Klasifikasi Nominal Uang Berbasis Stacking Ensemble pada Celengan IoT dengan Sensor TCS34725
DOI:
https://doi.org/10.30865/json.v8i1.10129Keywords:
Celengan IoT, Klasifikasi Uang, Machine Learning, Stacking Ensemble, TCS34725Abstract
Celengan konvensional memiliki kelemahan mendasar berupa ketiadaan pencatatan saldo otomatis, sedangkan upaya identifikasi mandiri menghadapi kendala degradasi fisik uang lusuh dan fluktuasi spektrum warna. Penelitian ini bertujuan mengembangkan sistem celengan cerdas berbasis Internet of Things (IoT) yang mampu mengklasifikasikan tujuh pecahan nominal uang kertas Rupiah emisi tahun 2016 dan 2022 secara otomatis dan memantau saldo tabungan secara real-time. Perangkat keras sistem mengintegrasikan mikrokontroler ESP32 dan sensor spektrum warna TCS34725 dengan server REST API berbasis Python Flask. Data masukan warna diproses melalui ekstraksi 21 fitur spektrum diskriminatif (mencakup warna mentah, rasio relatif, kontras, ruang warna HSV, proyeksi trigonometri Hue, diferensial warna lawan, rasio logaritmik, dan luminansi), penyaringan noise sensor menggunakan Edited Nearest Neighbours (ENN), penstandaran skala StandardScaler, serta penyeimbangan distribusi kelas data latih menggunakan SMOTE. Proses klasifikasi menerapkan arsitektur Stacking Ensemble berbasis pohon keputusan yang menggabungkan Random Forest, XGBoost, CatBoost, dan Gradient Boosting sebagai model dasar, serta Random Forest Classifier sebagai model meta. Hasil pengujian menunjukkan bahwa model Stacking Ensemble memperoleh akurasi sebesar 93,07% dengan weighted average F1-score sebesar 0,93 pada 202 data uji murni. Pengujian langsung pada perangkat fisik celengan IoT menggunakan 70 lembar uang kertas asli independen dalam tiga kondisi fisik (baru, sedang, dan lusuh) menghasilkan tingkat keberhasilan pembacaan real-time sebesar 87,1%. Sistem berhasil mengintegrasikan pencatatan transaksi ke database MySQL dan pemantauan saldo pada dashboard web secara terpadu.
References
[1] B. Indonesia, “Unsur Pengaman dan Spesifikasi Ciri Uang Kertas Rupiah Emisi 2022,” Bank Indonesia Official Web Portal. [Online]. Available: https://www.bi.go.id/id/rupiah/gambar-uang/Default.aspx
[2] A. Syahputra, Syahwin, and M. Z. Siambaton, “Rancang Bangun Sistem Monitoring Smart Savings Anak Sekolah pada Celengan Uang Kertas Berbasis Internet of Things ( IoT ),” J. Ilmu Komput., pp. 0–4, 2025, doi: 10.56211/helloworld.v4i2.959.
[3] R. M. Komarudin, W. Suteddy, and D. A. R. Agustini, “Rancang Bangun Celengan Pintar Berbasis IoT Menggunakan RFID dan Notifikasi Telegram,” Jambura J. Electr. Electron. Eng., vol. 6, no. 2, pp. 240–248, 2024, doi: 10.37905/jjeee.v6i2.26179.
[4] M. H. Alfarisi, A. L. Prasasti, and R. R. Septiawan, “IoT-Based Banknotes Saving Automation System,” CEPAT J. Comput. Eng. Progress, Appl. Technol., vol. 2, no. 2, pp. 62–69, 2023, doi: 10.25124/cepat.v2i01.5499.
[5] F. A. T. Putra, A. B. P. Negara, and H. Sastypratiwi, “Perbandingan Kinerja Naive Bayes dan Random Forest dengan Penanganan Imbalance Data,” J. Inform. Polinema, vol. 12, no. 2, pp. 401–408, 2026, doi: 10.33795/jip.v12i2.8424.
[6] B. A. Sadewa and Y. Yamasari, “Implementasi Deep Transfer Learning untuk Klasifikasi Nominal Uang Kertas Rupiah,” J. Informatics Comput. Sci., vol. 05, pp. 543–551, 2024, doi: 10.26740/jinacs.v5n04.p543-551.
[7] R. K. Habsari, M. I. Khoiri, Purbandini, B. D. Satoto, A. H. Salim, and T. Kristanto, “Identification of Indonesian Rupiah Paper Currency Denominations Using First Order Statistical Feature Extraction and k-Nearest Neighbor,” Int. J. INFORMATICS Vis., vol. 9, no. May, 2025, doi: 10.62527/joiv.9.3.3037.
