Action recognition using optimized deep autoencoder and CNN for surveillance data streams of non-stationary environments

Action recognition is a challenging research area in which several convolutional neural networks (CNN) based action recognition methods are recently presented. However, such methods are inefficient for real-time online data stream processing with satisfied accuracy. Therefore, in this paper we propo...

Full description

Saved in:
Bibliographic Details
Published in:Future generation computer systems Vol. 96; pp. 386 - 397
Main Authors: Ullah, Amin, Muhammad, Khan, Haq, Ijaz Ul, Baik, Sung Wook
Format: Journal Article
Language:English
Published: Elsevier B.V 01.07.2019
Subjects:
Action recognition
Big data processing
Convolutional neural network
Machine learning
Non-stationary environment
Online data stream analysis
Optimized deep autoencoder
Action recognition
Convolutional neural network
Machine learning
Big data processing
Online data stream analysis
Optimized deep autoencoder
Non-stationary environment
ISSN:0167-739X, 1872-7115
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Action recognition is a challenging research area in which several convolutional neural networks (CNN) based action recognition methods are recently presented. However, such methods are inefficient for real-time online data stream processing with satisfied accuracy. Therefore, in this paper we propose an efficient and optimized CNN based system to process data streams in real-time, acquired from visual sensor of non-stationary surveillance environment. Firstly, frame level deep features are extracted using a pre-trained CNN model. Next, an optimized deep autoencoder (DAE) is introduced to learn temporal changes of the actions in the surveillance stream. Furthermore, a non-linear learning approach, quadratic SVM is trained for the classification of human actions. Finally, an iterative fine-tuning process is added in the testing phase that can update the parameters of trained model using the newly accumulated data of non-stationary environment. Experiments are conducted on benchmark datasets and results reveal the better performance of our system in terms of accuracy and running time compared to state-of-the-art methods. We believe that our proposed system is a suitable candidate for action recognition in surveillance data stream of non-stationary environments. •Action recognition in online data stream acquired from non-stationary surveillance.•Efficient CNN model is used for frame-level representation.•An optimized deep autoencoder is presented for learning sequences and squeezing high. dimensional features.•Investigated a non-linear learning approach for action recognition.•Iterative fine-tuning of the trained recognition model for newly accumulated data.
ISSN:0167-739X
1872-7115
DOI:10.1016/j.future.2019.01.029