Please use this identifier to cite or link to this item:
Title: Activity recognition
Authors: Dong, Wen
Keywords: DRNTU::Engineering::Computer science and engineering::Computing methodologies::Image processing and computer vision
Issue Date: 2016
Abstract: Computer vision is a technological subject to generate numerical or symbolic information by including methods for acquiring, processing, analyzing, understanding images and video sequences. With the rapid development of it, computer vision technology gradually shows its importance in many different fields due to more and more applications interact with it. For example, computer vision is the key component of surveillance systems which can analyze the normal and abnormal actions from the detected video. The world around us is full of human action since we are moving all the days. We cannot process anything without moving. And we could easily acquire and differentiate other peoples’ action through our vision. But all these things are difficult for computer because what is more important besides collect actions is to understand the context and reason behind it. Therefore, more efforts need to be put in it if we want to achieve more. Nowadays, Motion History Image (MHI) is the basis representation of action which needs to be presented. It is a static image template which is a method used to identify movements within a row of images in time. It is commonly used in this field because of it is simple to implement and also effective in describing actions. Besides MHI, Local Binary Pattern (LBP) which is a type of feature used for classification in computer vision is also commonly applied in this field. It is famous of its low complexity of computation and also improves the detection performance on some datasets.
Rights: Nanyang Technological University
Fulltext Permission: restricted
Fulltext Availability: With Fulltext
Appears in Collections:EEE Student Reports (FYP/IA/PA/PI)

Files in This Item:
File Description SizeFormat 
Final FYP Report.pdf
  Restricted Access
1.69 MBAdobe PDFView/Open

Page view(s)

Updated on Jun 23, 2021


Updated on Jun 23, 2021

Google ScholarTM


Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.