A Novel Kernel Density Estimation Background Subtraction Design
碩士 === 元智大學 === 電機工程學系 === 103 === In this thesis, we present an algorithm of kernel density estimation (KDE) for the foreground-background detection. Our method can be divided into three main parts. Firstly, we use Mean Integrated Squared Error (MISE) criteria to find the optimal bandwidth for the...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Others |
| Language: | zh-TW |
| Online Access: | http://ndltd.ncl.edu.tw/handle/42545992471234903834 |