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

Quite a few people in the world have to stay under permanent surveillance forhealth reasons; they include diabetic people or people with some other chronicconditions, the elderly and the disabled.These groups may face heightened riskof having life-threatening falls or of being struck by a syncope. Due tolimited availability of resources a substantial part of people at risk can notreceive necessary monitoring and thus are exposed to excessive danger.Nowadays, this problem is usually solved via applying Human ActivityRecognition (HAR) methods. HAR is a perspective and fast-paced Data Sciencefield, which has a wide range of application areas such as healthcare, sport,security etc. However, the currently techniques of recognition are markedlylacking in accuracy, hence, the present paper suggests a highly accurate methodfor human activity classification. Wepropose a new workflow to address the HARproblem and evaluate it on the UniMiB SHAR dataset, which consists of theaccelerometer signals. The model we suggest is based on continuous wavelettransform (CWT) and convolutional neural networks (CNNs). Wavelet transformlocalizes signal features both in time and frequency domains and after that aCNN extracts these features and recognizes activity. It is also worth notingthat CWT converts 1D accelerometer signal into 2D images and thus enables toobtain better results as 2D networks have a significantly higher predictivecapacity. In the course of the work we build a convolutional neural network andvary such model parameters as number of spatial axes, number of layers, numberof neurons in each layer, image size, type of mother wavelet, the order of zeromoment of mother wavelet etc. Besides, we also apply models with residualblocks which resulted in significantly higher metric values. Finally, wesucceed to reach 99.26 % accuracy and it is a worthy performance for thisproblem.

Human Activity Recognition using Continuous Wavelet Transform and Convolutional Neural Networks