| Presentation | 2019-03-05 An optimization method of wireless communication bandwidth in Drone object recognition using edge computing PengFei Sun, Akihiro Nakao, |
|---|---|
| PDF Download Page |  PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | Real-time object detection is considered challenging in many UAV applications, such as detection, surveillance, search and rescue, etc. There have been many real-time object recognition models based on deep learning, which however require high computational resources. Generally, the cloud and edge servers are adopted to perform computing tasks based on images captured by the drone cameras. The most challenging problem with this approach is that, high-definition images are captured and transmitted to the cloud server, the need for network bandwidth is extremely high for real-time object detection, the service by multiple drones is not to scaled out. We suppose the object-recognition is performed via deep neutral network, which is usually applied to the object features abstracted from images as intermediary data. For this reason, our proposed method splits a deep neural network model and executes a part of computation to obtain intermediary data, which is supposed to be smaller than the size of original images. Our approach preserves the accuracy of object recognition at a high level while reducing the volume of transferred wireless data. Moreover, we use additional computational resources for extra layers of the neural network, yet lightweight enough to be executed on a drone, to compress the intermediary data so as to reduce the wireless network bandwidth. Our data compression is optimized for machine learning instead of human eyes so that it reduces the data transmission by as much as 64% compared with the general approach, while the accuracy of object detection is reduced only by 3% . |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | edge computing / AI / Object detection / drone / 5G |
| Paper # | NS2018-250 |
| Date of Issue | 2019-02-25 (NS) |
| Conference Information | |
| Committee | IN / NS |
|---|---|
| Conference Date | 2019/3/4(2days) |
| Place (in Japanese) | (See Japanese page) |
| Place (in English) | Okinawa Convention Center |
| Topics (in Japanese) | (See Japanese page) |
| Topics (in English) | General |
| Chair | Takuji Kishida(NTT-AT) / Yoshikatsu Okazaki(NTT) |
| Vice Chair | Kenji Ishida(Hiroshima City Univ.) / Akihiro Nakao(Univ. of Tokyo) |
| Secretary | Kenji Ishida(KDDI Research) / Akihiro Nakao(KDDI Research) |
| Assistant | / Kenichi Kashibuchi(NTT) |
| Paper Information | |
| Registration To | Technical Committee on Information Networks / Technical Committee on Network Systems |
|---|---|
| Language | ENG-JTITLE |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | An optimization method of wireless communication bandwidth in Drone object recognition using edge computing |
| Sub Title (in English) | |
| Keyword(1) | edge computing |
| Keyword(2) | AI |
| Keyword(3) | Object detection |
| Keyword(4) | drone |
| Keyword(5) | 5G |
| 1st Author's Name | PengFei Sun |
| 1st Author's Affiliation | The University of Tokyo(Tokyo Univ.) |
| 2nd Author's Name | Akihiro Nakao |
| 2nd Author's Affiliation | The University of Tokyo(Tokyo Univ.) |
| Date | 2019-03-05 |
| Paper # | NS2018-250 |
| Volume (vol) | vol.118 |
| Number (no) | NS-465 |
| Page | pp.pp.337-342(NS), |
| #Pages | 6 |
| Date of Issue | 2019-02-25 (NS) |