Using acceleration measurements for activity recognition: An effective learning algorithm for constructing neural classifiers

Yang, Jhun-Ying; Wang, Jeen-Shing; Chen, Yen-Ping

doi:10.1016/j.patrec.2008.08.002

Cited by 302 publications

(158 citation statements)

References 16 publications

Supporting

Mentioning

155

Contrasting

Unclassified

Order By: Relevance

“…Heuristic features including SMA [10] and Inter axis correlation [20,13] were derived from a fundamental understanding of how a specific movement would produce a distinguishable sensor signal. For instance, there are obvious correlations, using Pearson correlation test, between left and right wrist movements during all golf swings, walking, pushing with two hands, weight lifting and clapping.…”

Section: Feature Selectionmentioning

confidence: 99%

Kinect vs. Low-cost Inertial Sensing for Gesture Recognition

Gowing

Ahmadi

Destelle

et al. 2014

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. In this paper, we investigate efficient recognition of human gestures / movements from multimedia and multimodal data, including the Microsoft Kinect and translational and rotational acceleration and velocity from wearable inertial sensors. We firstly present a system that automatically classifies a large range of activities (17 different gestures) using a random forest decision tree. Our system can achieve near real time recognition by appropriately selecting the sensors that led to the greatest contributing factor for a particular task. Features extracted from multimodal sensor data were used to train and evaluate a customized classifier. This novel technique is capable of successfully classifying various gestures with up to 91 % overall accuracy on a publicly available data set. Secondly we investigate a wide range of different motion capture modalities and compare their results in terms of gesture recognition accuracy using our proposed approach. We conclude that gesture recognition can be effectively performed by considering an approach that overcomes many of the limitations associated with the Kinect and potentially paves the way for low-cost gesture recognition in unconstrained environments.

show abstract

Section: Feature Selectionmentioning

confidence: 99%

Kinect vs. Low-cost Inertial Sensing for Gesture Recognition

Gowing

Ahmadi

Destelle

et al. 2014

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…In Table 2 some work did not report accurately on the measured sub-tasks. In situations where target activities such as running were not explicitly defined [113,131] it was assumed that authors meant Running and not Running(treadmill) (found in Table 5) since involvement of a treadmill has not been mentioned. Although the two running categories relate to the same concept, it is worth noting that running on a treadmill is di↵erent from free running due to the treadmill setting the pace -as opposed to self-selected pace.…”

Section: Locomotion Transitions and Posturementioning

confidence: 99%

Classification and suitability of sensing technologies for activity recognition

Woznowski

Kaleshi

Oikonomou

et al. 2016

Computer Communications

View full text Add to dashboard Cite

Wider availability of sensors and sensing systems has pushed research in the direction of automatic activity recognition (AR) either for medical or other personal benefits e.g. wellness or fitness monitoring. Researchers apply di↵erent AR techniques/algorithms and use a wide range of sensors to discover home activities. However, it seems that the AR algorithms are purely technology-driven rather than informing studies on the type and quality of input required. There is an expectation to over-instrument the environment or the subjects and then develop AR algorithms, where instead the problem should be approached from a di↵erent angle i.e. what sensors (type, quality and quantity) a given algorithm requires to infer particular activities with a certain confidence? This paper introduces the concept of activity recognition, its taxonomy and familiarises the reader with sub-classes of sensor-based AR. Furthermore, it presents an overview of existing health services Telecare and Telehealth solutions, and introduces the hierarchical taxonomy of human behaviour analysis tasks. This work is a result of a systematic literature review and it presents the reader with a comprehensive set of home-based activities of daily living (ADL) and sensors proven to recognise these activities. Apart from reviewing usefulness of various sensing technologies for homebased AR algorithms, it highlights the problem of technology-driven cycle of development in this area. IntroductionIn the last two decades sensors have become cheaper, smaller and widely available, residing at the edge of the Internet. Some such examples are wearable personal activity (PA) trackers (e.g. Fitbit, Nike+ FuelBand, etc.). However, the available commercial o↵-the-shelf (COTS) sensors are only capable of 'sensing' a small subset of user activities -mostly outdoor sport activities (type of activity, distance covered, time taken, etc.) and estimation of additional information such as energy expenditure (either kcal or self-crafted metrics e.g. Nike's fuel-points). However, a large part of our lives, and increasingly so in the advanced age, is spent in the home, yet very little is known about our activities and behaviour in there.We are surrounded by a multitude of sensing devices and Mark Weiser's vision of ubiquitous computing [120] is starting to materialise in the advances made in embedded networked systems currently addressed as the Internet of Things (IoT). The significant increase in devices streaming low-level information over the Web presents many new challenges. Whilst many researchers present this as a big data challenge, we believe that many of the environments and applications will require to justify the value and process relatively small data, making this a two-faceted problem requiring to consider the highly distributed, non-interoperable, small and relatively "lonely" data. E cient and accurate activity recognition (AR) algorithms are needed in order to make sense of this data and provide useful/actionable information and services in the human...

show abstract

“…DT is applied to classify twenty physical activities [16]. An ANN based approach [14] using acceleration features first separates the static (standing, sitting) and dynamic (walking, running) activities and then classifies the activities in each class, where Principal Component Analysis (PCA) is used to obtain the well performing features. ANN is also compared with auto generated and domain knowledge based DTs for activity classification, where auto generated DT shows better accuracy while ANN suffers from over fitting [23].…”

Section: Related Workmentioning

confidence: 99%

“…Activity segmentation is performed using different techniques, sliding windows [7], relative weighting of objects in adjacent activities [8] or pattern mining [9], just to name a few. Segmented activity instances are classified in activity classes using different learning models such as Hidden Markov Model (HMM) [10], Conditional Random Fields (CRF) [11], Naive Bayes (NB) [12], Support Vector Machine (SVM) [13], Artificial Neural Network (ANN) [14,15], and Decision Tree (DT) [16]. In activity classification, a false assignment could occur due to the unreliable nature of sensor data [17], incorrect execution of an activity [18], similar activities due to overlapping in features [19] or inability of a learning algorithm to assign the correct label [20].…”

Section: Introductionmentioning

confidence: 99%

Activity recognition in smart homes with self verification of assignments

2015

View full text Add to dashboard Cite

This is the accepted version of the paper.This version of the publication may differ from the final published version. Permanent repository link AbstractActivity recognition in smart homes provides valuable benefits in the field of health and elderly care by remote monitoring of patients. In health care, capabilities of both performing the correct recognition and reducing the wrong assignments are of high importance. The novelty of the proposed activity recognition approach lies in being able to assign a category to the incoming activity, while measuring the confidence score of the assigned category that reduces the false positives in the assignments. Multiple sensors deployed at different locations of a smart home are used for activity observations. For multi-class activity classification, we propose a binary solution using support vector machines, which simplifies the problem to correct/incorrect assignments. We obtain the confidence score of each assignment by estimating the activity distribution within each class such that the assignments with low confidence are separated for further investigation by a human operator. The proposed approach is evaluated using a comprehensive performance evaluation metrics. Experimental results obtained from nine publicly available smart home datasets demonstrate a better performance of the proposed approach compared to the state of the art.

show abstract

Using acceleration measurements for activity recognition: An effective learning algorithm for constructing neural classifiers

Cited by 302 publications

References 16 publications

Kinect vs. Low-cost Inertial Sensing for Gesture Recognition

Kinect vs. Low-cost Inertial Sensing for Gesture Recognition

Classification and suitability of sensing technologies for activity recognition

Activity recognition in smart homes with self verification of assignments

Contact Info

Product

Resources

About