Human Activity and Motion Pattern Recognition within Indoor Environment Using Convolutional Neural Networks Clustering and Naive Bayes Classification Algorithms.

Ashraf Ali,Andrew Ware,Weam Samara,Omar A Saraereh,Doaa Alhaddad

doi:10.3390/s22031016

Abstract

Human Activity Recognition (HAR) systems are designed to read sensor data and analyse it to classify any detected movement and respond accordingly. However, there is a need for more responsive and near real-time systems to distinguish between false and true alarms. To accurately determine alarm triggers, the motion pattern of legitimate users need to be stored over a certain period and used to train the system to recognise features associated with their movements. This training process is followed by a testing cycle that uses actual data of different patterns of activity that are either similar or different to the training data set. This paper evaluates the use of a combined Convolutional Neural Network (CNN) and Naive Bayes for accuracy and robustness to correctly identify true alarm triggers in the form of a buzzer sound for example. It shows that pattern recognition can be achieved using either of the two approaches, even when a partial motion pattern is derived as a subset out of a full-motion path.

Highlights

This paper proposes an approach for classification detected motion patterns using
The features are extracted from detected motion according to spatio-temporal parameters that are reflected over the blueprint design of the building
Naive Bays algorithm is used to classify the detected motion pattern to be normal or abnormal motion depending on training data collected previously

Summary

Introduction

Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. Modern smart intruder alarm and Human Activity Recognition (HAR) systems consist of networks of integrated electronic devices and sensors connected to a centralised control unit to protect against intruders by distinguishing between legitimate and illegitimate activity. In contrast to conventional security systems that respond to a single sensor trigger, the intelligent system uses machine learning techniques and IoT (Internet of Things) sensor infrastructure to detect and classify complex motion patterns. Different types of sensors (such as the Passive Infrared Sensor PIR, motion sensor and the ultrasonic sensor) are often used to detect different movement patterns.

Methods

Results

Conclusion