| Presentation | 2008-07-15 Illumination Invariant Method using Cross-correlation and Square-difference for Still Object Detection Katsuya YAMASHITA, Kazuyuki ISO, Kumi JINZENJI, Hiroyuki ARAI, Akira KOJIMA, Hitoshi NAKAZAWA, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | This paper proposes a still-object detection method that is robust against background changes due to environmental illumination changes, the camera's AGC and AWB responses and so on. The object that remains still for some period is taken to be the foreground region. Our basic idea for foreground detection is to compare the weighted sum of the cross-correlation and square-difference of the region surrounding each pixel between the input image and the background image, which is calculated as the weighted average of the last few images and is updated sequentially. To improve the reliability of foreground detection, we also adopt the Gaussian model to determine the threshold for the background subtraction. Evaluation experiments using both synthesized images and real images prove the robustness and effectiveness of our proposed method. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | still object detection / cross-correlation / square-difference / gausian model / illumination variation |
| Paper # | IE2008-39,MVE2008-33 |
| Date of Issue |
| Conference Information | |
| Committee | MVE |
|---|---|
| Conference Date | 2008/7/7(1days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | |
| Chair | |
| Vice Chair | |
| Secretary | |
| Assistant | |
| Paper Information | |
| Registration To | Media Experience and Virtual Environment (MVE) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Illumination Invariant Method using Cross-correlation and Square-difference for Still Object Detection |
| Sub Title (in English) | |
| Keyword(1) | still object detection |
| Keyword(2) | cross-correlation |
| Keyword(3) | square-difference |
| Keyword(4) | gausian model |
| Keyword(5) | illumination variation |
| 1st Author's Name | Katsuya YAMASHITA |
| 1st Author's Affiliation | Cyber Space Laboratories, NTT Corporation() |
| 2nd Author's Name | Kazuyuki ISO |
| 2nd Author's Affiliation | Cyber Space Laboratories, NTT Corporation |
| 3rd Author's Name | Kumi JINZENJI |
| 3rd Author's Affiliation | Cyber Space Laboratories, NTT Corporation |
| 4th Author's Name | Hiroyuki ARAI |
| 4th Author's Affiliation | Cyber Space Laboratories, NTT Corporation |
| 5th Author's Name | Akira KOJIMA |
| 5th Author's Affiliation | Cyber Solution Laboratories, NTT Corporation |
| 6th Author's Name | Hitoshi NAKAZAWA |
| 6th Author's Affiliation | Cyber Space Laboratories, NTT Corporation |
| Date | 2008-07-15 |
| Paper # | IE2008-39,MVE2008-33 |
| Volume (vol) | vol.108 |
| Number (no) | 128 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |