Grading of Apples Using Multiple Features

Title

Grading of Apples Using Multiple Features

Creator

Sunil S.R.; Anita H.B.; Renupriya P.

Description

Apple is the most demanding food product that has the utmost importance when it comes to drupes. Food is the very basic necessity for our survival. Every new day brings a change, and the demand for a better quality is no greed. Quality food benefits the health of the living beings, and thus, it increases the economic growth of our country. There is a huge possibility that identifying the different varieties of apples is quite a tedious job for these traders and time consuming. Generally, identification is done manually by the very three basic senses: sight, hearing and smell. In the proposed work, an image processing technique is used to differentiate between the varieties of apples such that the manual process can be eliminated. Commercially available seven varieties of apple with various size, shape and color are considered to create database. Apples are purchased from different places across Karnataka, India to create the database. Various spatial and frequency domain based features are extracted from the images of apple. Naive Bayes, Random Forest and Multilayer perceptron (MLP) classifiers are used and got motivating results. An average accuracy of 78.47% is obtained using methods like Fourier Transform and Discrete Cosine Transform. 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Source

Lecture Notes in Networks and Systems, Vol-356, pp. 191-200.

Date

2022-01-01

Publisher

Springer Science and Business Media Deutschland GmbH

Subject

Apple; Discrete cosine transform; Drupes; Fast Fourier transform; Image processing; MATLAB; MLP; Naive Bayes; Random Forest; WEKA

Coverage

Sunil S.R., Department of Computer Science, CHRIST (Deemed To Be University), Karnataka, Bengaluru, India; Anita H.B., Department of Computer Science, CHRIST (Deemed To Be University), Karnataka, Bengaluru, India; Renupriya P., Department of Computer Science, CHRIST (Deemed To Be University), Karnataka, Bengaluru, India

Rights

Restricted Access

Relation

ISSN: 23673370; ISBN: 978-981167951-3

Format

Online

Language

English

Type

Conference paper

Identifier

https://doi.org/10.1007/978-981-16-7952-0_18

https://www.scopus.com/inward/record.uri?eid=2-s2.0-85130361613&doi=10.1007%2f978-981-16-7952-0_18&partnerID=40&md5=c0460b83a93181bca693b53ba6950a4e