IMPLEMENTASI ALGORITMA CONVOLUTIONAL NEURAL NETWORK (CNN) MOBILENETV2 DAN MEDIAPIPE FACE MESH UNTUK PREDIKSI USIA BERBASIS MOBILE

Authors

  • Muhammad Ersa Putra Nasution Universitas Islam Sumatera Utara, Medan, Indonesia Author
  • Khairuddin Nasution Universitas Islam Sumatera Utara, Medan, Indonesia Author
  • Mhd.Zulfansyuri Siambaton Universitas Islam Sumatera Utara, Medan, Indonesia Author

DOI:

https://doi.org/10.71282/jurmie.v3i11.2738

Keywords:

age estimation, CNN, MobileNetV2, MediaPipe Face Mesh, Flutter, TensorFlow Lite

Abstract

Age estimation from facial images can be applied in various fields, but most existing systems require a server and an internet connection. This study aims to implement the Convolutional Neural Network (CNN) MobileNetV2 algorithm and MediaPipe Face Mesh for age estimation in an Android mobile application that runs offline. The MobileNetV2 model was trained using transfer learning on the UTKFace dataset of 13,081 facial photos through three training stages, and then converted to the TensorFlow Lite format with a size of 4.9 MB. MediaPipe Face Mesh is used to detect faces and crop the facial area before the image is processed by the model. The application, named AgeDetect, was built using the Flutter framework and the Dart programming language, with an SQLite database to store the testing history. Black box testing on seven application pages achieved a 100% success rate across 38 scenarios. Age estimation testing using 20 facial photos from Lingkungan II Harjosari I, Medan City, produced a Mean Absolute Error (MAE) of 3.50 years and an accuracy of 75% at a tolerance of ±5 years. Without MediaPipe Face Mesh, the MAE increased to 4.19 years, indicating that the use of MediaPipe Face Mesh improved accuracy by 16.5%. Predictions that exceeded the tolerance were influenced by differences between Southeast Asian facial characteristics and the training dataset, which is dominated by Caucasian faces, as well as by photo lighting conditions.

Downloads

Download data is not yet available.

References

APJII. (2023). Laporan survei internet Indonesia 2023. Asosiasi Penyelenggara Jasa Internet Indonesia.

Arifandi, A. (2022). Identifikasi dan prediksi umur serta jenis kelamin berdasarkan citra wajah menggunakan algoritma Convolutional Neural Network (CNN). RAINSTEK: Jurnal Terapan Sains dan Teknologi, 4(2), 89–96. https://doi.org/10.21067/jtst.v4i2.6985

Arsyanda, & Fakhri, M. M. (2024). StudySync mobile application design for student academic activity management based on SQLite database. Journal of Embedded Systems, Security and Intelligent Systems, 5(3), 223–232. https://doi.org/10.59562/jessi.v5i3.5042

Faizin, M. D., & Asmunin. (2026). Sistem deteksi citra wajah manusia hasil generatif AI berbasis mobile menggunakan model transfer learning MobileNetV2. Jurnal Manajemen Informatika, 15(2), 406–413.

Flutter Community. (2023). sqflite: SQLite plugin for Flutter. pub.dev. https://pub.dev/packages/sqflite

Isdaryani, F., Farida, N., & Lugina, A. (2022). Sistem pengenalan penggunaan masker dan pemantauan suhu tubuh penumpang untuk mendukung protokol kesehatan. Jurnal Nasional Teknik Elektro dan Teknologi Informasi (JNTETI), 11(2), 140–147.

Kencana, N. W., Umar, R., & Murinto. (2023). Implementasi transfer learning untuk klasifikasi jenis ras ayam menggunakan arsitektur MobileNetV2. JIP (Jurnal Informatika Polinema), 9(3), 147–154.

Meilany, M., & Rahmadewi, R. (2025). Implementasi deteksi ekspresi wajah, usia, dan gender real time berbasis TensorFlow. JATI (Jurnal Mahasiswa Teknik Informatika), 9(2), 3203–3210.

Munarto, R., Fahrizal, R., & Darma, A. (2021). Klasifikasi gender dan usia berdasarkan citra wajah manusia menggunakan convolutional neural network. Jurnal Ilmiah Setrum, 10(2), 30–43. https://doi.org/10.36055/setrum.v10i2.12991

Abbood, Z. T., Al-Turfi, M. N., & Almajmaie, L. K. A. (2023). An abbreviated review of deep learning-based image classification models. Indonesian Journal of Electrical Engineering and Computer Science, 30(1), 491–500. https://doi.org/10.11591/ijeecs.v30.i1.pp491-500

Akuji, I. I. (2023). Artificial intelligence and machine learning in modern technology: A comprehensive review. International Journal of Advanced Research in Mathematics and Sciences (IJARMS). Al-Hikmah University.

