KLI

Recognition of Multiple Panamanian Watermelon Varieties Based on Feature Extraction Analysis

Metadata Downloads
Abstract
In this paper we present a multi-watermelon varieties recognition analysis using the color and texture information. The goal is to classify the watermelon taking into account its features, despite its variety. To improve the efficacy of the proposed method the images were preprocessed using morphological and adaptive threshold methods. Also, to extract the candidate region for feature extraction, vertical and horizontal histograms were used. Finally, the model was tested on a group of watermelons to prove the effectiveness of the implementation; the result shows it is capable of classifying watermelon varieties by just considering the presence or not of stripes, as well as the color information. The importance of this study is that it uses as test subjects local varieties (Panamanian) of watermelons, thus helping the knowledge of national export products with computer vision.
Author(s)
Javier E. Sanchez-GalanAnel HenryFatima RangelEmmy Saez조강현다닐로 카세레스
Issued Date
2021
Type
Article
Keyword
ClusteringFeature extractionModel fittingRecognitionWatermelon
DOI
10.1007/978-3-030-84529-2_6
URI
https://oak.ulsan.ac.kr/handle/2021.oak/9163
https://ulsan-primo.hosted.exlibrisgroup.com/primo-explore/fulldisplay?docid=TN_cdi_springer_books_10_1007_978_3_030_84529_2_6&context=PC&vid=ULSAN&lang=ko_KR&search_scope=default_scope&adaptor=primo_central_multiple_fe&tab=default_tab&query=any,contains,Recognition%20of%20Multiple%20Panamanian%20Watermelon%20Varieties%20Based%20on%20Feature%20Extraction%20Analysis&offset=0&pcAvailability=true
Publisher
Lecture Notes in Computer Science
Location
스위스
Language
영어
ISSN
0302-9743
Citation Volume
12837
Citation Number
1
Citation Start Page
65
Citation End Page
75
Appears in Collections:
Engineering > IT Convergence
Authorize & License
  • Authorize공개
Files in This Item:
  • There are no files associated with this item.

Items in Repository are protected by copyright, with all rights reserved, unless otherwise indicated.