| Presentation | 2003/2/14 Application of Recurrent Neural Network for Vehicle Detection Keiichirou INAGAKI, Shozo SATO, Taizo UMEZAKI, |
|---|---|
| PDF Download Page | PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | In this paper, a method of detecting rear vehicle with single camera using the Recurrent Neural Networks (RNN) model is proposed. The conventional vehicle detection methods have been successful in good visibility condition, but often fail to detect in the case that the captured image quality is reduced by the varied optical environment. In addition, the shapes of the vehicles view becomes to large and changes, as the target is approaching to the camera. In the vehicle detection by image sensor, it is important to solve the problem described previously. An RNN of which the output has dependency on the past network states learns the consecutive scenes in that the target vehicle is projected in the proposed method. We evaluated this method for the consecutive image frames containing the frames under the reduced visibility condition. The detected rate 93.4% is obtained at the 624 unlearning image frames in the experimental results. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Recurrent Neural Network / Vehicle Detection / ITS / Back-Propagation Through TIme |
| Paper # | PRMU2002-218 |
| Date of Issue |
| Conference Information | |
| Committee | PRMU |
|---|---|
| Conference Date | 2003/2/14(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 | Pattern Recognition and Media Understanding (PRMU) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Application of Recurrent Neural Network for Vehicle Detection |
| Sub Title (in English) | |
| Keyword(1) | Recurrent Neural Network |
| Keyword(2) | Vehicle Detection |
| Keyword(3) | ITS |
| Keyword(4) | Back-Propagation Through TIme |
| 1st Author's Name | Keiichirou INAGAKI |
| 1st Author's Affiliation | Faculty of Engineering, Chubu University() |
| 2nd Author's Name | Shozo SATO |
| 2nd Author's Affiliation | B&E research Center, Nihon Fukushi University |
| 3rd Author's Name | Taizo UMEZAKI |
| 3rd Author's Affiliation | Faculty of Engineering, Chubu University |
| Date | 2003/2/14 |
| Paper # | PRMU2002-218 |
| Volume (vol) | vol.102 |
| Number (no) | 652 |
| Page | pp.pp.- |
| #Pages | 6 |
| Date of Issue | |