Video Surveillance Under The Guidance of Smt. D.Neelima M.Tech., Asst. Professor Submitted by G....

17
Video Surveillance Under The Guidance of Smt. D.Neelima M.Tech., Asst. Professor Submitted by G. Subrahmanyam Roll No: 10021F0013 M.C.A

description

Existing system: Motion detection in consequent images is nothing but the detection of the moving object in the scene. In video surveillance, motion detection refers to the capability of the surveillance system to detect motion and capture the events. There are two conventional approaches to moving object detection, background subtraction and optical flow.

Transcript of Video Surveillance Under The Guidance of Smt. D.Neelima M.Tech., Asst. Professor Submitted by G....

Page 1: Video Surveillance Under The Guidance of Smt. D.Neelima M.Tech., Asst. Professor Submitted by G. Subrahmanyam Roll No: 10021F0013 M.C.A.

Video Surveillance

Under The Guidance of Smt. D.Neelima M.Tech., Asst. Professor

Submitted byG. SubrahmanyamRoll No: 10021F0013 M.C.A

Page 2: Video Surveillance Under The Guidance of Smt. D.Neelima M.Tech., Asst. Professor Submitted by G. Subrahmanyam Roll No: 10021F0013 M.C.A.

Abstract:

Video pre-processing steps such as frame separation, binary operation, gray enhancement and filter operation were conducted.

Then the detection and extraction of moving object was carried out on images according to frame difference-based dynamic background refreshing method.

Page 3: Video Surveillance Under The Guidance of Smt. D.Neelima M.Tech., Asst. Professor Submitted by G. Subrahmanyam Roll No: 10021F0013 M.C.A.

Existing system:Motion detection in consequent images is nothing but the

detection of the moving object in the scene. In video surveillance, motion detection refers to the capability of the surveillance system to detect motion and capture the events.

There are two conventional approaches to moving object detection, background subtraction and optical flow.

Page 4: Video Surveillance Under The Guidance of Smt. D.Neelima M.Tech., Asst. Professor Submitted by G. Subrahmanyam Roll No: 10021F0013 M.C.A.

Proposed system:

Frame difference method is used for detecting the moving objects is extracted according to the differences among two or three continuous frames.

Frames are converted to grayscale images, now subtracting the background from the sequential frames for foreground detection.

Then morphological operations are applied and moving objects were shown with a rectangular box in the output.

Page 5: Video Surveillance Under The Guidance of Smt. D.Neelima M.Tech., Asst. Professor Submitted by G. Subrahmanyam Roll No: 10021F0013 M.C.A.

Modules: Extraction Gray scale image Subtraction Adjacent frame difference Method

Page 6: Video Surveillance Under The Guidance of Smt. D.Neelima M.Tech., Asst. Professor Submitted by G. Subrahmanyam Roll No: 10021F0013 M.C.A.

Extraction:

First insert the video file which is in AVI format and read video file. After reading the input video file, extracted the red, green and blue intensities separately.

Gray scale image:

Gray scale images are images without color, or achromatic images. The levels of a gray scale range from 0 (black) to 1 (white).

Page 7: Video Surveillance Under The Guidance of Smt. D.Neelima M.Tech., Asst. Professor Submitted by G. Subrahmanyam Roll No: 10021F0013 M.C.A.

Subtraction:

This proposed system dynamically extracting the background from incoming all video frames, it is subtracted from every subsequent frame and compared with the background threshold. If it is greater than the background threshold, it assumed as foreground otherwise it is background. The Background is updated in each and every frame.

Page 8: Video Surveillance Under The Guidance of Smt. D.Neelima M.Tech., Asst. Professor Submitted by G. Subrahmanyam Roll No: 10021F0013 M.C.A.

Adjacent frame difference Method:

In this method, moving object is extracted according to the differences among two or three continuous frames.

since the time interval between two images is quite short, illumination changes have little influence on difference images, so the detection is effective and stable.

Page 9: Video Surveillance Under The Guidance of Smt. D.Neelima M.Tech., Asst. Professor Submitted by G. Subrahmanyam Roll No: 10021F0013 M.C.A.

Hardware Requirements:

• Processor : Intel® Core™ i3 2.53 GHz / Above

• Ram : 2GB • Hard Disk : 120GB• Input Devices : Standard Keyboard & Mouse• Output Device : Monitor 15”

Page 10: Video Surveillance Under The Guidance of Smt. D.Neelima M.Tech., Asst. Professor Submitted by G. Subrahmanyam Roll No: 10021F0013 M.C.A.

Software Requirements:

Operating System : Windows XP / AboveSoftware : MatLab2011a

Page 11: Video Surveillance Under The Guidance of Smt. D.Neelima M.Tech., Asst. Professor Submitted by G. Subrahmanyam Roll No: 10021F0013 M.C.A.

Experimental Result:

The moving object detecting was adopted in an experiment on a short video. Difference processing was conducted on the two frames for the detection is effective and stable.

Page 12: Video Surveillance Under The Guidance of Smt. D.Neelima M.Tech., Asst. Professor Submitted by G. Subrahmanyam Roll No: 10021F0013 M.C.A.

Screenshots:

Persons in Motion input video

Page 13: Video Surveillance Under The Guidance of Smt. D.Neelima M.Tech., Asst. Professor Submitted by G. Subrahmanyam Roll No: 10021F0013 M.C.A.

Persons in Motion output video

Page 14: Video Surveillance Under The Guidance of Smt. D.Neelima M.Tech., Asst. Professor Submitted by G. Subrahmanyam Roll No: 10021F0013 M.C.A.

Persons in Motion output video

Page 15: Video Surveillance Under The Guidance of Smt. D.Neelima M.Tech., Asst. Professor Submitted by G. Subrahmanyam Roll No: 10021F0013 M.C.A.

Conclusion:

Video Surveillance can be seen from analysis and examples of the computer language Matlab.

It has the characteristics of simple programming, easy operation and high processing rate.

Page 16: Video Surveillance Under The Guidance of Smt. D.Neelima M.Tech., Asst. Professor Submitted by G. Subrahmanyam Roll No: 10021F0013 M.C.A.

References:Barron J, Fleet D, Beauchemin S. Performance of

optical flow techniques. International Journal of Computer Vision, 1994, 12 (1) :42 - 77

Y Du, F Chen, W Xu, Activity recognition through multi-scale motion detail analysis, Neurocomputing, 2008, 71(16-18): 3561-3574

A Lipton, H Fujiyoshi, Moving target classification and tracking from real - time video. Proc ofWACV'98, 1998: 8 – 14

[11] David Kuncicky, MatLAB Programming, Prentice Hall: New York, 2003

Page 17: Video Surveillance Under The Guidance of Smt. D.Neelima M.Tech., Asst. Professor Submitted by G. Subrahmanyam Roll No: 10021F0013 M.C.A.

Thank You