Fruit classification using Viola-Jones detectors

In agriculture, the use of machine vision technology has been increasing throughout the years. The machine vision technology is highly essential in the process of fruit harvesting in order to increase the production rate and decrease the cost incurred by hiring human workers. Fruit detection and cla...

Full description

Bibliographic Details
Main Authors: Saeed S, Almehmadi Tarig, Moussa Hassan, Tariq Zeyad, Htike@Muhammad Yusof, Zaw Zaw, Nyein Naing, Wai Yan, Khan, Sheroz
Format: Conference or Workshop Item
Language:English
Published: 2014
Subjects:
Online Access:http://irep.iium.edu.my/38050/
http://irep.iium.edu.my/38050/
http://irep.iium.edu.my/38050/1/Fruit_classification.pdf
Description
Summary:In agriculture, the use of machine vision technology has been increasing throughout the years. The machine vision technology is highly essential in the process of fruit harvesting in order to increase the production rate and decrease the cost incurred by hiring human workers. Fruit detection and classification have been a challenging issue for machine vision researchers. Several fruit detection techniques using color-based and shape-based features have been developed by researchers. However, because the same type of fruits can have different colors and shapes, existing methods have produced low accuracy rates. This paper proposes a novel fruit classification system based on Haar-like features in conjunction with multiple Viola-Jones detectors. Experiments have been performed on two types of fruits: apple and pineapple. Satisfactory experimental results encourage us to extend the proposed framework on a variety of other fruits.