| Presentation | 2000/11/10 Face Detection in Images Using Genetic Algorithms and Its Implementation onto an Evolvable Hardware Systems Koji Takashima, Taro Nakamura, Moritoshi Yasunaga, Ikuo Yoshihara, |
|---|---|
| PDF Download Page | PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | In this paper, we propose a new face-detection algorithm based on the kernel-based method instead of the conventional template matching. The algorithm estimates distribution function by superposition of the kernels. The kernel does not need complex calculation, thus it can shorten detection time. Moreover, it can detect multiple faces at once. We also propose extention of the kernel-based method, where each kernel size is optimized to approximate the distribution function more precisely. We make use of the genetic algorithms in order to otpimize the kernel sizes. The proposed algorithm has high parallelism and can be easily implemented onto the evolvable hardware(the hardware that combines reconfigurable LSIs(FPGAs) and evolutional algorithms). This hardware implementation approach leads to the high speed image detection systems with small hardware sizes. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Face Detection / Template Matching / Genetic Algorithms / Kernel-based Method / Evolvable Hardware |
| Paper # | PRMU2000-117 |
| Date of Issue |
| Conference Information | |
| Committee | PRMU |
|---|---|
| Conference Date | 2000/11/10(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) | Face Detection in Images Using Genetic Algorithms and Its Implementation onto an Evolvable Hardware Systems |
| Sub Title (in English) | |
| Keyword(1) | Face Detection |
| Keyword(2) | Template Matching |
| Keyword(3) | Genetic Algorithms |
| Keyword(4) | Kernel-based Method |
| Keyword(5) | Evolvable Hardware |
| 1st Author's Name | Koji Takashima |
| 1st Author's Affiliation | Tsukuba University() |
| 2nd Author's Name | Taro Nakamura |
| 2nd Author's Affiliation | Tsukuba University |
| 3rd Author's Name | Moritoshi Yasunaga |
| 3rd Author's Affiliation | Tsukuba University |
| 4th Author's Name | Ikuo Yoshihara |
| 4th Author's Affiliation | Miyazaki University |
| Date | 2000/11/10 |
| Paper # | PRMU2000-117 |
| Volume (vol) | vol.100 |
| Number (no) | 443 |
| Page | pp.pp.- |
| #Pages | 8 |
| Date of Issue | |