TY - JOUR
T1 - Research on tracking method of flower target
AU - Chen, Yunqiang
AU - Wang, Qing
AU - Chen, Hong
AU - Li, Xiang
AU - Song, Xiaoyu
N1 - Publisher Copyright:
© Published under licence by IOP Publishing Ltd.
PY - 2020/3/3
Y1 - 2020/3/3
N2 - Flowers are a common species in people's lives, but people's understanding of flowers is not comprehensive. The traditional UI display method lacks immersion and interactivity. This paper proposes a method based on the deep learning neural network flower recognition algorithm to recognize the flower target information and replace the traditional UI with the AR information to display the flower information. The tracking and positioning technology used in the floral AR display method is a very important underlying technology. This paper focuses on the tracking technology of flower target, and proposes a server-based flower recognition neural network algorithm recognition algorithm. After identifying the flower information and detection frame, it outputs the detection frame parameters of the flower target, and then combines the OpenCV KCF tracker pair. The method of performing AR display of the flower information has high tracking accuracy and stability.
AB - Flowers are a common species in people's lives, but people's understanding of flowers is not comprehensive. The traditional UI display method lacks immersion and interactivity. This paper proposes a method based on the deep learning neural network flower recognition algorithm to recognize the flower target information and replace the traditional UI with the AR information to display the flower information. The tracking and positioning technology used in the floral AR display method is a very important underlying technology. This paper focuses on the tracking technology of flower target, and proposes a server-based flower recognition neural network algorithm recognition algorithm. After identifying the flower information and detection frame, it outputs the detection frame parameters of the flower target, and then combines the OpenCV KCF tracker pair. The method of performing AR display of the flower information has high tracking accuracy and stability.
UR - http://www.scopus.com/inward/record.url?scp=85081160264&partnerID=8YFLogxK
U2 - 10.1088/1742-6596/1453/1/012155
DO - 10.1088/1742-6596/1453/1/012155
M3 - 会议文章
AN - SCOPUS:85081160264
SN - 1742-6588
VL - 1453
JO - Journal of Physics: Conference Series
JF - Journal of Physics: Conference Series
IS - 1
M1 - 012155
T2 - 2019 2nd International Conference on Computer Information Science and Artificial Intelligence, CISAI 2019
Y2 - 25 October 2019 through 27 October 2019
ER -