/Background-Segmentor-OpenCv

Motion detector based on moving average background subtraction method implemented in C++ with OpenCV library

Primary LanguageC++

BackGround-ForeGround-Segmentor

Motion detection based on moving average background subtraction method implemented in C++ with OpenCV library

Overview

This is BackGround-ForeGround-Segmentor based on moving average background subtraction method implemented in C++ with OpenCV library, the algorithm simply model the background by taking the moving average of pixels.
mask = (1 - alpha) * background + alpha * frame

Algorithm Steps:

1- Convert frame to gray-level image  
2- Filter gray image with Gaussian Filter with kernal of size 7*7  
3- Compute difference between image and background (if first frame set first frame to be background)  
4- Apply threshold to foreground image (the difference result)  
5- Accumulate background with equ: background = (1 - alpha) * background + alpha * frame

Results image

alt text alt text
alt text alt text alt text