Smart Sensing for Food Nutrition and Safety: A Low-Cost, Portable Spectral Approach

PhD Thesis Proposal Defence


Title: "Smart Sensing for Food Nutrition and Safety: A Low-Cost, Portable 
Spectral Approach"

by

Miss Haiyan HU


Abstract:

The growing demand for accessible food nutrition and safety analysis has 
highlighted limitations in conventional approaches, which often rely on 
expensive laboratory equipment, expert operation, or visual-based methods 
with poor chemical specificity. This thesis addresses these challenges by 
developing novel low-cost spectral sensing systems that leverage 
near-infrared (NIR) spectroscopy and smartphone integration to enable 
ubiquitous food analysis. We overcome critical challenges including low 
signal-to-noise ratio in affordable hardware, sparse spectral sampling 
limitations, consumer imaging sensor constraints, and smartphone optical 
system deficiencies. By integrating physics-aware algorithms with optimized 
hardware designs, this work demonstrates that laboratory-grade food analysis 
can be democratized through portable, <$100 solutions. The proposed 
approaches bridge the gap between specialized instrumentation and daily 
nutritional monitoring, showing significant improvements in accuracy (22-38% 
over baselines), cost-effectiveness (10-100x reduction), and usability 
across solid/liquid foods under real-world conditions.


Date:                   Wednesday, 23 April 2025

Time:                   11:30am - 1:30pm

Venue:                  Room 2408
                        Lifts 17/18

Committee Members:      Prof. Qian Zhang (Supervisor)
                        Prof. Mo Li (Chairperson)
                        Dr. Xiaomin Ouyang