HMM-based Activity Recognition with a Ceiling RGB-D Camera

Liciotti, Daniele; Frontoni, Emanuele; Zingaretti, Primo; Bellotto, Nicola; Duckett, Tom

doi:10.5220/0006202305670574

Cited by 10 publications

(7 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The human activity recognition concept was first introduced in the 1990s. In the first phase, video-based recognition was studied [8,9]. Many applications for activity recognition have been introduced over the years, such as for healthcare, care of the elderly, to monitor daily living activities, and in security systems [10,11].…”

Section: Related Workmentioning

confidence: 99%

SARM: Salah Activities Recognition Model Based on Smartphone

et al. 2019

View full text Add to dashboard Cite

Alzheimer's is a chronic neurodegenerative disease that frequently occurs in many people today. It has a major effect on the routine activities of affected people. Previous advancement in smartphone sensors technology enables us to help people suffering from Alzheimer's. For people in the Muslim community, where it is mandatory to offer prayers five times a day, it may mean that they are struggling in their daily life prayers due to Alzheimer's or lack of concentration. To deal with such a problem, automated mobile sensor-based activity recognition applications can be supportive to design accurate and precise solutions with an objective to direct the Namazi (worshipper). In this paper, a Salah activities recognition model (SARM) using a mobile sensor is proposed with the aim to recognize specific activities, such as Al-Qayam (standing), Ruku (standing to bowing), and Sujud (standing to prostration). This model entails the collection of data, selection and placement of sensor, data preprocessing, segmentation, feature extraction, and classification. The proposed model will provide a stepping edge to develop an application for observing prayer. For these activities' recognition, data sets were collected from ten subjects, and six different features sets were used to get improved results. Extensive experiments were performed to test and validate the model features to train random forest (RF), K-nearest neighbor (KNN), naive Bayes (NB), and decision tree (DT). The predicted average accuracy of RF, KNN, NB, and DT was 97%, 94%, 71.6%, and 95% respectively.

show abstract

Section: Related Workmentioning

confidence: 99%

SARM: Salah Activities Recognition Model Based on Smartphone

et al. 2019

View full text Add to dashboard Cite

show abstract

“…Furthermore, researchers in Brulin et al and Yao used fuzzy logic techniques to recognize user activities while considering uncertainty of sensory data. Moreover, the probabilistic learning techniques have been exploited to deal with uncertainty, such as hierarchical hidden Markov (HMM), and its variations have been used in several studies . Nonprobabilistic methods, namely data mining techniques, have also been applied for activity recognition in smart environments.…”

Section: Literature Reviewmentioning

confidence: 99%

“…Moreover, the probabilistic learning techniques have been exploited to deal with uncertainty, such as hierarchical hidden Markov (HMM), and its variations have been used in several studies. [28][29][30] Nonprobabilistic methods, namely data mining techniques, have also been applied for activity recognition in smart environments. Those methods include frequent sequence mining, 31 frequent-periodic pattern mining, 32 and mining frequent temporal relationships.…”

Section: Literature Reviewmentioning

confidence: 99%

Evidence theory‐based framework for improving automation in home automation system

Chemoun

Gilg

Laghrouche

et al. 2018

Int J Communication

View full text Add to dashboard Cite

Summary An easy and better way of life is the topic of the day and stays an important theme for the future. New researches focusing on this aspect are increasingly emerging. In recent years, improving automation in home automation systems and making homes even smarter seem to be a major issue. Automation of daily life functioning of home devices makes life easier for inhabitants, and then saves them from manual adjustments. Home automation systems encompass different communication and networking technologies. They are embedded with sensor networks that perceive information about the environment and occupants. The collected data can mainly be processed to infer sequential patterns defining the home behavior. However, the imperfection, particularly uncertainty of sensor readings, is a significant factor to consider. Also, time is an important aspect when reasoning about home behavior, as each pattern (scenario) performs at a given time throughout the day. Evidence theory is recently gaining great attention. It is suitable for dealing with imperfection of sensor data and for including temporal features into the reasoning process to discover the appropriate pattern to launch. This paper presents a new framework based on evidence theory with the inclusion of time features, for improving automation in home automation systems. The effectiveness of the proposed approach is evaluated by conducting experiments on synthetic dataset. The results show the efficiency of the discovery process when multiple working sensors are available. The results demonstrate also that reliability and accuracy of the process improve significantly when temporal information is provided.

show abstract

“…Xu et al [5] presented a method to predict anomalies based on multiple one-class Support Vector Machine (SVM) models, offering an anomaly score for video sequences. In [6], instead, an RGB-D camera was used to detect Activities of Daily Living (ADLs) using Hidden Markov Models.…”

Section: Related Workmentioning

confidence: 99%

Entropy-based abnormal activity detection fusing RGB-D and domotic sensors

Fernández-Carmona

Coşar

Coppola

et al. 2017

2017 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI)

Self Cite

View full text Add to dashboard Cite

The automatic detection of anomalies in Active and Assisted Living (AAL) environments is important for monitoring the wellbeing and safety of the elderly at home. The integration of smart domotic sensors (e.g. presence detectors) and those ones equipping modern mobile robots (e.g. RGB-D cameras) provides new opportunities for addressing this challenge. In this paper, we propose a novel solution to combine local activity levels detected by a single RGB-D camera with the global activity perceived by a network of domotic sensors. Our approach relies on a new method for computing such a global activity using various presence detectors, based on the concept of entropy from information theory. This entropy effectively shows how active a particular room or environment's area is. The solution includes also a new application of Hybrid Markov Logic Networks (HMLNs) to merge different information sources for local and global anomaly detection. The system has been tested with a comprehensive dataset of RGB-D and domotic data containing data entries from 37 different domotic sensors (presence, temperature, light, energy consumption, door contact), which is made publicly available. The experimental results show the effectiveness of our approach and its potential for complex anomaly detection in AAL settings.

show abstract

HMM-based Activity Recognition with a Ceiling RGB-D Camera

Cited by 10 publications

References 25 publications

SARM: Salah Activities Recognition Model Based on Smartphone

SARM: Salah Activities Recognition Model Based on Smartphone

Evidence theory‐based framework for improving automation in home automation system

Entropy-based abnormal activity detection fusing RGB-D and domotic sensors

Contact Info

Product

Resources

About