TY - GEN
T1 - AI Management System to Prevent Accidents in Construction Zones Using 4K Cameras Based on 5G Network
AU - Nozaki, Daichi
AU - Okamoto, Koki
AU - Mochida, Toru
AU - Qi, Xin
AU - Wen, Zheng
AU - Myint, San Hlaing
AU - Tokuda, Kiyohito
AU - Sato, Takuro
AU - Tamesue, Kazuhiko
PY - 2019/5/10
Y1 - 2019/5/10
N2 - Accident prevention for trucks, cranes and work vehicles at construction sites is important. Here, we use high-precision surveillance cameras with 4K cameras as IoT terminals to implement safe workplace environments. In this research, the system transmits photographic images from trucks, cranes, and other construction equipment fitted with 4K cameras at the work site to a database via 5G wireless networks, and uses AI to assess the interactions and movements of workers in the database. We introduce a system that makes it possible to avoid crashes by informing truck and crane drivers, and notifying automatic driving trucks and crane cars of that information. It is conceivable that uplink traffic could become congested due to many vehicles equipped with 4K cameras simultaneously transmitting images over the 5G uplink. Here, as a basic 5G characteristic evaluation, 4K images from trucks or cranes moving at a low speed are transmitted to the database according to moving speed and distance to the 5G terminal connecting the base station and IoT device. Based on the error environment, we report the relation with the error of the system that judges and identifies the surrounding environment using AI when video quality is deteriorated.
AB - Accident prevention for trucks, cranes and work vehicles at construction sites is important. Here, we use high-precision surveillance cameras with 4K cameras as IoT terminals to implement safe workplace environments. In this research, the system transmits photographic images from trucks, cranes, and other construction equipment fitted with 4K cameras at the work site to a database via 5G wireless networks, and uses AI to assess the interactions and movements of workers in the database. We introduce a system that makes it possible to avoid crashes by informing truck and crane drivers, and notifying automatic driving trucks and crane cars of that information. It is conceivable that uplink traffic could become congested due to many vehicles equipped with 4K cameras simultaneously transmitting images over the 5G uplink. Here, as a basic 5G characteristic evaluation, 4K images from trucks or cranes moving at a low speed are transmitted to the database according to moving speed and distance to the 5G terminal connecting the base station and IoT device. Based on the error environment, we report the relation with the error of the system that judges and identifies the surrounding environment using AI when video quality is deteriorated.
KW - 4K
KW - 5G
KW - AI
KW - Artificial Intelligence
KW - Autonomous driving technologies
KW - Image processing
KW - Machine learning
KW - Object detection
KW - Tensorflow
UR - http://www.scopus.com/inward/record.url?scp=85066318061&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85066318061&partnerID=8YFLogxK
U2 - 10.1109/WPMC.2018.8712896
DO - 10.1109/WPMC.2018.8712896
M3 - Conference contribution
AN - SCOPUS:85066318061
T3 - International Symposium on Wireless Personal Multimedia Communications, WPMC
SP - 462
EP - 466
BT - 2018 21st International Symposium on Wireless Personal Multimedia Communications, WPMC 2018
PB - IEEE Computer Society
T2 - 21st International Symposium on Wireless Personal Multimedia Communications, WPMC 2018
Y2 - 25 November 2018 through 28 November 2018
ER -