Three-dimensional ionospheric tomography using observation data of GPS ground receivers and ionosonde by neural network

The inheritance and innovation of ancient architecture decoration art is an important way for the development of the construction industry. The data process of traditional ancient architecture decoration art is relatively backward, which leads to the obvious distortion of the digitalization of ancient architecture decoration art. In order to improve the digital effect of ancient architecture decoration art, based on neural network, this paper combines the image features to construct a neural network-based ancient architecture decoration art data system model, and graphically expresses the static construction mode and dynamic construction process of the architecture group. Based on this, three-dimensional model reconstruction and scene simulation experiments of architecture groups are realized. In order to verify the performance effect of the system proposed in this paper, it is verified through simulation and performance testing, and data visualization is performed through statistical methods. The result of the study shows that the digitalization effect of the ancient architecture decoration art proposed in this paper is good.

Download Full-text

Automatic method for glaucoma diagnosis using a three-dimensional convoluted neural network

Neurocomputing ◽

10.1016/j.neucom.2020.07.146 ◽

2021 ◽

Vol 438 ◽

pp. 72-83

Author(s):

Nonato Rodrigues de Sales Carvalho ◽

Maria da Conceição Leal Carvalho Rodrigues ◽

Antonio Oseas de Carvalho Filho ◽

Mano Joseph Mathew

Keyword(s):

Neural Network ◽

Three Dimensional ◽

Automatic Method ◽

Glaucoma Diagnosis

Download Full-text

Hardware Acceleration Implementation of Three-Dimensional Convolutional Neural Network on Vector Digital Signal Processors

2020 4th International Conference on Robotics and Automation Sciences (ICRAS) ◽

10.1109/icras49812.2020.9135062 ◽

2020 ◽

Author(s):

Weiwen Chen ◽

Yaohua Wang ◽

Chao Yang ◽

Yong Li

Keyword(s):

Neural Network ◽

Convolutional Neural Network ◽

Three Dimensional ◽

Hardware Acceleration ◽

Digital Signal ◽

Digital Signal Processors ◽

Signal Processors

Download Full-text

Estimation of 6D Object Pose Using a 2D Bounding Box

Sensors ◽

10.3390/s21092939 ◽

2021 ◽

Vol 21 (9) ◽

pp. 2939

Author(s):

Yong Hong ◽

Jin Liu ◽

Zahid Jahangir ◽

Sheng He ◽

Qing Zhang

Keyword(s):

Neural Network ◽

Loss Function ◽

Three Dimensional ◽

Unit Vector ◽

Prediction Algorithm ◽

Computational Time ◽

Bounding Box ◽

Dimensional Unit ◽

Bounding Boxes ◽

Rgb Image

This paper provides an efficient way of addressing the problem of detecting or estimating the 6-Dimensional (6D) pose of objects from an RGB image. A quaternion is used to define an object′s three-dimensional pose, but the pose represented by q and the pose represented by -q are equivalent, and the L2 loss between them is very large. Therefore, we define a new quaternion pose loss function to solve this problem. Based on this, we designed a new convolutional neural network named Q-Net to estimate an object’s pose. Considering that the quaternion′s output is a unit vector, a normalization layer is added in Q-Net to hold the output of pose on a four-dimensional unit sphere. We propose a new algorithm, called the Bounding Box Equation, to obtain 3D translation quickly and effectively from 2D bounding boxes. The algorithm uses an entirely new way of assessing the 3D rotation (R) and 3D translation rotation (t) in only one RGB image. This method can upgrade any traditional 2D-box prediction algorithm to a 3D prediction model. We evaluated our model using the LineMod dataset, and experiments have shown that our methodology is more acceptable and efficient in terms of L2 loss and computational time.

Download Full-text

Three-Dimensional Neural Network to Automatically Assess Liver Tumor Burden Change on Consecutive Liver MRIs

Journal of the American College of Radiology ◽

10.1016/j.jacr.2020.06.033 ◽

2020 ◽

Vol 17 (11) ◽

pp. 1475-1484

Author(s):

Alexander Goehler ◽

Tzu-Ming Harry Hsu ◽

Ronilda Lacson ◽

Isha Gujrathi ◽

Raein Hashemi ◽

...

Keyword(s):

Neural Network ◽

Liver Tumor ◽

Three Dimensional ◽

Tumor Burden

Download Full-text

Modeling and Performance Evaluation of Precise Positioning and Time-Frequency Transfer with Galileo Five-Frequency Observations

Remote Sensing ◽

10.3390/rs13152972 ◽

2021 ◽

Vol 13 (15) ◽

pp. 2972

Author(s):

Wei Xu ◽

Wen-Bin Shen ◽

Cheng-Hui Cai ◽

Li-Hong Li ◽

Lei Wang ◽

...

Keyword(s):

Three Dimensional ◽

Satellite Orbit ◽

Average Frequency ◽

Satellite System ◽

Observation Data ◽

Navigation Satellite ◽

Dual Frequency ◽

Time Frequency ◽

Combination Model ◽

Frequency Transfer

The present Global Navigation Satellite System (GNSS) can provide at least double-frequency observations, and especially the Galileo Navigation Satellite System (Galileo) can provide five-frequency observations for all constellation satellites. In this contribution, precision point positioning (PPP) models with Galileo E1, E5a, E5b, E5 and E6 frequency observations are established, including a dual-frequency (DF) ionospheric-free (IF) combination model, triple-frequency (TF) IF combination model, quad-frequency (QF) IF combination model, four five-frequency (FF) IF com-bination models and an FF uncombined (UC) model. The observation data of five stations for seven days are selected from the multi-GNSS experiment (MGEX) network, forming four time-frequency links ranging from 454.6 km to 5991.2 km. The positioning and time-frequency transfer performances of Galileo multi-frequency PPP are compared and evaluated using GBM (which denotes precise satellite orbit and clock bias products provided by Geo Forschung Zentrum (GFZ)), WUM (which denotes precise satellite orbit and clock bias products provided by Wuhan University (WHU)) and GRG (which denotes precise satellite orbit and clock bias products provided by the Centre National d’Etudes Spatiales (CNES)) precise products. The results show that the performances of the DF, TF, QF and FF PPP models are basically the same, the frequency stabilities of most links can reach sub10−16 level at 120,000 s, and the average three-dimensional (3D) root mean square (RMS) of position and average frequency stability (120,000 s) can reach 1.82 cm and 1.18 × 10−15, respectively. The differences of 3D RMS among all models are within 0.17 cm, and the differences in frequency stabilities (in 120,000 s) among all models are within 0.08 × 10−15. Using the GRG precise product, the solution performance is slightly better than that of the GBM or WUM precise product, the average 3D RMS values obtained using the WUM and GRG precise products are 1.85 cm and 1.77 cm, respectively, and the average frequency stabilities at 120,000 s can reach 1.13 × 10−15 and 1.06 × 10−15, respectively.

Download Full-text