TY - GEN
T1 - Low-Cost CNN Design for Intelligent Surveillance System
AU - Wei Yang, Liang
AU - Yen Su, Chung
N1 - Publisher Copyright:
© 2018 IEEE.
PY - 2018/11/1
Y1 - 2018/11/1
N2 - In today's world, video surveillance appears everywhere in our life. The challenge for a video surveillance system is to recognize the object of interest for analysis. The convolutional neural network (CNN) models achieve high accuracy on image recognition, but the models require powerful calculations. The models cannot be applied to most smart surveillance systems directly. In this paper, we propose a low-cost CNN design for the application of surveillance systems. Instead of using GPUs, we use a hardware accelerator called Neural Compute Stick (NCS) accompanied with the Rock64 to build the system. The NCS is a low-cost and low-power USB device, which has the advantages in the high-speed calculation of images. As a result, we use the NCS to load the Single Shot MultiBox Detector (SSD) network for human detection. Our system can get each detected image in 0.15 sec. It is six times faster than other single-board surveillance systems. Furthermore, the cost of building the real-time surveillance system is less than 100. Therefore, our system can achieve a low-cost and high-performance intelligent surveillance system.
AB - In today's world, video surveillance appears everywhere in our life. The challenge for a video surveillance system is to recognize the object of interest for analysis. The convolutional neural network (CNN) models achieve high accuracy on image recognition, but the models require powerful calculations. The models cannot be applied to most smart surveillance systems directly. In this paper, we propose a low-cost CNN design for the application of surveillance systems. Instead of using GPUs, we use a hardware accelerator called Neural Compute Stick (NCS) accompanied with the Rock64 to build the system. The NCS is a low-cost and low-power USB device, which has the advantages in the high-speed calculation of images. As a result, we use the NCS to load the Single Shot MultiBox Detector (SSD) network for human detection. Our system can get each detected image in 0.15 sec. It is six times faster than other single-board surveillance systems. Furthermore, the cost of building the real-time surveillance system is less than 100. Therefore, our system can achieve a low-cost and high-performance intelligent surveillance system.
UR - http://www.scopus.com/inward/record.url?scp=85057578373&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85057578373&partnerID=8YFLogxK
U2 - 10.1109/ICSSE.2018.8520133
DO - 10.1109/ICSSE.2018.8520133
M3 - Conference contribution
AN - SCOPUS:85057578373
T3 - 2018 International Conference on System Science and Engineering, ICSSE 2018
BT - 2018 International Conference on System Science and Engineering, ICSSE 2018
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2018 International Conference on System Science and Engineering, ICSSE 2018
Y2 - 28 June 2018 through 30 June 2018
ER -