The week's pick
Random Articles
Reseach Article
NutriSnap: Mobile-based Food Recognition with Caloric and Macronutrient Estimation using MobileNetV2 and YOLOv8n
| International Journal of Computer Applications |
| Foundation of Computer Science (FCS), NY, USA |
| Volume 187 - Number 93 |
| Year of Publication: 2026 |
| Authors: Andrei Carl L. Castro, Nathan Sheary G. Muñoz, Neo Jezer A. Pare, Joey S. Aviles |
10.5120/ijca2026926609
|
Andrei Carl L. Castro, Nathan Sheary G. Muñoz, Neo Jezer A. Pare, Joey S. Aviles . NutriSnap: Mobile-based Food Recognition with Caloric and Macronutrient Estimation using MobileNetV2 and YOLOv8n. International Journal of Computer Applications. 187, 93 ( Mar 2026), 31-37. DOI=10.5120/ijca2026926609
Abstract
This paper presents NutriSnap, a mobile-based food recognition system that estimates caloric and macronutrient content from user-captured images. The system integrates MobileNetV2 for image classification and YOLOv8n for object detection in a modular two-stage pipeline. Trained on the Food-101, UEC-256, Food2K, and a custom Filipino food dataset (Phil23), the MobileNetV2 model achieved a Top-1 validation accuracy of 73.19% across 189 food categories, with a macro-averaged F1-score of 0.73 and a weighted F1 of 0.73. The YOLOv8n model, trained using a three-stage fine-tuning approach with synthetic data augmentation, attained 96.1% precision, 92.9% recall, and 97.3% mAP50 on the validation set. Both models were converted to TensorFlow Lite (TFLite) and integrated into a Flutter-based Android application. Nutritional values are retrieved from the USDA FoodData Central and Philippine Food Composition Tables (PhilFCT) databases using a proportion-based formula keyed to user-entered serving weight. System usability was evaluated using the System Usability Scale (SUS) with 68 participants, yielding a mean score of 80.62, categorized as "Excellent." Comprehensive experimental results, including training convergence curves, per-class performance analysis, stage-wise detection metrics, and comparative evaluation against related works, demonstrate that the integrated pipeline is effective for real-time dietary monitoring on resource-constrained mobile devices.
References
- Agarwal, R., Choudhury, T., Ahuja, N. J., and Sarkar, T. 2023. Hybrid Deep Learning Algorithm-Based Food Recognition and Calorie Estimation. Journal of Food Processing and Preservation, 2023, 1–15.
- Amugongo, L. M., Kriebitz, A., Boch, A., and Lütge, C. 2022. Mobile Computer Vision-Based Applications for Food Recognition and Volume and Calorific Estimation: A Systematic Review. Healthcare, 11(1), 59.
- Food and Nutrition Research Institute. 2025. DOST-FNRI presents the latest PH nutrition situation. https://fnri.dost.gov.ph
- Gulzar, Y. 2023. Fruit Image Classification Model Based on MobileNetV2 with Deep Transfer Learning Technique. Sustainability, 15(3), 1906.
- Haque, R. U., Khan, R. H., Shihavuddin, A. S. M., Syeed, M. M. M., and Uddin, M. F. 2022. Lightweight and Parameter-Optimized Real-Time Food Calorie Estimation from Images Using CNN-Based Approach. Applied Sciences, 12(19), 9733.
- Ji, H., Chen, X., Bai, J., and Gong, C. 2025. Lightweight SCL-YOLOv8. Sensors, 25(16), 5147.
- Konstantakopoulos, F. S., Georga, E. I., and Fotiadis, D. I. 2024. A Review of Image-Based Food Recognition and Volume Estimation AI Systems. IEEE Reviews in Biomedical Engineering, 17, 136–152.
- Kumar, R. D., et al. 2021. Recognition of food type and calorie estimation using neural network. Journal of Supercomputing, 77(8), 8172–8193.
- Min, W., et al. 2023. Food2K: A Large-Scale Benchmark for Fine-Grained Food Recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence.
- Mixon, H., and Davis, M. E. 2020. Thinking about food: An analysis of calorie estimation accuracy. Journal of Integrated Social Sciences, 10(1), 102–125.
- Mohanty, S. P., et al. 2022. The Food Recognition Benchmark. Frontiers in Nutrition, 9, 875143.
- Moumane, K., et al. 2023. Food Recognition and Nutrition Estimation using MobileNetV2 CNN architecture and Transfer Learning. SITA 2023.
- NCD Risk Factor Collaboration. 2024. Worldwide trends in underweight and obesity from 1990 to 2022.
- Rokhva, S., Teimourpour, B., and Soltani, A. H. 2024. Computer vision in the food industry: Accurate, real-time food recognition with MobileNetV2. Food and Humanity, 3, 100378.
- Sajith, R., et al. 2023. Nutrient Estimation from Images of Food for Diet Management in Diabetic Patients. WCONF 2023.
- Shao, Z., et al. 2022. Towards the Creation of a Nutrition and Food Group Based Image Database. Procedia Computer Science.
- Shen, Z., et al. 2020. Machine Learning Based Approach on Food Recognition and Nutrition Estimation. Procedia Computer Science, 174, 448–453.
- Vera, N., et al. 2024. A Practical Computer Vision-Based Advanced Nutritional Analyzer for Adult Diets. Procedia Computer Science, 251, 192–199.
- Zheng, J., et al. 2024. Artificial Intelligence Applications to Measure Food and Nutrient Intakes: Scoping Review. JMIR, 26, e54557.
- Zhou, J., et al. 2018. Calorie Estimation From Pictures of Food: Crowdsourcing Study. Interactive Journal of Medical Research, 7(2), e17.
- Bangor, A., Kortum, P., and Miller, J. 2009. Determining what individual SUS scores mean: Adding an adjective rating scale. Journal of Usability Studies, 4(3), 114–123.
Index Terms
Keywords