Neural Network Training System for Marker Encoding
- 作者: Wang L.1, Grinchuk O.2, Tsurkov V.2,3
-
隶属关系:
- Nanjing University of Aeronautics and Astronautics
- Moscow Institute of Physics and Technology
- Federal Research Center Computer Science and Control, Russian Academy of Sciences
- 期: 卷 58, 编号 3 (2019)
- 页面: 434-440
- 栏目: Artificial Intelligence
- URL: https://journals.rcsi.science/1064-2307/article/view/220387
- DOI: https://doi.org/10.1134/S1064230719030195
- ID: 220387
如何引用文章
详细
In this paper, we propose a training system for visual markers that provides the generation and subsequent recognition (under real-world conditions) of stylized images that contain information encoded by a sequence of bits. New types of neural network layers that make the recognition process tolerant to external noise are developed. The training process is based on the end-to-end principle, which enables automatic training for the intermediate stages of the model. The experimental results that demonstrate the performance of the system in encoding and decoding artificial markers are presented.
作者简介
L. Wang
Nanjing University of Aeronautics and Astronautics
Email: tsurkov@ccas.ru
中国, Nanjing
O. Grinchuk
Moscow Institute of Physics and Technology
编辑信件的主要联系方式.
Email: oleg.grinchuk@phystehc.edu
俄罗斯联邦, Dolgoprudny, 141701
V. Tsurkov
Moscow Institute of Physics and Technology; Federal Research Center Computer Science and Control, Russian Academy of Sciences
编辑信件的主要联系方式.
Email: tsurkov@ccas.ru
俄罗斯联邦, Dolgoprudny, 141701; Moscow, 119991
![](/img/style/loading.gif)