Scribd
Upload
39 views1 page

Module 4

Uploaded by

kyakrnahetujhe
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
39 views1 page

Module 4

Uploaded by

kyakrnahetujhe
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 1

Module 4

1. Explain background subtraction and how it is used to identify moving objects in video

frames.

Ans: Bckground subtraction is a widely used approach which is used for detecting moving objects
from static cameras. The method used is comparing all frames and pixels with the static background
image. In simple words set the background and check what changes are occurring.
The steps are:

Background Modelling: A background model is created using the constant parts of the scene. It is the
task of extracting the static image from the video. Also, it can be mathematically represented as
probability distribution for each pixel in the image. The Gaussian Mixture Madel is usually used for
background modelling , where each pixel is modelled as several Gaussian distributions.

Foreground detection: The next step is foreground detection. Here the foreground is detetcted by
comparing the current pixel value with the background model. Using a threshold value a pixel is
classifies as foreground.

Background Update: There may be some changes in the background, so to adapt to these changes
the background model should also get updated. Running average or exponential moving average is
used to update the mean and variance values of pixels.

2. Describe adaptive background modeling and its advantages over simple background

subtraction.

Ans: Adaptive background modelling is a better version of simple background modelling. In simple
background subtraction , the background the assumed to static but in adaptive modelling the
background model keeps updating to adjust the changes in the scene.

Advantages :

Adjustment to changes

Better handling of lighting and shadows

3. Implement a Gaussian Mixture Model (GMM) for background subtraction on a video

sequence and evaluate its effectiveness.

Ans:

You might also like