GCBICT: Green Coffee Bean Identification Command-line Tool

Coffee is one of the most important agricultural commodities in commodity markets. The quality of coffee beverages strongly depends on that of green coffee beans. However, the conventional selection technique mainly relies on personnel visual inspection, which is subjective and time-consuming. Based...

Full description

Saved in:
Bibliographic Details
Main Authors: Shu-Min Tan, Shih-Hsun Hung, Je-Chiang Tsai
Format: Article
Language:English
Published: Elsevier 2024-12-01
Series:SoftwareX
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2352711024002140
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846150150839336960
author Shu-Min Tan
Shih-Hsun Hung
Je-Chiang Tsai
author_facet Shu-Min Tan
Shih-Hsun Hung
Je-Chiang Tsai
author_sort Shu-Min Tan
collection DOAJ
description Coffee is one of the most important agricultural commodities in commodity markets. The quality of coffee beverages strongly depends on that of green coffee beans. However, the conventional selection technique mainly relies on personnel visual inspection, which is subjective and time-consuming. Based on our recently discovered site-specific color characteristics of the seat coat of green coffee beans and support vector machines (a machine learning classifier), the Python-based identification/evaluation scheme of beans, GCBICT, provides an affordable, effective, and user-friendly way to identify qualified beans and their growing sites.The command-line tool consists of two functions: (1) the Qualified-Defective Separator and (2) the Mixed Separator. The Qualified-Defective Separator function is to distinguish between qualified and defective green coffee beans. Due to the site-specific property of our color characteristics of beans, the training set can be small. The Mixed Separator can identify qualified beans from different growing sites if coffee distributors mix them for cost in their business. Moreover, this function is unique to our evaluation scheme.
format Article
id doaj-art-1664d0414f06467bb33ccdca87ff1a28
institution Kabale University
issn 2352-7110
language English
publishDate 2024-12-01
publisher Elsevier
record_format Article
series SoftwareX
spelling doaj-art-1664d0414f06467bb33ccdca87ff1a282024-11-29T06:23:57ZengElsevierSoftwareX2352-71102024-12-0128101843GCBICT: Green Coffee Bean Identification Command-line ToolShu-Min Tan0Shih-Hsun Hung1Je-Chiang Tsai2Department of Mathematics, National Tsing Hua University, No. 101, Sec. 2, Kuang-Fu Road, Hsinchu 300, TaiwanDepartment of Mathematics, National Tsing Hua University, No. 101, Sec. 2, Kuang-Fu Road, Hsinchu 300, TaiwanDepartment of Mathematics, National Tsing Hua University, No. 101, Sec. 2, Kuang-Fu Road, Hsinchu 300, Taiwan; National Center for Theoretical Sciences, No. 1, Sec. 4, Roosevelt Road, Taipei 106, Taiwan; Corresponding author at: Department of Mathematics, National Tsing Hua University, No. 101, Sec. 2, Kuang-Fu Road, Hsinchu 300, Taiwan.Coffee is one of the most important agricultural commodities in commodity markets. The quality of coffee beverages strongly depends on that of green coffee beans. However, the conventional selection technique mainly relies on personnel visual inspection, which is subjective and time-consuming. Based on our recently discovered site-specific color characteristics of the seat coat of green coffee beans and support vector machines (a machine learning classifier), the Python-based identification/evaluation scheme of beans, GCBICT, provides an affordable, effective, and user-friendly way to identify qualified beans and their growing sites.The command-line tool consists of two functions: (1) the Qualified-Defective Separator and (2) the Mixed Separator. The Qualified-Defective Separator function is to distinguish between qualified and defective green coffee beans. Due to the site-specific property of our color characteristics of beans, the training set can be small. The Mixed Separator can identify qualified beans from different growing sites if coffee distributors mix them for cost in their business. Moreover, this function is unique to our evaluation scheme.http://www.sciencedirect.com/science/article/pii/S2352711024002140Green coffee beansComputer vision evaluation schemeColor characteristicsMachine learning classifier
spellingShingle Shu-Min Tan
Shih-Hsun Hung
Je-Chiang Tsai
GCBICT: Green Coffee Bean Identification Command-line Tool
SoftwareX
Green coffee beans
Computer vision evaluation scheme
Color characteristics
Machine learning classifier
title GCBICT: Green Coffee Bean Identification Command-line Tool
title_full GCBICT: Green Coffee Bean Identification Command-line Tool
title_fullStr GCBICT: Green Coffee Bean Identification Command-line Tool
title_full_unstemmed GCBICT: Green Coffee Bean Identification Command-line Tool
title_short GCBICT: Green Coffee Bean Identification Command-line Tool
title_sort gcbict green coffee bean identification command line tool
topic Green coffee beans
Computer vision evaluation scheme
Color characteristics
Machine learning classifier
url http://www.sciencedirect.com/science/article/pii/S2352711024002140
work_keys_str_mv AT shumintan gcbictgreencoffeebeanidentificationcommandlinetool
AT shihhsunhung gcbictgreencoffeebeanidentificationcommandlinetool
AT jechiangtsai gcbictgreencoffeebeanidentificationcommandlinetool