23 December 2019 in IEEE Sensors Journal
In this paper, we present a machine learning model to estimate the blood pressure (BP) of a person using only his photoplethysmogram (PPG) signal. We propose algorithms to better detect some critical points of the PPG signal, such as systolic and diastolic peaks, dicrotic notch and inflection point. These algorithms are applicable to different PPG signal morphologies and improve the precision of feature extraction. We show that the logarithm of dicrotic notch reflection index, the ratio of low-to high-frequency components of heart rate (HR) variability signal, and the product of HR multiplied by the modified Normalized Pulse Volume (mNPV) are the key features in accurately estimating the BP using PPG signal. Our proposed method has achieved higher accuracies in estimating BP compared to the previously reported methods that only use PPG signal. For the systolic BP, the achieved correlation coefficient between the estimated values and the real values is 0.78, the mean absolute error of the estimated values is 8.22 mmHg, and their standard deviation is 10.38 mmHg. For the diastolic BP, the achieved correlation coefficient between the estimated and the real values is 0.72, the mean absolute error of the estimated values is 4.17 mmHg, and their standard deviation is 4.22 mmHg. The achieved results fall within Grade A for diastolic, Grade C for systolic and Grade B for mean BP based on BHS standard.
18 December 2017 in IEEE Sensors Journal
This article proposes a new method for viewinvariant action recognition that utilizes the temporal position of skeletal joints obtained by Kinect sensor. In this method, the actions are represented as sequences of several pre-defined poses. After pre-processing, which includes skeleton alignment and scaling, the appropriate feature vectors are obtained for recognizing and discriminating the pose of every frame by the proposed Fisherposes method. The proposed regularized Mahalanobis distance metric is used in order to recognize both the involuntary and highly made-up actions at the same time. Hidden Markov Model (HMM) is then used to classify the action related to an input sequence of poses. For taking into account the motion in the actions which are not separable by solely their temporal poses, histograms of trajectories are also proposed. The proposed action recognition method is capable of recognizing both the voluntary and involuntary actions, as well as posebased and trajectory-based ones with a high accuracy rate. The effectiveness of the proposed method is experimented on three publicly available datasets, TST fall detection, UTKinect, and UCFKinect datasets.
13 January 2021 in IEEE Sensors Journal
This paper proposes a novel 3D action recognition technique that uses time-series information extracted from depth image sequences for use in systems of human daily activity monitoring. To this end, each action is represented as a multi-dimensional time series, where each dimension represents the position variation of one skeleton joint over time. The time series is then mapped onto a vector space using Dynamic Time Warping (DTW) distance. Furthermore, to employ the correlation-distinctiveness relationship of the sequences in recognition, this vector space is remapped onto a discriminative space using the regularized Fisher method, where final decisions about the actions are made. Unlike other available methods, the time-warping used in the mapping strategy makes the feature space robust to temporal variations of the motion sequences. Moreover, our method eliminates the need for a complicated design method for extracting the static and dynamic information of a motion sequence. Furthermore, most existing methods treat all skeletal joints identically for different actions, while some joints are more discriminative to distinguish a specific action. Thanks to the nature of the proposed features, we propose to use a separate set of discriminative joints, called joint importance map for each class of action. Evaluation results on three well-known datasets, TST, UTKinect, UCFKinect, and NTU RGB+D show competitive performance with the state-of-the-art methods in human action recognition.
1 August 2018 in Journal of Visual Communication and Image Representation
6 May 2018 in Scientia Iranica
Human action recognition has been one of the most active fields of research in computer vision over the last years. Two dimensional action recognition methods are facing serious challenges such as occlusion and missing the third dimension of data. Development of depth sensors has made it feasible to track positions of human body joints over time. This paper proposes a novel method for action recognition which usestemporal 3D skeletal Kinect data. This method introduces the definition of body states and then every action is modeled as a sequence of these states. The learning stage uses Fisher Linear Discriminant Analysis (LDA) to construct discriminant feature space for discriminating the body states. Moreover, this paper suggests the use of the Mahalonobis distance as an appropriate distance metric for the classification of the states of involuntary actions. Hidden Markov Model (HMM) is then used to model the temporal transition between the body states in each action. According to the results, this method significantly outperforms other popular methods, with recognition (recall) rate of 88.64% for eight different actions and up to 96.18% for classifying the class of all fall actions versus normal actions.