Affordable Access

Access to the full text

New effective techniques for automatic detection and classification of external olive fruits defects based on image processing techniques

Authors
  • Hussain Hassan, Nashat M.1
  • Nashat, Ahmed A.1
  • 1 Fayoum University, Electronics and Communication Engineering Department, El Faiyûm, 63514, Egypt , El Faiyûm (Egypt)
Type
Published Article
Journal
Multidimensional Systems and Signal Processing
Publisher
Springer US
Publication Date
Mar 31, 2018
Volume
30
Issue
2
Pages
571–589
Identifiers
DOI: 10.1007/s11045-018-0573-5
Source
Springer Nature
Keywords
License
Yellow

Abstract

One of the major concerns for fruit selling companies, at present, is to find an effective way for rapid classification and detection of fruit defects. Olive is one of the most important agricultural product, which receives great attention from fruit and vegetables selling companies, for its utilization in various industries such as oils and pickles industry. The small size and multiple colours of the olive fruit increases the difficulty of detecting the external defects. This paper presents new efficient methods for detecting and classifying automatically the external defects of olive fruits. The proposed techniques can separate between the defected and the healthy olive fruits, and then detect and classify the actual defected area. The proposed techniques are based on texture analysis and the homogeneity texture measure. The results and the performance of proposed techniques were compared with varies techniques such as Canny, Otsu, local binary pattern algorithm, K-means, and Fuzzy C-Means algorithms. The results reveal that proposed techniques have the highest accuracy rate among other techniques. The simplicity and the efficiency of the proposed techniques make them appropriate for designing a low-cost hardware kit that can be used for real applications.

Report this publication

Statistics

Seen <100 times