Human Activity Recognition using Smart Phone Sensors
Vinish Kumar1, Anuj Sharma2
1Vinish Kumar, Department of Computer Science and Engineering, Institute of Technology, Roorkee (Uttarakhand), India.
2Anuj Sharma, Department of Computer Science and Engineering, Institute of Technology, Roorkee (Uttarakhand), India.
Manuscript received on 05 August 2019 | Revised Manuscript received on 28 August 2019 | Manuscript Published on 05 September 2019 | PP: 445-450 | Volume-8 Issue-2S7 July 2019 | Retrieval Number: B10820782S719/2019©BEIESP | DOI: 10.35940/ijrte.B1082.0782S719
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: In the new age technology, there exists many smart devices, which are using human activity data in reshaping the modern world dynamics of every aspect of our life be it health trackers, smartphones, intelligent systems. One futuristic concept is the connected devices that are way more efficient, adaptive, responsive and flexible to any conditions and reacts according to the data. For some connected devices to work more efficiently, human activity data is required. This data can be used to make devices smarter and using it can be useful in solving many problems of healthcare, efficient surveillance. Our work is an effort in efficient surveillance and using deep learning models, we detect the presence of human activities in different environments and use the data to analyze better to have efficient and effective surveillance. Many different models of deep learning model are used in our work from the likes of CNN (Convolutional Neural Networks) to LSTM (Long Short-Term Memory Networks. The data collected is from sensors’ data which is present in the mobile and can make the predictions about various activities like sitting, walking, jumping and some other human activities. The prime focus here is to detect various canonical activities that are not given to the system.
Keywords: Ubiquitous Sensing, Machine Learning, CNN, Human Activity Recognition, Tensor flow, Deep-Learning.
Scope of the Article: Pattern Recognition