Human Fall Detection Algorithm Design Based on Sensor Fusion and Multi-threshold Comprehensive Judgment

Abstract

The use of a single method of acceleration threshold discrimination cannot fully characterize the change in human fall behavior, which can easily result in misjudgment. In this paper, we propose a human fall detection algorithm that combines human posture, support vector machine (SVM), and quadratic threshold decision. Firstly, a large number of human posture data are collected through a six-axis inertial measurement module (MPU6050). A fall detection model is established through filtering preprocessing, eigenvalue extraction, classification, and SVM training. Secondly, a first-level threshold determination is performed through a wearable wristband device. When a suspected fall occurs, six eigenvalues will be captured and uploaded to a cloud platform to trigger second-level SVM fall judgments. By matching the eigenvalues with the fall detection model, it can be determined accurately whether a fall has taken place. The experimental results show that the fall detection has a recognition rate of 92.2%, a false rate of 3.593%, and missing rate of 2.187%, which can better distinguish other nonfall actions.

Research on Pose Recognition Algorithm for Sports Players Based on Machine Learning of Sensor Data

“…Taking basketball as a benchmark, the completion of basketball movements is mainly done through the overall movement of the player's upper and lower limbs in a coordinated manner, so when identifying basketball movements, the upper and lower limb movements need to be discussed separately [16,27,31]. In the process of data acquisition, the collected data of upper limb movements and lower limb movements are discussed and recognized separately according to the different positions of sensor nodes placed on the body.…”

“…In the above parametric environment, the 3D point data in the video are extracted using the Google tracking-related API of Kinect SDK [27,28], the video mainly includes the gymnast's competition video and Weizmann human behavior database; the ratio of training data and test data is about 1 : 1, where the relevant gymnastic postures are mainly walking, jumping, prone, squatting, standing, and raising hands.…”

A data fusion-based method for detecting difficult yoga postures

“…The human posture detection algorithm is mainly used to determine and classify the recognized human skeleton point data. 18 and 25 human skeleton point output formats are provided by Openpose, the difference between the two is mainly in the number of foot key points, because the foot key point information is not high for the final classification result image, but the increase of key points will greatly affect the recognition speed [12] . Therefore, the format of 18 key points is chosen here, which can increase the recognition speed as much as possible while ensuring the accuracy.…”

Detecting Medication-Taking Gestures Using Machine Learning and Accelerometer Data Collected via Smartwatch Technology: Instrument Validation Study

“…Independent reports [13,[28][29][30][31][32] have also confirmed the usability of smartwatches to detect other human motion behaviors, such as eating, physical activity, and foot movement. In the last decade, inspired by the introduction of smart wearable devices, human activity recognition has expanded to include activities such as cigarette smoking [20,22], falls [33][34][35], and sleep [36,37]. Sleep activity has been studied further using sensor data obtained from electroencephalograms and electromyogram devices to develop neural network models [38].…”

“…This problem can be easily overcome with the presence of a sensing device on each hand. Although not common presently, the arrival of smart wristbands, rings, and other forms of wearable devices is likely to provide a more complete picture of a person's daily activities [27,30,33,35,69].…”