Accepted for/Published in: JMIR Formative Research
Date Submitted: Jul 23, 2019
Date Accepted: Oct 2, 2020
Use of different food image recognition platforms in dietary assessment: a comparison study.
ABSTRACT
Background:
In the domain of dietary assessment, there has been an increasing amount of critique on memory-based techniques such as food frequency questionnaires or 24h recalls. A non-memory-based technique is logging pictures of consumed food followed by an automatic image recognition analysis that provides information on type and amount of food in the picture.
Objective:
Evaluation of the performance of different food image recognition platforms in the context of dietary assessment.
Methods:
A comparison study of currently available image recognition platforms was conducted. A variety of foods and beverages was photographed in a range of standardized settings. All pictures (n=185) were fed into selected recognition platforms (n=7) and estimates were saved. Accuracy was determined along with totality (=successful estimation of most dish components) of the estimate in the case of multiple component dishes.
Results:
None of the platforms were capable of estimating the amount of food. Top 1 accuracies ranged from 63% for Calorie Mama API® to 9% for Google Vision API®. Top 5 accuracies ranged from 88% for Calorie Mama API® to 24% for Google Vision API®. The totality of top 5 estimates ranged from 71% for Foodvisor to less than 20% for Google Vision API®.
Conclusions:
The platforms perform poorly while others perform decently. Nevertheless, important obstacles such as the accurate estimation of food quantity need to be overcome before these platforms can be used as a real alternative for traditional dietary assessment methods. Clinical Trial: NA
Citation
Request queued. Please wait while the file is being generated. It may take some time.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.