Nutrition Meets Tech: An Automated Food Analysis Cameras

Abstract

A healthy and balanced diet plays a crucial role in preventing chronic diseases such as obesity, cardiovascular disorders, and cancer. However, conventional dietary assessment methods rely on self-reporting techniques that are time-consuming, inaccurate, and often associated with underreporting and low user adherence. Recent advances in artificial intelligence (AI), computer vision, and wearable sensing technologies have enabled automated and passive dietary monitoring solutions, addressing these limitations. This study integrates AI-based food detection, mobile health (mHealth) applications, automated quality control, and artificial vision systems for dietary assessment and food-related analysis. Egocentric image datasets acquired using a wearable camera device were analysed using convolutional neural network-based algorithms to automatically detect food-related activities in real-world environments. The datasets included images of free-living daily activities such as dining, cooking, shopping, and physical exercise. Cross-dataset evaluations demonstrated high accuracy, sensitivity, and specificity in classifying food and non-food images, even from low-quality wearable camera data, thereby reducing manual processing burden and privacy concerns. In addition, automated quality control frameworks using data-centric AI paradigms were explored to evaluate multiple quality factors of packaged food products. These frameworks integrated deep learning and traditional computer vision approaches, achieving rapid prediction speeds and high classification accuracy. Artificial vision systems, including hyperspectral imaging, further enhanced food quality assessment by detecting defects and internal features beyond human visual capabilities. Overall, AI-driven dietary monitoring and food quality assessment systems show strong potential to improve nutritional data accuracy, support effective mHealth interventions, reduce food waste, and enable scalable industrial applications. Continued advancements in AI and vision technologies are expected to significantly impact personalised nutrition, public health, and food system sustainability.