Food recognition using enhanced squirrel search optimisation algorithm and convolutional neural network

Megha Chopra and Archana Purwar

International Journal of Data Analysis Techniques and Strategies, 2023, vol. 15, issue 3, 238-254

Abstract: Owning to the sedentary lifestyle, dietary assessment has become a significant research area. Automated food assessment initiates with food classification. Image classification commences with segmentation. Apparently, thresholding is the elemental method to perform segmentation. Although, there are many ways to optimise the solution of multi-level thresholding, this paper proposes a squirrel search algorithm (SSA)-based optimised solution for multi-level thresholding. It applies convolutional neural network (CNN) to recognise food images. Further, the paper has proposed a new enhanced squirrel search algorithm (ESSA) to improve the food recognition accuracy. The results show that ESSA improves the performance of image segmentation and classification. The performance of the proposed algorithm is evaluated using food datasets UEC-256 and UEC-100 and accuracy of 83.1% and 82.1% was obtained respectively. Proposed algorithm is also compared with existing work taken under this study and it has been observed that it outperformed them.

Keywords: food recognition; squirrel search algorithm; SSA; enhanced squirrel search algorithm; ESSA; thresholding; segmentation. (search for similar items in EconPapers)
Date: 2023
References: Add references at CitEc
Citations:

Downloads: (external link)
http://www.inderscience.com/link.php?id=133023 (text/html)
Access to full text is restricted to subscribers.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:ids:injdan:v:15:y:2023:i:3:p:238-254

Access Statistics for this article

More articles in International Journal of Data Analysis Techniques and Strategies from Inderscience Enterprises Ltd
Bibliographic data for series maintained by Sarah Parker ().