[8] Irma, D. Hendryadi, and J. Iryani, “Prototipe alat pemilah buah kopi otomatis pada panen racutan (stripping) menggunakan sensor warna TCS34725 dan modul Arduino,” J. Syst. Inf. Comput. Inst. Teknol. Dan Bisnis Bina Adinata, vol. 2, no. 1, 2024, doi: 10.63989/ammatoa.v2i1.74.
[9] D. N. Masmur and M. Snae, “Rancang bangun sistem sortir buah cabai berdasarkan warna berbasis arduino uno,” J. Teknol. Inf., vol. 15, no. c, pp. 33–42, 2024, doi: 10.52972/hoaq.vol15no1.p33-42.
[10] Sudarto and Kusrini, “Klasifikasi Tsunami Gempa Bumi dengan Teknik Stacking Ensemble Machine Learning,” J. Inform. Polinema, vol. 10, no. 4, pp. 511–519, 2024, doi: 10.33795/jip.v10i4.5655.
[11] N. Anggraini, S. J. Putra, L. K. Wardhani, F. D. U. Arif, N. Hakiem, and I. M. Shofi, “A Comparative Analysis of Random Forest , XGboost , and LightGBM Algorithms for Emotion Classification in Reddit Comments,” J. Tek. Inform., vol. 17, no. 1, pp. 88–97, 2024, doi: 10.15408/jti.v17i1.38651.
[12] K. H. Rambe et al., “Model Ensemble Stacking untuk Klasifikasi Big Data Stunting Berbasis XGBoost dan MLP,” Remik Ris. dan E-Jurnal Manaj. Inform. Komput., vol. 10, no. 1, pp. 385–395, 2026, doi: 10.33395/remik.v10i1.15903.
[13] N. A. Siagian, S. P. Sipayung, A. Rikki, and N. Marbun, “Integrating SMOTE with XGBoost for Robust Classification on Imbalanced Datasets : A Dual-Domain Evaluation,” J. dan Penelit. Tek. Inform., vol. 9, no. 3, pp. 1094–1107, 2025, doi: 10.33395/sinkron.v9i3.15029.
[14] M. A. Hamid and E. R. Subhiyakto, “Performance Comparison of Random Forest , SVM , and XGBoost Algorithms with SMOTE for Stunting Prediction,” J. Appl. Informatics Comput., vol. 9, no. 4, pp. 1163–1169, 2025, doi: 10.30871/jaic.v9i4.9701.
[15] D. Trisatya and P. Haryoko, “Improving Customer Churn Detection Through Balanced Ensemble Learning,” J. Inform. Polinema, vol. 12, no. 3, pp. 431–438, 2026, doi: 10.33795/jip.v12i3.9500.
[16] E. Systems, “ESP32 Series Datasheet: Hardware Specifications and Wi-Fi Microcontroller Capabilities,” Espressif Systems Official Technical Documentation. [Online]. Available: https://www.espressif.com/en/products/socs/esp32
[17] I. K. G. D. Putra, Pengolahan Citra Digital dan Pengenalan Pola. Yogyakarta: Penerbit Andi, 2021.
[18] N. A. Levina and M. Rahardi, “Analysis of Stacking Ensemble Method in Machine Learning Algorithms to Predict Student Depression,” J. Appl. Informatics Comput., vol. 9, no. 6, pp. 3283–3294, 2025, doi: 10.30871/jaic.v9i6.11453.
[19] R. Sukmawardani, L. Nurpulaela, and R. Rahmadewi, “Implementasi Arsitektur Resnet152 untuk Klasifikasi Uang Kerta Rupiah dengan Metode Transfer Learning,” J. Mhs. Tek. Inform., vol. 8, no. 4, pp. 5650–5657, 2024, doi: 10.36040/jati.v8i4.10001.
[20] D. Pratama, Sucipto, and B. C. Octariadi, “Penerapan Teknik Stacking Ensemble Learning untuk Identifikasi Infeksi Saluran Kemih,” J. Mhs. Tek. Inform., vol. 10, no. 4, pp. 6035–6040, 2026, doi: 10.36040/jati.v10i4.18612.
[21] J. Bahaswan, Sopingi, and H. Hasanah, “Perbandingan Random Forest dan XGBoost dalam Prediksi Status Pinjaman Menggunakan Cross Validation dan Analisis Feature Importance,” J. Mhs. Tek. Inform., vol. 10, no. 4, pp. 6908–6915, 2026, doi: 10.36040/jati.v10i4.18823.
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Jurnal Sistem Komputer dan Informatika (JSON)
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
This work is licensed under a Creative Commons Attribution 4.0 International License
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under Creative Commons Attribution 4.0 International License that allows others to share the work with an acknowledgment of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgment of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (Refer to The Effect of Open Access).
