IMPLEMENTASI KLASIFIKASI DUA TAHAP BERBASIS MOBILENETV2 PADA APLIKASI ANDROID UNTUK DETEKSI DAN KLASIFIKASI VARIETAS MANGGA
DOI:
https://doi.org/10.29407/6dkt6g67Abstract
Indonesia memiliki beragam varietas mangga lokal. Namun, kemiripan morfologi antarvarietas membuat proses identifikasi secara manual rentan mengalami kesalahan. Penelitian ini bertujuan untuk mengembangkan aplikasi Android berbasis MobileNetV2 dengan pendekatan Two-Stage Classification untuk mendeteksi objek mangga dan mengklasifikasikan varietasnya secara otomatis. Dataset yang digunakan terdiri atas 400 citra untuk model deteksi dan 600 citra untuk model klasifikasi varietas mangga. Tahap preprocessing meliputi resize citra, normalisasi, dan augmentasi data. Model hasil pelatihan kemudian dikonversi ke format TensorFlow Lite agar dapat diintegrasikan ke dalam aplikasi mobile. Hasil evaluasi internal menunjukkan bahwa model deteksi memperoleh akurasi sebesar 88%, sedangkan model klasifikasi memperoleh akurasi sebesar 91%. Pengujian aplikasi pada data riil dilakukan menggunakan 25 citra, yang terdiri atas 20 citra mangga dan 5 citra non-mangga. Dari 20 citra mangga, sistem berhasil mengklasifikasikan 18 citra dengan benar dan mengalami 2 kesalahan prediksi. Sementara itu, seluruh 5 citra non-mangga berhasil dideteksi dengan benar sebagai objek non-mangga. Meskipun demikian, faktor latar belakang, pencahayaan, dan sudut pengambilan gambar masih memengaruhi nilai probabilitas sistem dalam melakukan prediksi pada data riil.
Keywords:
Aplikasi Mobile, CNN, Klasifikasi, Mangga, MobileNetV2##plugins.themes.default.displayStats.downloads##
References
[1] C. Rozikin, A. Susilo, and Y. Irawan, “Klasifikasi Jenis Buah Mangga dengan Metode
Backpropagation,” pp. 1–12.
[2] Y. A. Sitorus, Y. A. Sari, and S. Adinugroho, “Evaluasi Komparatif Arsitektur
Lightweight CNN , MobileNetV2 , dan,” vol. 9, no. 8, pp. 1–6, 2025.
[3] A. K. Ratha, N. K. Barpanda, P. K. Sethy, and S. K. Behera, “Automated Classification
of Indian Mango Varieties Using Machine Learning and MobileNet-v2 Deep Features,”
Trait. du Signal, vol. 41, no. 2, pp. 669–679, 2024, doi: 10.18280/ts.410210.
[4] R. Mardianto, Stefanie Quinevera, and S. Rochimah, “Perbandingan Metode Random
Forest, Convolutional Neural Network, dan Support Vector Machine Untuk Klasifikasi
Jenis Mangga,” J. Appl. Comput. Sci. Technol., vol. 5, no. 1, pp. 63–71, 2024, doi:
10.52158/jacost.v5i1.742.
[5] Y. Gulzar, “Fruit Image Classification Model Based on MobileNetV2 with Deep
Transfer Learning Technique,” Sustain., vol. 15, no. 3, 2023, doi: 10.3390/su15031906.
[6] C. Wang et al., “Improved You Only Look Once v . 8 Model Based on Deep Learning :
Precision Detection and Recognition of Fresh Leaves from Yunnan Large-Leaf Tea
Tree,” pp. 1–21, 2024.
[7] H. A. Faudyta, J. T. Sinaga, and E. R. Subhiyakto, “Implementation of MobileNet
Architecture for Skin Cancer Disease Classification,” vol. 8, no. 2, pp. 589–597, 2024.
[8] I. Wulandari, H. Yasin, T. Widiharih, D. Statistika, and U. Diponegoro, “Klasifikasi citra
digital bumbu dan rempah dengan algoritma convolutional neural network (cnn) 1,2,3,”
vol. 9, pp. 273–282, 2020.
[9] C. Aggarwal, “Data Mining”.
[10] C. Buiu, V. R. Dănăilă, and C. N. Răduţă, “MobileNetV2 ensemble for cervical
precancerous lesions classification,” Processes, vol. 8, no. 5, 2020, doi:
10.3390/PR8050595.
[11] Y. Aufar and T. P. Kaloka, “Robusta coffee leaf diseases detection based on
MobileNetV2 model,” Int. J. Electr. Comput. Eng., vol. 12, no. 6, pp. 6675–6683, 2022,
doi: 10.11591/ijece.v12i6.pp6675-6683.
Downloads
Published
Issue
Section
License

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Copyright on any article is retained by the author(s).
- The author grants the journal, right of first publication with the work simultaneously licensed under a Creative Commons Attribution 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.
- The article and any associated published material is distributed under the Creative Commons Attribution-ShareAlike 4.0 International License