Real time human activity recognition in control and unconditional environment

Real time human recognition in controlled and uncontrolled environment is the research project. We design a automated system. Optical flow base motion features are extracted. Noise in the videos creates a problem for extraction of region of interest. Improve the problem of

Project Title

Project Area of Specialization

Artificial Intelligence

Project Summary

Real time human recognition in controlled and uncontrolled environment is the research project. We design a automated system. Optical flow base motion features are extracted. Noise in the videos creates a problem for extraction of region of interest. Improve the problem of noise. we extract human from video frames. Design Deep learning based frame work along with motion features.

In our project, we will improve surveillance systems by introducing human activity recognition real time with controlled and uncontrolled environment.

Project Objectives

We work on real-time human activities reorganization in controlled and uncontrolled environment. We detect the motion of human in real-time. We have problem of occlusion, change in human variation, similarities among different action, clatter background, in controlled environment we will control that kind of problems but in uncontrolled environment we will focus on that problems and reduce that problem as much as possible.

Project Implementation Method

Benefits of the Project

CCTV has been used daily for many purposes such as crime investigation, traffic control, chemical process record, production control, and security surveillance. Public places such as offices, hospitals, shops, tourism resorts and also houses install CCTV for Security surveillance. In its development, the use of CCTV has transformed from passive surveillance into integrated intelligent surveillance system.

Recognizing human behavior in an automated manner is essential in many ambient intelligence applications such as smart homes, health monitoring and assistance applications, emergency services, and transportation assistance services. It is foreseen that soon smart environments that interact with the people according to their specialized needs will be become an inseparable part of daily life. From the last fewer airs, with the increase in video cameras, massive amount of data is received for surveillance which is almost impossible for human operators to analyze and respond. Thus, we are in need to develop an automated system which efficiently analyzes the video/image data and respond as per requirements.

Security camera systems, such as Closed-circuit Television (CCTV) are widely used and can be found in places where monitoring is needed. The cheaper price and the increasing use of CCTVs in many areas are making it to be easily deployable. However, this traditional CCTV requires human to monitor the scene continuously. It is very costly and inefficient to have human monitoring the scene. Monitoring activity is dull in nature and the process of analyzing series of videos afterward can also be time-consuming. Therefore, a system with human activity recognition is needed for this purpose.

In our project, we will improve these Surveillance systems by introducing human activity recognition real time with controlled and uncontrolled environment.

Technical Details of Final Deliverable

We design a automated system. Optical flow base motion features are extracted. Noise in the videos creates a problem for extraction of region of interest. Improve the problem of noise. we extract human from video frames. Design Deep learning based frame work along with motion features.

In our project we have following problems;

•Noise

•Change in human variation

•Similarities among different action

•Clutter background

•Useful features extraction

In our project, we will improve these Surveillance systems by introducing human activity recognition real time with controlled and uncontrolled environment.