Please use this identifier to cite or link to this item:
|Title:||Adaptive Background Mixture Models for Motion Detection and Scale Invariant Feature Transform for Object Recognition|
|Abstract:||Background subtraction is a very common technique for segmentation of moving objects in image sequences. Generally objects in the foreground (cars, humans, etc.) are a video’s regions of interest. After a few preprocessing steps and region of interest localization this method is used for motion detection. The difference in the approaches to this problem lies in the type of background model used and the procedure used to update the model. One such approach is modeling each pixel as a mixture of Gaussians and using an on-line approximation to update the model. Each pixel is classified based on whether the Gaussian distribution which represents it most effectively is considered part of the background model. This results in a stable, real-time motion detection which reliably deals with lighting changes, repetitive motions from clutter, and long-term scene changes. For object recognition Scale-invariant feature transform is used to detect and describe local features in images. These features can be used to perform robust matching invariant to image scale and rotation between different views of an object or scene. This approach to recognition can robustly identify objects among clutter and occlusion while achieving near real-time performance.|
|Appears in Collections:||03. EE|
Files in This Item:
|Adaptive Background Mixture Models for Motion Detection and Scale Invariant Feature Transform for Object Recognition.pdf||1.69 MB||Adobe PDF||View/Open Request a copy|
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.