Detection of the Maturity Level of Lempuyangan Robusta Coffee Fruit Using HSV Segmentation and K-Means Clustering

Authors

  • Muhammad Ferdy Irmawan UIN Walisongo Semarang
  • Wenty Dwi Yuniarti UIN Walisongo Semarang
  • Moh Hadi Subowo UIN Walisongo Semarang

DOI:

https://doi.org/10.30871/jaic.v10i2.12541

Keywords:

Clustering, Coffee maturity detection, HSV segmentation, Image processing, Silhouette score

Abstract

This study proposes an image based approach to detect the ripeness level of Lempuyangan Robusta coffee using HSV color segmentation and K-Means clustering. Manual sorting conducted by farmers is subjective and inconsistent due to lighting variation and perceptual differences. A total of 901 coffee fruit images were collected and processed through resizing, denoising using Non-Local Means, RGB to HSV conversion, and color based segmentation. Mean HSV features were extracted and clustered into three maturity levels: unripe, semi-ripe, and ripe. Cluster quality was evaluated using the Silhouette Score to measure compactness and separation, and an additional evaluation metric, the Davies Bouldin Index (DBI), to assess the quality of separation between clusters by measuring the ratio between intra cluster dispersion and inter cluster distance. Experimental results show that the proposed method achieves an overall Silhouette Score of 0.6119 and a Davies Bouldin Index of 0.5019, indicating good cluster structure stability with minimal difference between training and testing subsets. The findings demonstrate that HSV segmentation combined with K-Means provides an effective and objective approach to support automated coffee sorting systems under real-field conditions.

Downloads

Download data is not yet available.

Author Biographies

Muhammad Ferdy Irmawan, UIN Walisongo Semarang

Students of the Information Technology Study Program, Walisongo State Islamic University, Semarang

Wenty Dwi Yuniarti, UIN Walisongo Semarang

Lecturer at the Information Technology Study Program, Walisongo State Islamic University, Semarang

Moh Hadi Subowo, UIN Walisongo Semarang

Lecturer at the Information Technology Study Program, Walisongo State Islamic University, Semarang

References

[1] Kementerian Pertanian RI, “Outlook Komoditas Perkebunan Kopi,” Pus. Data dan Sist. Inf. Pertan. Sekr. Jendral - Kementrian Pertan. 2022, pp. 1–103, 2022.

[2] Agus AP, “Dari Total 35 Kabupaten dan Kota di Jateng, 29 Daerah Memiliki Perkebunan Kopi,” https://radarsemarang.jawapos.com/, 2021. https://radarsemarang.jawapos.com/service/721385309/dari-total-35-kabupaten-dan-kota-di-jateng-29-daerah-memiliki-perkebunan-kopi

[3] B. B. Bengkulu, “Dan Mutu Green Bean Kopi Robusta the Effect of Fruit Maturity Levels on the Yield Losses and Quality of,” vol. 8, pp. 67–78, 2021.

[4] I. N. Alam, W. Warkoyo, and D. D. Siskawardani., “Karakteristik Tingkat Kematangan Buah Kopi Robusta ( Coffea,” Food Technol. Halal Sci. J., vol. 5, no. 2, pp. 169–185, 2022.

[5] A. D. Riastuti, S. Komarayanti, and A. P. Utomo, “Karakteristik Morfologi Biji Kopi Tobusta (Coffea Canephora) pascapanen di kawasan lereng Meru Betiri sebagai sumber belajar smk dalam bentuk e-modul,” J. Ilmu Pendidik., vol. 5, no. 2, pp. 1–13, 2021.

[6] K. P. R. Dirjen Perkebunan, “Akselerasi Nilai Tambah Untuk Mendorong Ekspor Melalui Perbaikan Pascapanen Kopi Di Jawa Tengah,” https://ditjenbun.pertanian.go.id/, 2019. https://ditjenbun.pertanian.go.id/akselerasi-nilai-tambah-untuk-mendorong-ekspor-melalui-perbaikan-pascapanen-kopi-di-jawa-tengah

[7] D. E. Maulina, Daryono, and Y. Purwanti, “Pengawasan Mutu pada Proses Pengolahan,” J. Technol. Food Process., vol. 2, no. 02, pp. 12–18, 2022.

[8] J. Rusman and N. Pasae, “Prototype Sistem Penyortir Buah Kopi Arabika Berdasarkan Tingkat Kematangan Menggunakan Metode Support Vector Machine Prototype of Arabica Coffee Sorting System Based on Maturity Level Using Support Vector Machine,” vol. 12, no. 1, pp. 65–72, 2023, doi: 10.34148/teknika.v12i1.602.

[9] Kementerian Pertanian RI, “BUKU OUTLOOK komoditas perkebunan,” Sekr. Jenderal Kementeri. Pertan., p. 90, 2023.

[10] S. Hira and S. Lande, “Detection of fruit ripeness using image processing,” Int. J. Health Sci. (Qassim)., vol. 6, no. May, pp. 3874–3886, 2022, doi: 10.53730/ijhs.v6ns6.10146.

[11] A. Tri Laksono, P. Citra Digital Buah, P. Wanarti Rusmamto, and M. Syariffuddien Zuhrie, “Pengolahan Citra Digital Buah Murbei Dengan Algoritma LDA (Linear Discriminant Analysis),” Indones. J. Eng. Technol., vol. 4, no. 2, pp. 71–78, 2022, [Online]. Available: https://journal.unesa.ac.id/index.php/inajet

[12] M. Fahmi Wibawa, M. A. Rahman, and A. W. Widodo, “Penerapan Ruang Warna HSV dan Ekstraksi Fitur Tekstur Local Binary Pattern untuk Tingkat Kematangan Sangrai Biji Kopi,” J. Pengemb. Teknol. Inf. dan Ilmu Komput., vol. 5, no. 7, pp. 2819–2825, 2021.

[13] S. R. Raysyah, A. Veri, and I. M. Dadang, “Classification of Coffee Fruit Maturity Level Based on Color Detection Using Knn and Pca Method,” JSiI (Journal Inf. Syst., vol. 8, no. 2, pp. 88–95, 2021.

[14] N. Auliasari, Rahma Nur; Novamizanti, Ledya; Ibrahim, “Identifikasi Kematangan Daun Teh Berbasis Fitur Warna Hue Saturation Intensity ( HSI ) dan Hue Saturation Value ( HSV ) ( Identification Maturity Tea Leaves Based on Color Feature Hue Saturation Intensity ( HSI ) and Hue Saturation Value,” JUITA J. Inform. e-ISSN, vol. 8, no. November, pp. 217–223, 2020.

[15] M. Y. Pusadan and I. Safitri, “JURNAL RESTI The Image Extraction Using the HSV Method to Determine the Maturity Level of Palm Oil Fruit with the k-nearest Neighbor Algorithm,” J. RESTI (Rekayasa Sist. dan Teknol. Informasi), vol. 5, no. 158, pp. 1448–1456, 2026.

[16] T. Andri; Paulus;Wong,Ng Poi ; Gunawan, “Segmentasi Buah Menggunakan Metode K-Means,” vol. 15, no. 2, pp. 91–100, 2014.

Downloads

Published

2026-04-26

How to Cite

[1]
M. F. Irmawan, W. D. Yuniarti, and M. H. Subowo, “Detection of the Maturity Level of Lempuyangan Robusta Coffee Fruit Using HSV Segmentation and K-Means Clustering”, JAIC, vol. 10, no. 2, pp. 1990–1995, Apr. 2026.

Similar Articles

<< < 46 47 48 

You may also start an advanced similarity search for this article.