Feature Space Reduction for Human Activity Recognition based on Multi-channel Biosignals

Human activity recognition (HAR) has been an active area in computer vision with a broad range of applications, such as education, security surveillance, and healthcare. HAR is a general time series classification problem. LSTMs are widely used for time series classification tasks. However, they work well with high-dimensional feature vectors, which reduce the processing speed of LSTM in real-time applications. Therefore, dimension reduction is required to create low-dimensional feature space. As it is experimented in previous study, LSTM with dimension reduction yielded the worst performance among other classifiers, which are not deep learning methods. Therefore, in this paper, a novel scale and rotation invariant human activity recognition system, which can also work in low dimensional feature space is presented. For this purpose, Kinect depth sensor is employed to obtain skeleton joints. Since angles are used, proposed system is already scale invariant. In order to provide rotation invariance, body relative direction in egocentric coordinates is calculated. The 3D vector between right hip and left hip is used to get the horizontal axis and its cross product with the vertical axis of global coordinate system assumed to be the depth axis of the proposed local coordinate system. Instead of using 3D joint angles, 8 number of limbs and their corresponding 3D angles with X, Y, and Z axes of the proposed coordinate system are compressed with several dimension reduction methods such as averaging filter, Haar wavelet transform (HWT), and discrete cosine transform (DCT) and employed as the feature vector. Finally, extracted features are trained and tested with LSTM (long short-term memory) network, which is an artificial recurrent neural network (RNN) architecture. Experimental and benchmarking results indicate that proposed framework boosts the performance of LSTM by approximately 30% accuracy in low-dimensional feature space.

Download Full-text

Feature Space Analysis for Human Activity Recognition in Smart Environments

2016 12th International Conference on Intelligent Environments (IE) ◽

10.1109/ie.2016.43 ◽

2016 ◽

Cited By ~ 3

Author(s):

Eris Chinellato ◽

David C. Hogg ◽

Anthony G. Cohn

Keyword(s):

Activity Recognition ◽

Human Activity ◽

Feature Space ◽

Human Activity Recognition ◽

Smart Environments ◽

Space Analysis

Download Full-text

Granular Region-oriented Fuzzy-Rough based kNN Improvization for Activity Recognition Modeling

Journal of Applied Information Science ◽

10.21863/jais/2015.3.1.005 ◽

2015 ◽

Vol 3 (1) ◽

Author(s):

Vijay Borges ◽

Wilson Jeberson

Keyword(s):

Pattern Recognition ◽

Activity Recognition ◽

Human Activity ◽

Test Object ◽

Feature Space ◽

Nearest Neighbors ◽

Human Activity Recognition ◽

Similarity Relations ◽

Fuzzy Similarity ◽

Knn Classifier

Activity recognition is a complex task of the Human Computer Interaction (HCI) domain with ever-increasing research interest. Human activity recognition has been specially addressed by the advances in pattern recognition. k-Nearest Neighbors(kNN) is a non-parametric classifier from pattern recognition theory, that mimics human decision making by taking previous experiences into consideration for segregating unknown objects. A novel fuzzy-rough model, based on granular computing for improvisation of the kNN classifier is proposed herewith. In this model, feature-wise fuzzy memberships are generated to fuzzify the feature space of the nearest neighbors of the test object. These neighbors fuzzified feature space are then aggregated into granules, based on their class-belongingness. From these, lower and upper approximation granules are generated using rough set theory to classify the test object. It is shown experimentally that this model outperforms the traditional kNN by 16.43% and Fuzzy-kNN by 10.25%, in the human activity recognition domain. Another novelty is in the efficient use of the fuzzy similarity relations in class-dependent granulated feature space, and, fuzzy-rough lower/upper approximations in the hybridization of the kNN classifier.

Download Full-text

Enhanced Human Activity Recognition Using Wearable Sensors via a Hybrid Feature Selection Method

Sensors ◽

10.3390/s21196434 ◽

2021 ◽

Vol 21 (19) ◽

pp. 6434

Author(s):

Changjun Fan ◽

Fei Gao

Keyword(s):

Feature Selection ◽

Activity Recognition ◽

Human Activity ◽

Inertial Sensors ◽

Wearable Sensors ◽

Feature Selection Method ◽

Feature Space ◽

Human Activity Recognition ◽

Wearable Electronics ◽

Benchmark Datasets

The study of human activity recognition (HAR) plays an important role in many areas such as healthcare, entertainment, sports, and smart homes. With the development of wearable electronics and wireless communication technologies, activity recognition using inertial sensors from ubiquitous smart mobile devices has drawn wide attention and become a research hotspot. Before recognition, the sensor signals are typically preprocessed and segmented, and then representative features are extracted and selected based on them. Considering the issues of limited resources of wearable devices and the curse of dimensionality, it is vital to generate the best feature combination which maximizes the performance and efficiency of the following mapping from feature subsets to activities. In this paper, we propose to integrate bee swarm optimization (BSO) with a deep Q-network to perform feature selection and present a hybrid feature selection methodology, BAROQUE, on basis of these two schemes. Following the wrapper approach, BAROQUE leverages the appealing properties from BSO and the multi-agent deep Q-network (DQN) to determine feature subsets and adopts a classifier to evaluate these solutions. In BAROQUE, the BSO is employed to strike a balance between exploitation and exploration for the search of feature space, while the DQN takes advantage of the merits of reinforcement learning to make the local search process more adaptive and more efficient. Extensive experiments were conducted on some benchmark datasets collected by smartphones or smartwatches, and the metrics were compared with those of BSO, DQN, and some other previously published methods. The results show that BAROQUE achieves an accuracy of 98.41% for the UCI-HAR dataset and takes less time to converge to a good solution than other methods, such as CFS, SFFS, and Relief-F, yielding quite promising results in terms of accuracy and efficiency.

Download Full-text

Human Activity Recognition in AAL Environments Using Random Projections

Computational and Mathematical Methods in Medicine ◽

10.1155/2016/4073584 ◽

2016 ◽

Vol 2016 ◽

pp. 1-17 ◽

Cited By ~ 34

Author(s):

Robertas Damaševičius ◽

Mindaugas Vasiljevas ◽

Justas Šalkevičius ◽

Marcin Woźniak

Keyword(s):

Activity Recognition ◽

Human Activity ◽

Assisted Living ◽

Ambient Assisted Living ◽

Feature Space ◽

Human Activity Recognition ◽

The State ◽

Random Projections ◽

Healthcare Applications ◽

Subject Identification

Automatic human activity recognition systems aim to capture the state of the user and its environment by exploiting heterogeneous sensors attached to the subject’s body and permit continuous monitoring of numerous physiological signals reflecting the state of human actions. Successful identification of human activities can be immensely useful in healthcare applications for Ambient Assisted Living (AAL), for automatic and intelligent activity monitoring systems developed for elderly and disabled people. In this paper, we propose the method for activity recognition and subject identification based on random projections from high-dimensional feature space to low-dimensional projection space, where the classes are separated using the Jaccard distance between probability density functions of projected data. Two HAR domain tasks are considered: activity identification and subject identification. The experimental results using the proposed method with Human Activity Dataset (HAD) data are presented.

Download Full-text