LanguageTerm (authority = ISO 639-3:2007); (type = text)
English
Abstract (type = abstract)
Background: The school lunch environment is a prime target for increasing a child’s consumption of fresh fruits and vegetables. Schools are using smarter lunchroom strategies to facilitate healthy choices. However, there is an increasing concern about food waste, especially at school food services. Plate waste at school lunch is used to assess menu performance and meals acceptance using a variety of methodologies. The gold standard for measuring plate waste is the weighing method which is time consuming and costly. This has led researchers to search for alternatives.
Objective: The study aims to test the feasibility and validate the accuracy of a digital image recognition model as a tool to quantify aggregate vegetable waste and compare it against the gold standard “weighing method” in a middle school.
Design: The study was divided in two phases. In phase I, images and weights of the salad plate pre and post consumption were recorded. The model was trained using these data to test the feasibility of model for predicting food classes and estimating physical weights of food. In Phase II, digital images and weights of the salad plates pre and post consumption were recorded and run through the trained model. Aggregate vegetable waste was calculated as the difference between the recorded weights, and the predicted weights assessed through the model.
Results: In Phase I, the image recognition model achieved overall classification accuracy of 85.7% of predicting nine food classes. The mean rank for recorded pre weight was (1.61 g + 0.43 g) and predicted pre weight was (1.01 g + 0.99 g) The feasibility results suggested that there was a significant difference between the recorded and predicted weights (p=0.009). In Phase II, the mean rank for recorded pre weight was (1.63 g + 0.45 g) and predicted weight was (1.73 g + 0.22 g) and did not elicit a statistically significant difference as compared to manually recorded weight (p = 0.341). The mean rank for recorded post weight was (0.62 g + 0.77 g) and weight predicted by the image recognition model was (0.63 g+ 0.80 g) with no statistically significant difference between the two (p=0.619). The mean rank for recorded plate waste was (0.68 % + 0.83%) and plate waste determined by the predicted weights by the image recognition model was (0.72 % + 0.91%). The Wilcoxon signed-rank test showed no statistically significant difference (p=0.177) in plate waste calculated using two methods.
Conclusion: The main findings from this study were that the image recognition model was feasible and accurate for identifying food classes and quantifying vegetable plate waste in a self-serve salad bar in a middle school and did not differ significantly from the gold standard weighing method. This study supports the use
of a digital image recognition model as a valid tool to semi automate data collection and estimate food waste.
Subject (authority = RUETD)
Topic
Food Science
Subject (authority = local)
Topic
Plate Waste
Subject (authority = LCSH)
Topic
Food waste -- Measurement
RelatedItem (type = host)
TitleInfo
Title
Rutgers University Electronic Theses and Dissertations
Identifier (type = RULIB)
ETD
Identifier
ETD_10264
PhysicalDescription
Form (authority = gmd)
InternetMediaType
application/pdf
InternetMediaType
text/xml
Extent
1 online resource (xi, 46 pages) : illustrations
Note (type = degree)
M.S.
Note (type = bibliography)
Includes bibliographical references
RelatedItem (type = host)
TitleInfo
Title
School of Graduate Studies Electronic Theses and Dissertations
Identifier (type = local)
rucore10001600001
Location
PhysicalLocation (authority = marcorg); (displayLabel = Rutgers, The State University of New Jersey)
I hereby grant to the Rutgers University Libraries and to my school the non-exclusive right to archive, reproduce and distribute my thesis or dissertation, in whole or in part, and/or my abstract, in whole or in part, in and from an electronic format, subject to the release date subsequently stipulated in this submittal form and approved by my school. I represent and stipulate that the thesis or dissertation and its abstract are my original work, that they do not infringe or violate any rights of others, and that I make these grants as the sole owner of the rights to my thesis or dissertation and its abstract. I represent that I have obtained written permissions, when necessary, from the owner(s) of each third party copyrighted matter to be included in my thesis or dissertation and will supply copies of such upon request by my school. I acknowledge that RU ETD and my school will not distribute my thesis or dissertation or its abstract if, in their reasonable judgment, they believe all such rights have not been secured. I acknowledge that I retain ownership rights to the copyright of my work. I also retain the right to use all or part of this thesis or dissertation in future works, such as articles or books.