Thorat, Ninad (2018) Human Activity Recognition Using Deep Learning. MTech thesis.
|PDF (Restricted upto 25/03/2021) |
Restricted to Repository staff only
In recent years, focus on the physical activities recognition has gained a lot of momentum due to the wearable device sensor. Compared to other types of visual sensory devices such as video camera has gain lot of popularity due to its nonintrusive nature. However, most of the activities recognition is done under certain laboratory and experimental conditions. These sensor has been tested on only small groups of people under certain conditions also
they carry some extra cost which is not possible in everyday life scenario. Alternatively, smartphone and smartwatches devices which have inertial measurement unit sensor (IMU)
in it can be used for robust activity recognition also abnormal activities can also be detected by it. In this research, we put forward our data collection techniques and compare itsperformance by using the neural network classifiers and deep learning techniques and its different types networks such as convolution neural network (CNN) and Long Short-Term Memory networks (LSTM) for robust human activity recognition.
|Item Type:||Thesis (MTech)|
|Uncontrolled Keywords:||IMU sensor; Neural networks ; Deep learning.|
|Subjects:||Engineering and Technology > Computer and Information Science > Image Processing|
|Divisions:||Engineering and Technology > Department of Computer Science Engineering|
|Deposited By:||IR Staff BPCL|
|Deposited On:||21 Mar 2019 17:34|
|Last Modified:||21 Mar 2019 17:34|
Repository Staff Only: item control page