Bimasakti, Muhammad Satryo Pamungkas (2026) Pengembangan Aplikasi Mobile Pelacakan Nutrisi Menggunakan Pemindaian Label Nutrisi Berbasis OCR. Other thesis, Institut Teknologi Sepuluh Nopember.
![[thumbnail of 05111840000070-Undergraduate_Thesis.pdf]](http://repository.its.ac.id/style/images/fileicons/text.png) |
Text
05111840000070-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (6MB) | Request a copy |
Abstract
Pemantauan asupan gizi harian merupakan aspek penting dalam menjaga kesehatan dan mencegah berbagai penyakit kronis. Seiring berkembangnya teknologi, banyak aplikasi mobile yang dirancang untuk membantu pengguna dalam melacak konsumsi makanan dan mengelola pola makan mereka secara lebih praktis. Salah satu teknologi yang berpotensi besar dalam mendukung proses ini adalah Optical character recognition (OCR), yang memungkinkan ekstraksi teks secara otomatis dari gambar, termasuk label informasi nilai gizi pada kemasan makanan. Penelitian ini mengembangkan aplikasi mobile berbasis Android yang mengekstraksi informasi nilai gizi menggunakan Google ML Kit dengan metode parsing berbasis posisi spasial (spatial positioning). Pendekatan ini memanfaatkan bounding box coordinates dari hasil OCR untuk mengenali struktur tabel dan mencocokkan setiap label nutrisi dengan nilainya berdasarkan posisi vertikal dan horizontal pada gambar. Evaluasi terhadap 50 sampel label gizi menunjukkan sistem mencapai precision 74,51%, recall 55%, dan F1-score 63,19%, dengan performa tertinggi pada bidang natrium dengan F1-score 83,72% dan terendah pada lemak jenuh dengan F1-score 43,48%. Aplikasi mengintegrasikan personalisasi gizi berdasarkan standar Angka Kecukupan Gizi (AKG) Kementerian Kesehatan RI 2019 untuk menghitung kontribusi nutrisi terhadap kebutuhan harian pengguna, sehingga mendukung pengambilan keputusan konsumsi yang lebih sehat.
=================================================================================================================================
Daily nutritional intake monitoring is a crucial aspect of maintaining health and preventing various chronic diseases. With the advancement of technology, many mobile applications have been developed to assist users in tracking their food consumption and managing their dietary habits more practically. One promising technology that can support this process is Optical Character Recognition (OCR), which enables automatic extraction of text from images, including nutritional information labels on food packaging.
This research develops an Android mobile application that extracts nutritional information using Google ML Kit with a spatial positioning-based parsing method. This approach utilizes bounding box coordinates from OCR results to recognize table structures and match each nutrition label with its value based on vertical and horizontal positions in the image. An evaluation of 50 nutrition label samples showed the system achieved 74.51% precision, 55% recall, and 63.19% F1-score, with the highest performance in sodium with an F1-score of 83.72% and the lowest in saturated fat with an F1-score of 43.48%. The application integrates nutritional personalization based on the 2019 Indonesian Ministry of Health's Nutritional Adequacy Intake (RDA) standards to calculate the nutritional contribution to users' daily needs, thus supporting healthier consumption decisions.
| Item Type: | Thesis (Other) |
|---|---|
| Uncontrolled Keywords: | Android, Angka kecukupan gizi, Google ml kit, Label nutrisi, Optical character recognition, Spatial Parsing Android, Google ML kit, Nutrition labels, Optical character recognition, Recommended dietary allowance, Spatial Parsing |
| Subjects: | T Technology > T Technology (General) > T57.5 Data Processing |
| Divisions: | Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Informatics Engineering > 55201-(S1) Undergraduate Thesis |
| Depositing User: | Muhammad Satryo Pamungkas Bimasakti |
| Date Deposited: | 28 Jan 2026 07:50 |
| Last Modified: | 28 Jan 2026 07:50 |
| URI: | http://repository.its.ac.id/id/eprint/130768 |
Available Versions of this Item
- Pengembangan Aplikasi Mobile Pelacakan Nutrisi Menggunakan Pemindaian Label Nutrisi Berbasis OCR. (deposited 28 Jan 2026 07:50) [Currently Displayed]
Actions (login required)
 |
View Item |