Deteksi Cacat Biji Kopi Berdasarkan Spesifikasi Specialty Coffee Association dengan YOLOv8
Abstract
Produksi kopi diprediksi akan menambah
sebanyak 5,8% pada tahun 2024. Kopi dapat dibagi menjadi
berbagai kualitas berdasarkan kecacatan yang ditemukan.
Pemeriksaan kualitas kopi biasanya dilakukan melalui inspeksi
visual, yang memakan waktu dan subjektif. Penelitian lain yang
telah dilakukan menerapkan metode yang hanya mendeteksi
adanya cacat, atau menggunakan tekstur untuk penilaian
kualitas kopi. Penelitian ini menggunakan pendekatan yang
berbeda, dengan metode You Only Look Once versi 8
(YOLOv8) untuk mendeteksi cacat berdasarkan standar
Specialty Coffee Association (SCA). Dataset yang disusun
adalah kumpulan 204 citra yang menampilkan 300 gram biji
kopi hijau arabika mandheling. Dengan menggunakan
pendekatan di mana model akan mendeteksi dan
mengklasifikasikan cacat berdasarkan standar SCA, model
dapat memberikan hasil yang lebih akurat dalam mendeteksi
cacat biji kopi hijau dan mempermudah inspeksi kualitas kopi.
Kontribusi utama dari penelitian ini adalah model yang dapat
mendeteksi biji kopi yang memiliki cacat berdasarkan standar
SCA. Model yang dibuat memiliki mean average precision
sebesar 0,14.
Kata kunci— biji kopi, deteksi objek, computer vision,
kualitas kopi.
References
International Coffee Organization. Coffee report and
outlook. Technical report, International Coffee
Organization, 2023.
SCAA green coffee beans classification,.
http://www.coffeeresearch. org/coffee/scaaclass.htm,
Accessed: 2024-04-01.
James Kosalos, Rob Stephen, Steven Diaz, Paul
Songer, Man´e Alves, Marty Curtis, and Steven SungYong Kil. Arabica Green Coffee Defect Handbook.
Specialty Coffee Association of America, 2013.
Nen-Fu Huang, Dong-Lin Chou, Chia-An Lee, FengPing Wu, An-Chi Chuang, Yi-Hsien Chen, and YinChun Tsai. Smart agriculture: real-time classification of
green coffee beans by using a convolutional neural
network. IET Smart Cities, 2, 10 2020.
Serawork Wallelign, Mihai Polceanu, Towfik Jemal,
and C´edric Buche. Coffee grading with convolutional
neural networks using small datasets with high variance.
Edwin R Arboleda, Arnel C Fajardo, and Ruji P
Medina. Classification of coffee bean species using
image processing, artificial neural network and k
nearest neighbors. In 2018 IEEE international
conference on innovative research and development
(ICIRD), pages 1–5. IEEE, 2018.
JR Arunkumar and Tagele berihun Mengist.
Developing ethiopian yirga-cheffe coffee grading
model using a deep learning classifier. International
Journal of Innovative Technology and Exploring
Engineering, 9(4):3303–3309, 2020.
Muhammad NS Akbar, Ema Rachmawati, and
Febryanti Sthevanie. Visual feature and machine
learning approach for arabica green coffee beans grade
determination. In Proceedings of the 6th International
Conference on Communication and Information
Processing, pages 97–104, 2020.
Edwin R. Arboleda, Arnel C. Fajardo, and Ruji P.
Medina. An image processing technique for coffee
black beans identification. In 2018 IEEE International
Conference on Innovative Research and Development
(ICI-RD), pages 1–5, 2018.
Shih-Yu Chen, Chuan-Yu Chang, Cheng-Syue Ou, and
Chou-Tien Lien. Detection of insect damage in green
coffee beans using vis-nir hyperspe-ctral imaging.
Remote Sensing, 12(15), 2020.
Carlito Pinto, Junya Furukawa, Hidekazu Fukai, and
Satoshi Tamura. Classification of green coffee bean
images basec on defect types using convolutional neural
network (cnn). In 2017 International Conference on
Advanced Informatics, Concepts, Theory, and
Applications (ICAICTA), pages 1–5, 2017.
Glenn Jocher, Ayush Chaurasia, and Jing Qiu.
Ultralytics YOLO. https:
//github.com/ultralytics/ultralytics, 2023. Accessed:
-04-11.
Juan Terven, Diana-Margarita C´ordova-Esparza, and
Julio-Alejandro Romero-Gonz´alez. A comprehensive
review of yolo architectures in com-puter vision: From
yolov1 to yolov8 and yolo-nas. Machine Learning and
Knowledge Extraction, 5(4):1680–1716, 2023.
Palmiro Poltronieri and Franca Rossi. Challenges in
specialty coffee pro-cessing and quality assurance.
Challenges, 7(2), 2016.
Amanda R Hale, Paul M Ruegger, Philippe Rolshausen,
James Borneman, and Jiue-in Yang. Fungi associated
with the potato taste defect in coffee beans from rwanda.
Botanical studies, 63(1):17, 2022.
Rafael Carlos Eloy Dias, Sebastian Ed Wieland Opitz,
and Chahan Ye-retzian. Bioactive compounds in blends
