Background-subtraction is a Computer-Vision technique that given a frame sequence from a fixed camera, processes pixels and separates them into foreground and background pixels.
Background Subtraction is a type of image segmentation which goal is to separate the parts of the image that are invariant over time (background) from the objects that are moving or changing (foreground).
The simplest techniques use frame differencing and more advanced techniques require using statistical methods. For example, a moving leaf of a tree would be considered foreground using simple frame differencing but with a proper statistical method it can be considered background as the leaf is always there, moving periodically.
Methods:
- Frame differencing
- Background as the average or the median of the previous nframes.
- Background as the running average.
- Pfinder (one Gaussian)
- Mixture of K Gaussians
