System and method for feature level foreground segmentation
US7916944B2 · kind B2 · utility
Assignee
Inventors
Key dates
| Filing date | Jan 31, 2007 |
| Grant date | Mar 29, 2011 |
| Priority date | — |
| Expiry date | Jan 28, 2030 |
Classification
- Technology area (CPC G)Physics
- CPC primaryG06T2207/10016
- WIPO fieldComputer technology
- WIPO sectorElectrical engineering
Abstract
Foreground segmentation in real world dynamic scenes, including under various lighting and shadow conditions, is disclosed. It may be used with one or multiple cameras for various automated tasks, such as classification of moving object, tracking moving objects, and event detection in various indoor or outdoor environments. Pixel to pixel subtraction is performed on each frame, followed by a feature-level based foreground segmentation to properly validate the foreground pixels. In this step, for each pixel in the image, a neighborhood of pixels is selected, and the aggregate change in the neighborhood image is used to classify foreground and background pixels. Normalized cross correlation is then applied to the neighborhood of each pixel that was confirmed to be foreground pixel.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.