TY - GEN
T1 - Landscape sand table system based on deep learning and augmented reality technology
AU - Ou, Pan
AU - Chen, Moran
AU - Yu, Qingfeng
N1 - Publisher Copyright:
© Springer Nature Singapore Pte Ltd. 2020.
PY - 2020
Y1 - 2020
N2 - Landscape sand table has been widely used in many fields, and higher requirements have been put forward for landscape sand table system. Therefore, the traditional landscape sand table system cannot meet the needs of current use. In addition, the traditional research has great limitations and cannot meet the requirements of high restoration and precision of the landscape sand table. In this paper, on the basis of deep study and augmented reality, combine the two, the in-depth study in augmented reality, and with landscape as the research object, combined with image processing and network technology knowledge, based on extensive analysis of the data source, this paper proposes a method to build a landscape sand table system with higher efficiency and restoration degree. This method is a deep learning algorithm constructed through parameter optimization and experimental verification, which can accurately identify scene information and realize augmented reality effect. In this paper, the stochastic gradient descent algorithm was used to conduct an experiment to improve the accuracy of the target detection in the landscape sand table. The experimental results also showed that the augmented reality technology could make the landscape sand table with high reproducibility and stereoscopic effect, and the use of the augmented reality effect could make the landscape sand table show better development.
AB - Landscape sand table has been widely used in many fields, and higher requirements have been put forward for landscape sand table system. Therefore, the traditional landscape sand table system cannot meet the needs of current use. In addition, the traditional research has great limitations and cannot meet the requirements of high restoration and precision of the landscape sand table. In this paper, on the basis of deep study and augmented reality, combine the two, the in-depth study in augmented reality, and with landscape as the research object, combined with image processing and network technology knowledge, based on extensive analysis of the data source, this paper proposes a method to build a landscape sand table system with higher efficiency and restoration degree. This method is a deep learning algorithm constructed through parameter optimization and experimental verification, which can accurately identify scene information and realize augmented reality effect. In this paper, the stochastic gradient descent algorithm was used to conduct an experiment to improve the accuracy of the target detection in the landscape sand table. The experimental results also showed that the augmented reality technology could make the landscape sand table with high reproducibility and stereoscopic effect, and the use of the augmented reality effect could make the landscape sand table show better development.
KW - Augmented reality
KW - Deep learning
KW - Landscape sand table
KW - Stochastic gradient descent
UR - https://www.scopus.com/pages/publications/85079530772
U2 - 10.1007/978-981-15-2568-1_76
DO - 10.1007/978-981-15-2568-1_76
M3 - 会议稿件
AN - SCOPUS:85079530772
SN - 9789811525674
T3 - Advances in Intelligent Systems and Computing
SP - 565
EP - 570
BT - Big Data Analytics for Cyber-Physical System in Smart City, BDCPS 2019
A2 - Atiquzzaman, Mohammed
A2 - Yen, Neil
A2 - Xu, Zheng
PB - Springer
T2 - 1st International Conference on Big Data Analytics for Cyber-Physical System in Smart City, BDCPS 2019
Y2 - 28 December 2019 through 29 December 2019
ER -