Wyniki wyszukiwania - BazTech

1

Recognition of sports exercises using inertial sensor technology

Krutz Pascal, Rehm Matthias, Schlegel Holger, Dix Martin

Applied Computer Science

|

2023

|

Vol. 19, no 1

152--163

EN

Supervised learning as a sub-discipline of machine learning enables the recognition of correlations between input variables (features) and associated outputs (classes) and the application of these to previously unknown data sets. In addition to typical areas of application such as speech and image recognition, fields of applications are also being developed in the sports and fitness sector. The purpose of this work is to implement a workflow for the automated recognition of sports exercises in the Matlab® programming environment and to carry out a comparison of different model structures. First, the acquisition of the sensor signals provided in the local network and their processing is implemented. Realised functionalities include the interpolation of lossy time series, the labelling of the activity intervals performed and, in part, the generation of sliding windows with statistical parameters. The preprocessed data are used for the training of classifiers and artificial neural networks (ANN). These are iteratively optimised in their corresponding hyper parameters for the data structure to be learned. The most reliable models are finally trained with an increased data set, validated and compared with regard to the achieved performance. In addition to the usual evaluation metrics such as F1 score and accuracy, the temporal behaviour of the assignments is also displayed graphically, allowing statements to be made about potential causes of incorrect assignments. In this context, especially the transition areas between the classes are detected as erroneous assignments as well as exercises with insufficient or clearly deviating execution. The best overall accuracy achieved with ANN and the increased dataset was 93.7 %.

2

Human activity recognition using improved complete ensemble EMD with adaptive noise and long short-term memory neural networks

Altuve Miguel, Lizarazo Paula, Villamizar Javier

Biocybernetics and Biomedical Engineering

|

2020

|

Vol. 40, no. 3

901--909

EN

The recognition of human activities is a topic of great relevance due to its wide range of applications. Different approaches have been proposed to recognize human activities, ranging from the comparison of signals with thresholds to the application of deep and machine learning techniques. In this work, the classification of six human activities (walking, walking downstairs, walking upstairs, standing, sitting, and lying down) is performed using bidirectional LSTM networks that exploit intrinsic mode function (IMF) representation of inertial signals. Records with inertial signals (accelerometer and gyroscope) of 2.56 s, available at the UCI Machine Learning Repository, were collected from 30 subjects using a smartphone. First, inertial signals were standardized to take them to the same scale and were decomposed into IMF using the improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN). IMF were then segmented (split) into nine segments of 1.28 s with 12.5% overlap and introduced to a first network with four outputs to identify the dynamic activities and the statics as a single class called ‘‘statics’’, giving 98.86% accuracy. Then, the non-segmented IMF of the records assigned to the statics class were introduced to a second network to classify their three activities, giving an accuracy of 88.46%. In total, 92.91% accuracy was obtained to classify the six human activities. This performance is because ICEEMDAN allowed the extraction of information that was embedded in the signal, and the segmentation of the IMF allowed the network to discriminate between static and dynamic activities.

3

Influence of accelerometer signal preprocessing and classification method on human activity recognition

Kupryjanow A., Kaszuba K., Czyżewski A.

Elektronika : konstrukcje, technologie, zastosowania

|

2010

|

Vol. 51, nr 3

18-23

EN

A study of data preprocessing influence on accelerometer-based human activity recognition algorithms is presented. The frequency band used to filter-out the accelerometer signals and the number of accelerometers involved were considered in terms of their influence on the recognition accuracy. In the tests four methods of classification were used: support vector machine, decision trees, neural network, k-nearest neighbor.

PL

W artykule przedstawiono wpływ przetwarzania wstępnego sygnału przyspieszenia na skuteczność rozpoznawania aktywności ruchowych. Przeanalizowano zależność filtracji sygnałów oraz ilości zastosowanych czujników na skuteczność klasyfikacji. W badaniach wykorzystano cztery różne klasyfikatory: maszynę wektorów wsparcia, drzewa decyzyjne, sztuczne sieci neuronowe oraz klasyfikator najbliższego sąsiada.

4

Rozpoznawanie kategorii ruchu ludzkiego na podstawie analizy sygnałów pochodzących z trójosiowych czujników przyspieszenia

Kupryjanow A., Kaszuba K.

Zeszyty Naukowe Wydziału Elektrotechniki i Automatyki Politechniki Gdańskiej

|

2009

|

Nr 26

53-56

PL

W artykule przedstawiono procedurę rejestracji sygnałów przyspieszenia pochodzących z czujników biomedycznych Shimmer, sposób ich rozmieszczenia na ciele oraz opisano klasyfikator pozwalający na rozpoznawanie wybranych kategorii ruchu ludzkiego. W części eksperymentalnej artykułu zbadano wpływ filtracji dolnoprzepustowej sygnałów na skuteczność rozpoznawania typu aktywności ruchowej.

EN

In many scientific fields, especially medicine, information about human activity is crucial. The analysis of acceleration data coming from the sensors mounted on human’s limbs and trunk allows automatic classification of patients’activities (e.g. sitting, walking, getting up, etc). In this paper, a neural network based motion activity classifier and the procedure for recording signals from accelerometers are described. Owing to a very fast development of microcontrollers, it is now possible to create devices which enable real-time recording and transmission of signals from accelerometers. Today’s miniaturization enables the integration of accelerometers, microcontrollers and Bluetooth transmitters into a single matchbox-size device. Research carried out by Intel resulted in highly integrated devices and software platforms designed for networks of sensors which communicate wirelessly. Small size and weight of such devices as well as low energy consumption make the montage of sensors on a human body technically possible and comfortable for patients. The research proved that the localization of sensors on a human body has a great impact on the accuracy of motion type recognition. Many experiments addressing this subject were conducted, and finally an optimal sensors configuration was chosen. A group of 16 healthy people was observed. The acceleration signals were sampled with the frequency of 51,2 Hz whereas the G force was set within the range of 0 to 4. The 64 sample windows with the 32 samples overlap were used for the analysis. For each window, a set of parameters was extracted, which allowed the classification of signals. The research showed that the motion classifier based on neural networks ensures satisfying efficiency of motion type classification. Activity recognition was performed off-line. The accuracy of detection depended on the type of activity and the way the activity was performed. It turned out that for a better network training and testing, a greater number of signals must be collected.