| Presentation | 2017-01-26 Development of species classification system by the deep learning using the broadband split-beam Atsushi Kinjo, Ikuo Matsuo, Tomohito Imaizumi, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | pecies classification using an acoustic sounder is important for fisheries. With schools of mixed species, it is necessary to isolate individual fish echoes in order to classify individual fish species from echoes. A broadband signal, which offered the advantage of high-range resolution, was applied for this purpose, and the positions of fish were estimated using the split-beam system. The target strength (TS) spectra of individual fish echoes could be computed from isolated echoes and estimated positions. In this paper, we proposed the fish classification system by using these TS spectra. We examined two methods, which were Nearest Neighbor Algorithm (NNA) and Deep Learning as machine learning. Subject species were chub mackerel (Scomber japonicas), Japanese jack mackerel (Trachurus japonicus) and Japanese anchovy (Engraulis japonicas). It was found that the classification rates using the Deep Learning were superior to those using the NNA and the classification rate using Deep Learning was about 68%. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | Machine Learning / Deep Learning / Fish classification / Split-Beam, |
| Paper # | US2016-108 |
| Date of Issue | 2017-01-18 (US) |
| Conference Information | |
| Committee | EA / US |
|---|---|
| Conference Date | 2017/1/25(2days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | Doshisha Univ. |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | [Joint Meeting on Acoustics and Ultrasonics Subsociety] Engineering/Electro Acoustics, Ultrasonic and Underwater Acoustics, and Related Topics |
| Chair | Mitsunori Mizumachi(Kyushu Inst. of Tech.) / Nobuyuki Endo(Kanagawa Univ.) |
| Vice Chair | Yoichi Haneda(Univ. of Electro-Comm.) / Suehiro Shimauchi(NTT) / Yoichi Itoh(Nihon Univ.) / Minoru Kurosawa(Tokyo Inst. of Tech.) |
| Secretary | Yoichi Haneda(KDDI R&D Labs.) / Suehiro Shimauchi(Akita Pref. Univ.) / Yoichi Itoh(Univ. of Electro-Comm.) / Minoru Kurosawa(Univ. of Toyama) |
| Assistant | Shigeto Takeoka(Shizuoka Inst. of Science and Tech.) / TREVINO Jorge(Tohoku Univ.) / Takeshi Morita(Univ. of Tokyo) |
| Paper Information | |
| Registration To | Technical Committee on Engineering Acoustics / Technical Committee on Ultrasonics |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Development of species classification system by the deep learning using the broadband split-beam |
| Sub Title (in English) | |
| Keyword(1) | Machine Learning |
| Keyword(2) | Deep Learning |
| Keyword(3) | Fish classification |
| Keyword(4) | Split-Beam, |
| 1st Author's Name | Atsushi Kinjo |
| 1st Author's Affiliation | Tohoku Gakuin University(TGU) |
| 2nd Author's Name | Ikuo Matsuo |
| 2nd Author's Affiliation | Tohoku Gakuin University(TGU) |
| 3rd Author's Name | Tomohito Imaizumi |
| 3rd Author's Affiliation | Fisheries Research Agency National Research Institute of Fisheries Engineering(FRA,NRIFE) |
| Date | 2017-01-26 |
| Paper # | US2016-108 |
| Volume (vol) | vol.116 |
| Number (no) | US-419 |
| Page | pp.pp.173-178(US), |
| #Pages | 6 |
| Date of Issue | 2017-01-18 (US) |