NIRSCam: A Mobile Near-Infrared Sensing System for Food Calorie Estimation

The calculation of calorie consumption is paramount importance for the human diet and health management. Most existing solutions use image-processing techniques to identify food type refer nutrition table compute total calorie, which quite challenging differentiate foods that look same but contain vastly different quantities calories. To address this issue, we propose leverage near-infrared spectroscopy (NIRS) derive concentration nutrients based on unique absorption spectrum foods. Instead using professional NIRS system bulky, expensive, impractical daily by nonexpert users, develop a low-cost portable commercial LEDs. As quality signals low-power LEDs relatively poor, carefully design modulation schemes interference elimination algorithms improve signal remove interference. Extensive experiments show NIRSCAM outperforms image-based baseline in achieving more accurate estimation, especially look-alike robust various environmental factors.