Automatic Food Intake Assessment Using Camera Phones

Obesity is becoming an epidemic phenomenon in most developed countries. The fundamental cause of obesity and overweight is an energy imbalance between calories consumed and calories expended. It is essential to monitor everyday food intake for obesity prevention and management. Existing dietary assessment methods usually require manually recording and recall of food types and portions. Accuracy of the results largely relies on many uncertain factors such as user's memory, food knowledge, and portion estimations. As a result, the accuracy is often compromised. Accurate and convenient dietary assessment methods are still blank and needed in both population and research societies. In this thesis, an automatic food intake assessment method using cameras, inertial measurement units (IMUs) on smart phones was developed to help people foster a healthy life style. With this method, users use their smart phones before and after a meal to capture images or videos around the meal. The smart phone will recognize food items and calculate the volume of the food consumed and provide the results to users. The technical objective is to explore the feasibility of image based food recognition and image based volume estimation. This thesis comprises five publications that address four specific goals of this work: (1) to develop a prototype system with existing methods to review the literature methods, find their drawbacks and explore the feasibility to develop novel methods; (2) based on the prototype system, to investigate new food classification methods to improve the recognition accuracy to a field application level; (3) to design indexing methods for large-scale image database to facilitate the development of new food image recognition and retrieval algorithms; (4) to develop novel convenient and accurate food volume estimation methods using only smart phones with cameras and IMUs. A prototype system was implemented to review existing methods. Image feature detector and descriptor were developed and a nearest neighbor classifier were implemented to classify food items. A reedit card marker method was introduced for metric scale 3D reconstruction and volume calculation. To increase recognition accuracy, novel multi-view food recognition algorithms were developed to recognize regular shape food items. To further increase the accuracy and make the algorithm applicable to arbitrary food items, new food features, new classifiers were designed. The efficiency of the algorithm was increased by means of developing novel image indexing method in large-scale image database. Finally, the volume calculation was enhanced through reducing the marker and introducing IMUs. Sensor fusion technique to combine measurements from cameras and IMUs were explored to infer the metric scale of the 3D model as well as reduce noises from these sensors.

Life cycle assessment of trayless dining

Universities in the United States are applying more sustainable approaches to their dining service operations. "The increase in social consciousness and environmental stewardship on college campuses has spurred an array of new and innovative sustainability programs"(ARAMARK Higher Education 2008). University residence dining is typically cafeteria style, with students using trays to carry food. Studies report that food served without trays substantially reduces food waste and water and electrical consumption associated with washing trays. Commonly, these reported results are estimates and not measurements taken under actual operating conditions. This study utilizes measurements recorded under actual dining service conditions in student residence halls at Michigan Technological University to develop the following: 1) operational-specific data on the issues and potential savings associated with a conversion to trayless dining and 2) life cycle assessment (LCA) cost and environmental impact analyses comparing dining with and without trays. For the LCA, the entire life cycle of the system is considered, from the manufacturing to the usage and disposal phases. The study shows that trayless dining reduces food waste because diners carry less food. The total savings for the diner shifts when not using trays for the standard academic year (205 days), with an average number of 700 diners, is 7,032 pounds of food waste from the pre-rinse area (33% reduction) and 3,157 pounds of food waste from the pan washing area (39% reduction). In addition, for each day of the study, the diners consumed more food during the trayless portion of the experiment. One possible explanation for the increased food consumption during this short duration study could be that the diners found it more convenient to eat the extra food on their plate rather than carrying it back for disposal. The trayless dining experiment shows a reduction in dishwasher water, steam, and electrical consumption for each day of the study. The average reduction of dishwasher water, steam, and electrical consumption over the duration of the study were 10.7%, 9.5%, and 6.4% respectively. Trayless dining implementation would result in a decrease of 4,305 gallons of consumption and wastewater discharge, 2.87 mm BTU of steam consumption, and 158 kWh of electrical consumption for the dinner shift over the academic year. Results of the LCA indicate a total savings of $190.4 when trays are not used during the dinner shift. Trayless dining requires zero CO2 eq and cumulative energy demand in the manufacturing stage, reductions of 1005 kg CO2 eq and 861 MJ eq in the usage phase, and reductions of 6458 kg CO2 eq and 1821 MJ eq in the end of the life cycle.