Feature detector-level fusion methods in food recognition

The analysis of dietary pattern is important in health-care to reduce the risk factors of getting diet-related chronic diseases. An automatic dietary pattern assessment via food recognition algorithm provide an alternative way to improve the overall quality of dietary pattern assessment. However, du...

Full description

Bibliographic Details
Main Authors: Razali @ Ghazali, Mohd Norhisham, Manshor, Noridayu
Format: Conference or Workshop Item
Language:English
Published: IEEE 2019
Online Access:http://psasir.upm.edu.my/id/eprint/36286/
http://psasir.upm.edu.my/id/eprint/36286/1/Feature%20detector-level%20fusion%20methods%20in%20food%20recognition.pdf
_version_ 1848848289907081216
author Razali @ Ghazali, Mohd Norhisham
Manshor, Noridayu
author_facet Razali @ Ghazali, Mohd Norhisham
Manshor, Noridayu
author_sort Razali @ Ghazali, Mohd Norhisham
building UPM Institutional Repository
collection Online Access
description The analysis of dietary pattern is important in health-care to reduce the risk factors of getting diet-related chronic diseases. An automatic dietary pattern assessment via food recognition algorithm provide an alternative way to improve the overall quality of dietary pattern assessment. However, due to very high variability of food images, the fusion of multiple type of features have become inevitable. The Bag of Features (BoF) model to encode the features are originally designed to consider a single type of features only. Therefore, this paper investigates the methods to fuse multiple features extracted from food images. Specifically, three fusion methods have been evaluated namely descriptor-level fusion, representation-level fusion and score-level fusion. The fusion is performed on the feature detector-level between Different of Hessian (DoH) and MSER detector. Speeded-up Robust Feature (SURF) is employed for feature description. The features are encoded by using k-means clustering and Support Vector Machine with linear kernel has been employed for classification. The evaluation has been performed on UECIOO-Food dataset.
first_indexed 2025-11-15T09:32:08Z
format Conference or Workshop Item
id upm-36286
institution Universiti Putra Malaysia
institution_category Local University
language English
last_indexed 2025-11-15T09:32:08Z
publishDate 2019
publisher IEEE
recordtype eprints
repository_type Digital Repository
spelling upm-362862020-06-15T07:29:33Z http://psasir.upm.edu.my/id/eprint/36286/ Feature detector-level fusion methods in food recognition Razali @ Ghazali, Mohd Norhisham Manshor, Noridayu The analysis of dietary pattern is important in health-care to reduce the risk factors of getting diet-related chronic diseases. An automatic dietary pattern assessment via food recognition algorithm provide an alternative way to improve the overall quality of dietary pattern assessment. However, due to very high variability of food images, the fusion of multiple type of features have become inevitable. The Bag of Features (BoF) model to encode the features are originally designed to consider a single type of features only. Therefore, this paper investigates the methods to fuse multiple features extracted from food images. Specifically, three fusion methods have been evaluated namely descriptor-level fusion, representation-level fusion and score-level fusion. The fusion is performed on the feature detector-level between Different of Hessian (DoH) and MSER detector. Speeded-up Robust Feature (SURF) is employed for feature description. The features are encoded by using k-means clustering and Support Vector Machine with linear kernel has been employed for classification. The evaluation has been performed on UECIOO-Food dataset. IEEE 2019 Conference or Workshop Item PeerReviewed text en http://psasir.upm.edu.my/id/eprint/36286/1/Feature%20detector-level%20fusion%20methods%20in%20food%20recognition.pdf Razali @ Ghazali, Mohd Norhisham and Manshor, Noridayu (2019) Feature detector-level fusion methods in food recognition. In: 2nd International Conference on Communication Engineering and Technology (ICCET 2019), 12-15 Apr. 2019, Nagoya, Japan. (pp. 134-138). 10.1109/ICCET.2019.8726897
spellingShingle Razali @ Ghazali, Mohd Norhisham
Manshor, Noridayu
Feature detector-level fusion methods in food recognition
title Feature detector-level fusion methods in food recognition
title_full Feature detector-level fusion methods in food recognition
title_fullStr Feature detector-level fusion methods in food recognition
title_full_unstemmed Feature detector-level fusion methods in food recognition
title_short Feature detector-level fusion methods in food recognition
title_sort feature detector-level fusion methods in food recognition
url http://psasir.upm.edu.my/id/eprint/36286/
http://psasir.upm.edu.my/id/eprint/36286/
http://psasir.upm.edu.my/id/eprint/36286/1/Feature%20detector-level%20fusion%20methods%20in%20food%20recognition.pdf