Deteksi Objek Makanan Berdasarkan Foto Menggunakan Yolov11 Pada Modul Food Diary Aplikasi Gizicare

Purba, Lihardo Marson (2026) Deteksi Objek Makanan Berdasarkan Foto Menggunakan Yolov11 Pada Modul Food Diary Aplikasi Gizicare. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5025211238-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (18MB)

Abstract

Macronutrient adalah nutrisi utama yang dibutuhkan tubuh dalam jumlah besar untuk menyediakan energi dan mendukung fungsi tubuh secara keseluruhan. Aplikasi GiziCare memiliki utilitas dalam melakukan monitoring terhadap masukan macronutrient yang dimiliki oleh modul food diary. Namun, dalam pengembangannya pengguna harus memasukkan daftar makanan secara manual, sehingga diperlukan penerapan teknologi computer-vision untuk melakukan otomasi untuk mengatasi permasalahan tersebut. Agar dapat mendeteksi multiple object dalam gambar makanan dengan mudah dan efisien implementasinya menggunakan teknologi object-detection. Objek makanan dideteksi menggunakan model yang telah dilatih menggunakan arsitektur YOLOv11. Pengerjaan tugas akhir dimulai dari pengumpulan data berupa foto sebagai data input untuk training model, melakukan pre-processing terhadap data yang dikumpulkan, training model dan pengembangan sub-modul image recognition pada modul food diary. Dataset yang digunakan dalam pengerjaan tugas akhir adalah Indonesian Food Image (Ashari, 2023) dan indonesian-food-segmentation Computer Vision dataset(indonesianfoodsegmentation,2025). Terdapat beberapa metode pre-processing yang digunakan dalam pengerjaan tugas akhir yaitu data cleaning, format conversion, anotasi, data splitting dan merging.Data dipecah dengan susunan 80% file gambar setiap kelas sebagai data latih (train), 10% sebagai data validasi (validation) dan 10% sebagai data uji (test). Dalam proses training, model dilatih menggunakan pre-trained model YOLOv11 ukuran nano dan dilatih menggunakan parameter epoch=100. Tugas akhir berhasil melakukan implementasi pengembangan sub-modul image recognition pada modul food diary dengan evaluasi model menggunakan mAP-50 sebesar 0,946 dan mAP-95 sebesar 0,808 terhadap semua kelas.
======================================================================================================================================
Macronutrients are the primary nutrients required by the body in large amounts to provide energy and support overall bodily functions. The nutrition monitoring application has the utility of monitoring the intake of macronutrients within the food diary module. However, during its development, users were required to manually input a list of foods, thus necessitating the application of computer-vision technology to automate this process and address the issue. To enable e!cient and e”ective detection of multiple objects in food images, the implementation utilizes object-detection technology. Food objects are detected using a model trained with the YOLOv11 architecture. The final project work began with data collection in the form of images as model input, followed by data pre-processing, model training, and the development of the image recognition sub-module within the food diary module. The datasets used in this study are the Indonesian Food Image dataset Ashari2023IndonesianFood and the Indonesian Food Segmentation Computer Vision Dataset IndonesianFoodSegmentation2025. Several pre-processing methods were applied, including data cleaning, format conversion, annotation, data splitting, and merging. The data were split such that 80% of image files per class were used as training data (train), 10% as validation data (validation), and 10% as testing data (test). During the training process, the model was trained using a pre-trained YOLOv11 nano model with a total of 100 epochs. This final project successfully implemented the development of the image recognition sub-module within the food diary module, achieving model evaluation results of mAP-50 of 0.946 and mAP-95 of 0.808 across all classes

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Computer Vision, Object-detection, Macronutrient, Machine Learning, mAP-50, mAP-95, Modul Food Diary, Sub-Modul image recognition, YOLOv11.
Subjects:	A General Works > AI Indexes (General) A General Works > AI Indexes (General) T Technology > T Technology (General) > T57.5 Data Processing T Technology > T Technology (General) > T58.8 Productivity. Efficiency
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Informatics Engineering > 55201-(S1) Undergraduate Thesis
Depositing User:	Lihardo Marson Purba
Date Deposited:	30 Jan 2026 02:36
Last Modified:	30 Jan 2026 02:36
URI:	http://repository.its.ac.id/id/eprint/131255

Actions (login required)

View Item