2019

A Survey of IoT: Advances in Smart and Dynamic Environmental Monitoring

8th International Conference on Environmental Engineering, Science and Management

2019 | Conference paper

Contributors: Otanasap, Nuth; Chalermsuk, Chanintorn; Bungkomkhun, Pornpimol

Show less detail

URL

https://www.researchgate.net/publication/333967805_A_Survey_of_IoT_Advances_in_Smart_and_Dynamic_Environmental_Monitoring

Contributors

Otanasap, Nuth (Author)

Chalermsuk, Chanintorn (Author)

Bungkomkhun, Pornpimol (Author)

Abstract

Due to the growing of heavy industry around the world and the various chemical substances with pollution occurring from the daily life of people, freed into the natural environment have established dangerous solicitudes on the impacts of such chemicals not only on the human health but also on the ecosystem. Consequently, nations around the world have raised their directions in titles of monitoring as well as regulating and handling consequent infection. Over the last several years, people have developed dramatically with the rush of the manufacturing revolt in which a new production of wireless communication facilitates ubiquitous connectivity among things. As the advancement of the internet of things (IoT), increasingly useful applications can be realized in various enterprises for now. IoT is regularly perceived as practical things, broadly distributed, with low repository ability and processing potential, with the purpose of enhancing authenticity, enforcement, and preservation of the smart living and its foundations. Nowadays IoT systems have become intelligent than before. As these reason, the IoT will be applied more to support dynamic, commercial and reliable performance of the system based on various perspectives. This paper not only provides IoT physical characteristics, network architecture, and MAC protocols, but also presents the idea of smart environmental monitoring applications using IoT. The paper directs to a significant criterion change of how to monitor, sense and track dynamic phenomena in real-time in the environment. From this review, it is visible that there are remaining a few exciting opportunities and challenges on improvement and deployment of IoT for smart and dynamic environmental monitoring.

Fuzzy Inference Based Pre-impact Fall Detection System Using Dynamic Threshold

ASEAN Journal of Scientific and Technological Reports

2019 | Journal article

Contributors: Otanasap, Nuth; Boonbrahm, Poonpong

Show less detail

URL

https://ph02.tci-thaijo.org/index.php/tsujournal/article/view/133232

Contributors

Otanasap, Nuth (Author)

Boonbrahm, Poonpong (Author)

Abstract

Problems of falling are particularly vital safety in seniors. For these reasons, increasing a useful fall prevention approach is necessary to relieve the infliction of falls. This study focuses on a dynamic threshold model for real-time pre-impact fall detection that enables the falls to be identified before the body crashes to the ground. The velocity of head and chest position and the center of gravity of the subject body used for the feature combination classified by fuzzy inference for pre-impact fall detection. It only needs subjects to wear some tiny coin-size sensors that combine with the Kinect sensor, vision-based, without recording any data due to privacy issue. The dynamic threshold-based model with stereotypes suitable for an individual one, is applied for real-time fall and non-fall classification for the longest lead time of pre-impact fall detection. Moreover, the various kinds of integration of single, multiple, and triple Kinect combined with and without the wearable device are evaluated. The 14 rules of Sugeno fuzzy set defining the falling posture, movement transitions, and comparison of the different combination of devices are inferred first, whereas the final decision is produced through thinking and trigger on such fuzzy sets. The experimentation result found that the highest lead-time of pre-impact fall detection is 549.83 ms. However, the integration method that combined multiple Kinect with the wearable device can reduce camera overlapping and obscurity with the highest accuracy about 98.09 percent. Vice versa, the method using only multiple Kinect without the wearable device provide lower accuracy than 93.00 percent.

Pre-impact fall detection system using real time dynamic threshold and human body bounding box by multiple kinects

SAU JOURNAL OF SCIENCE & TECHNOLOGY

2019 | Journal article

Contributors: Otanasap, Nuth

Show less detail

URL

https://ph01.tci-thaijo.org/index.php/saujournalst/article/view/170324

Contributors

Otanasap, Nuth (Author)

Abstract

This work contributes on the fusion of multiple Kinect based skeletons, based on dynamic threshold and bounding box posture analysis which is the only research work reported so far. As the second leading cause of accidental death extensive is unintentional falls, which is a vital cause of personal harm, particularly with the venerable. Accordingly, many studies in healthcare are achieving on the improvement of the pre-fall detection system to secure the protection of those who are possible to be concerned. Pre-impact fall detection system has to overcome many difficulties to improve an effective system. Some of the particular difficulties are obtrusion, occlusion, and overlap in the vision-based method.

In this research, the purpose of adopting the bounding box and head velocity method compare with a real-time dynamic threshold is for analyzing the fall and non-fall incident accurately. Furthermore, the skeleton joint position provided by multiple Kinect viewpoints are utilized for the reason of resolving in obtrusion, occlusion, and overlap issues without demanding of markers. Though, the various fuzzy rule base methods also are applied for the final decision of lead time detection and triggering fall alarm. The demonstration of subjects completion is performed 1,100 actions were included 700 times for activities of daily living and 400 times for falling. All activities performed by ten different volunteers, seven healthy young males, and three healthy young females.

The results have shown that 98.55% of the proposed method is higher accurately detected. However, the proposed method provided the lowest specificity at 97.71%, vice versa it offered the highest sensitivity at 100.00%. It implies that during system provided higher accuracy and sensitivity in pre-impact fall detection, the recognized precision of normal activities will be reduced. Moreover, the multiple Kinect methods not only provide higher accuracy and sensitivity but also offer higher average lead-time as 505.86 ms.