bangladesh.
2Ahmed Al Marouf, Currently Pursuing M.Sc. Engineering, in Computer Science and Engineering (CSE) from Islamic University of Technology (IUT), Gazipur, Bangladesh.
3Rezwana Sultana, Has Completed Her M.Sc. in CSE and B. Sc. in CSE from University of Dhaka.

Manuscript received on 01 May 2019 | Revised Manuscript received on 15 May 2019 | Manuscript published on 30 May 2019 | PP: 1277-1281 | Volume-8 Issue-7, May 2019 | Retrieval Number: F4017048619/19©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: This paper presents the very first image dataset of mango leaves of different species which are originated in Bangladesh. This dataset contains the RGB, binary and grayscale; three versions of each image. Being the national tree of Bangladesh, mango is a sweet and juicy drupe having numerous species of trees. The scientific name for the commonly found mango species is Megnifera Indica. After the Jamdani Saree and Hilsha, different species of mango such as Khir sha, Langra, Aswina, Fazli, Haribhanga are the future geographical identification (GI) products of Bangladesh. Therefore, being highly demandable fruit, the identification of different species from its leaf images could be a challenging task. Agriculture specialists, farmers and general people may have difficulty to recognize samples just by observing the leaves. In this paper, we have formulated an image dataset of mango leaves of six different species namely, Aswina, Fazli, Gopalvog, Khirsha, Langra and Lokhna. After data retrieval, cleaning and processing, we have created an image dataset consisting of 7905 sample images. The images are collected using Smart phone having enough image pixel information for applying image processing tools. The dataset contains a minimal amount (1.5%) of blur and noise. The dataset could be considered as the basic ground-truth dataset for species recognition, disease recognition etc. from mango leaves in the area of computer vision.
Keyword: Computer Vision, Mango, Megnifera Indica, Mango Leaves Image Dataset, Species Recognition.
Scope of the Article: Image analysis and Processing.