Unique Action Identifier by Using Magnetometer, Accelerometer and Gyroscope: KNN Approach

Abstract

In today's world, where technology is advancing every single day, new methodologies are being developed, and are brought in everyday use making our lives simpler, faster, safer, and powerful. Similarly, Human Activity Recognition (HAR) is getting more popular with all the revolutions made in the technologies. Sensor Network Technology is used in industrial applications, smart homes and system. A massive amount of data can be obtained from these sensors which are linked to the human body. Recognition of Human Activities using these sensors, and wearable technologies has been actively studied. Behavior Recognition seeks to distinguish one or more people's activities and goals through a collection of observations on the actions and environmental conditions of the person. Health surveillance, aged treatment, and plenty of other domains can be used to automatically understand the behavioral context. An existing dataset consisting of 10 subjects (5 females, 5 males) is being used in the paper, which incorporates both young and old volunteers between 19 and 60 years of old with weights ranging from 55 to 85 kg. The dataset reflects motion data collected when subjects are engaged in 11 separate (static and dynamic) smart home activities: computer usage (1 min), telephone conversation (1 min), vacuum cleaning (1 min), book reading (1 min), TV watching (1 min), ironing (1 min), walking (1 min), exercise (1 min), cooking (1 min), drinking (20 times), hair brushing (1 min) (20 times). Most of the activities are similar because of the multi sensor environment which makes it more difficult. Using three tri axial IMU (inertial measurement unit), Magnetometer, Accelerometer, Gyroscope sensors attached to

Comparative Study Between Machine learning algorithms and feature ranking techniques on UI-PRMD dataset

“…[4] Shows the architecture of the Extra tree algorithm. Palimkar et al [26] implemented an Extra Tree classifier for Human Activity Recognition (HAR) using a combination of magnetometer, accelerometer, and gyroscope data. The classifier was able to accurately identify activities, even in dynamic environments, showing its robustness across a wide range of scenarios.…”

Skeletal Keypoint-Based Transformer Model for Human Action Recognition in Aerial Videos

“…Traditional methods for action recognition used various sensing modalities, including accelerometers, magnetometers, and gyroscopes, to capture body movements, frequency of motion, angles and orientation of body parts, velocity, and acceleration along with some other advance features [8]- [11]. Although these methods are computationally efficient, robust to noise and illumination changes, and easily implementable, they are limited in terms of their scalability, accuracy and adaptability as compared to the computer vision-based methods.…”

DriverSVT: Smartphone-Measured Vehicle Telemetry Data for Driver State Identification

“…The latter systems can use data collected by smartphones' sensors, such as a gyroscope and an accelerometer, which can be analyzed as indicators to detect aggressive driving such as sudden braking, speeding, and sudden turns. Palimkar et al [15] used machine learning algorithms to recognize different human activities using data collected from various sensors (magnetometer, accelerometer, and gyroscope) attached to the human body.…”