AL-Jibory, F. K. (2023). Predicting age and gender using AlexNet. TEM Journal, 12(1), 512–518. https://doi.org/10.18421/TEM121-61

Alsaleh, A., & Perkgoz, C. (2023). A space and time efficient convolutional neural network for age group estimation from facial images. PeerJ Computer Science, 9, e1395. https://doi.org/10.7717/peerj-cs.1395

Ameen, S. Y., & Mohammed, D. Y. (2022). Developing cross-platform library using Flutter. European Journal of Engineering and Technology Research, 7(2), 18–21. https://doi.org/10.24018/ejeng.2022.7.2.2740

Bhat, V. S., Patilkulkarni, S. A., Karkannavar, P., Aditi, M., & Chandunavar, M. (2022). Gender and age classification using deep learning. International Research Journal of Engineering and Technology (IRJET), 9(5).

ELKarazle, K., Raman, V., & Then, P. (2022). Facial age estimation using machine learning techniques: An overview. Big Data and Cognitive Computing, 6(4), 128. https://doi.org/10.3390/bdcc6040128

Fuad, M. T. H., Fime, A. A., Sikder, D., Iftee, M. A. R., Rabbi, J., Al-Rakhami, M. S., Gumaei, A., Sen, O., Fuad, M., & Islam, M. N. (2021). Recent advances in deep learning techniques for face recognition. IEEE Access, 9, 99112–99142. https://doi.org/10.1109/ACCESS.2021.3096136

George, G., Uppin, C., & Bello, U. A. (2024). Human age estimation from face images with deep convolutional neural networks using transfer learning. Preprints, 2024010350. https://doi.org/10.20944/preprints202401.0350.v1

Google. (2023). Save data using SQLite. Android Developers Documentation. https://developer.android.com/training/data-storage/sqlite

Google. (2023). TensorFlow Lite: Deploy machine learning models on mobile and edge devices. TensorFlow. https://www.tensorflow.org/lite

Krichen, M. (2023). Convolutional neural networks: A survey. Computers, 12(8), 151. https://doi.org/10.3390/computers12080151

Kumar, A., Baiju, A., Timofte, R., & Ignatov, D. (2026). MobileAgeNet: Lightweight facial age estimation for mobile deployment. arXiv:2604.17007.

Li, Z., Liu, F., Yang, W., Peng, S., & Zhou, J. (2022). A survey of convolutional neural networks: Analysis, applications, and prospects. IEEE Transactions on Neural Networks and Learning Systems, 33(12), 6999–7019. https://doi.org/10.1109/TNNLS.2021.3084827

Lovrić, L., Fischer, M., Röderer, N., & Wünsch, A. (2023). Evaluation of the cross-platform framework Flutter using the example of a cancer counselling app. Proceedings of the 9th International Conference on ICT4AWE, 135–142. https://doi.org/10.5220/0011824500003476

Mounir, S. E. A. (2023). Software engineering for mobile applications, a survey on challenges and solutions. arXiv:2301.00602.

Russell, S. J., & Norvig, P. (2022). Artificial intelligence: A modern approach (4th ed., Global Edition). Pearson Education Limited.

Thaman, B., Cao, T., & Caporusso, N. (2022). Face mask detection using MediaPipe Facemesh. 2022 45th Jubilee International Convention on Information, Communication and Electronic Technology (MIPRO), 378–382. https://doi.org/10.23919/MIPRO55190.2022.9803531

Wang, X., Peng, J., Zhang, S., Chen, B., Wang, Y., & Guo, Y. (2022). A survey of face recognition. arXiv:2212.13038.

Yu, F., Xiu, X., & Li, Y. (2022). A survey on deep transfer learning and beyond. Mathematics, 10(19), 3619. https://doi.org/10.3390/math10193619

Downloads

Published

22-11-2026

How to Cite

IMPLEMENTASI ALGORITMA CONVOLUTIONAL NEURAL NETWORK (CNN) MOBILENETV2 DAN MEDIAPIPE FACE MESH UNTUK PREDIKSI USIA BERBASIS MOBILE. (2026). Jurnal Riset Multidisiplin Edukasi, 3(11), 612-632. https://doi.org/10.71282/jurmie.v3i11.2738

Most read articles by the same author(s)

1 2 > >> 

Similar Articles

71-80 of 274

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