English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Automatic food detection is an emerging topic of interest due to its widearray of applications ranging from detecting food images on social mediaplatforms to filtering non-food photos from the users in dietary assessmentapps. Recently, during the COVID-19 pandemic, it has facilitated enforcing aneating ban by automatically detecting eating activities from cameras in publicplaces. Therefore, to tackle the challenge of recognizing food images with highaccuracy, we proposed the idea of a hybrid framework for extracting andselecting optimal features from an efficient neural network. There on, anonlinear classifier is employed to discriminate between linearly inseparablefeature vectors with great precision. In line with this idea, our methodextracts features from MobileNetV3, selects an optimal subset of attributes byusing Shapley Additive exPlanations (SHAP) values, and exploits kernel extremelearning machine (KELM) due to its nonlinear decision boundary and goodgeneralization ability. However, KELM suffers from the 'curse of dimensionalityproblem' for large datasets due to the complex computation of kernel matrixwith large numbers of hidden nodes. We solved this problem by proposing a novelmulticolumn kernel extreme learning machine (MCKELM) which exploited the k-dtree algorithm to divide data into N subsets and trains separate KELM on eachsubset of data. Then, the method incorporates KELM classifiers into parallelstructures and selects the top k nearest subsets during testing by using thek-d tree search for classifying input instead of the whole network. Forevaluating a proposed framework large food/non-food dataset is prepared usingnine publically available datasets. Experimental results showed the superiorityof our method on an integrated set of measures while solving the problem of'curse of dimensionality in KELM for large datasets.

Novel Multicolumn Kernel Extreme Learning Machine for Food Detection via Optimal Features from CNN