China
Africa
Explore chapters and articles related to this topic
Automated Methods for Vessel Segmentation in X-ray Coronary Angiography and Geometric Modeling of Coronary Angiographic Image Sequences: A Survey
Published in Kayvan Najarian, Delaram Kahrobaei, Enrique Domínguez, Reza Soroushmehr, Artificial Intelligence in Healthcare and Medicine, 2022
Zijun Gao, Kritika Iyer, Lu Wang, Jonathan Gryak, C. Alberto Figueroa, Kayvan Najarian, Brahmajee K. Nallamothu, Reza Soroushmehr
Recently, researchers applied deep learning to stenosis detection with XCA image patches. Antczak and Liberadzki (2018) created an artificial dataset with an algorithm that mainly employed Bézier curves to train convolutional neural networks of up to five layers from scratch and then used real XCA images patches for further training. Au et al. (2018) proposed a deep learning framework for handling XCA images to achieve real-time stenosis detection with three network modules that serve to localize, segment, and identify clinically significant stenosis regions. Inputs of the segmentation network module are small patches obtained from the localization module. Ovalle-Magallanes et al. (2020) focused on transfer learning for stenosis detection in XCA image patches by introducing a network-cut and fine-tuning hybrid method. Various setups and fine-tuning strategies were tested on different network structures, with both the classification results and activation maps for a visual explanation presented.
Resilient UAV Networks: Solutions and Trends
Published in Fei Hu, Xin-Lin Huang, DongXiu Ou, UAV Swarm Networks, 2020
Zhiyong Xia, Fei Hu, Nathan Jeong, Iftikhar Rasheed
The three fundamental applications of a resilient UAV-based sensor system are briefly discussed. The first application is the localization function. In the sensor network, if the sensor's position information is not available or not accurate, then the localization function has to be activated. In Adler, the UAV-based localization method is proposed. Basically, it divides the target area into grids, then designs the flight trajectory of UAVs to cover all sensor points, and it finally calculates the location coordinates of each sensor via the received signal strength indicator (RSSI). The RSSI is utilized to calculate the distance between a sender and a receiver, and its formula is shown in Eqn. (13.4). RSSI=A+10nlog10(d)
Single Electron Devices and Applications
Published in Simon Deleonibus, Electronic Device Architectures for the Nano-CMOS Era, 2019
Jacques Gautier, Xavier Jehl, Marc Sanquert
In a Single Electron Device, SED, the flow of current between electrodes is quantized by the electron charge. This is the main difference to conventional electronic devices, MOSFET or BJT, where this flow is continuous. Of course, to obtain such a behavior, there are conditions to meet. Especially, at each time, it is required to have an integer number of electron, or hole, in the body of the device, which implies a localization of the electron wave function. The most obvious way to achieve that is to implement two tunneling junctions, or potential barriers, which define an island in between. Their equivalent resistance RT should be higher than the quantum of resistance RQ = h/2e2 ~ 13kΩ.2 When a third electrode is added for an electrostatic control of the island potential, a Single Electron Transistor, SET, is obtained (Fig. 1).
Image processing algorithm for mechanical properties testing of high-temperature materials based on time-frequency analysis
Published in Journal of Experimental Nanoscience, 2023
It can be seen from the comparison data of the above two tables that when the size of the sub-region is the same, the calculation time of the two algorithms will increase with the increase of the image size, but the time of the point-by-point search method is much longer than that of the fast ZNC algorithm. In the case of the same image size, the calculation time of the point-by-point search method increases significantly with the increase of the sub-region, but the calculation time of the fast ZNC algorithm remains basically unchanged. It can be seen that the fast ZNC algorithm is separated from the sub-region. Due to the limitation of the region size, the calculation speed is only affected by the image size, which is also consistent with the previous derivation results. Therefore, this paper chooses this algorithm as the integer pixel localization algorithm of the system.
Sound absorption by perforated walls along boundaries
Published in Applicable Analysis, 2022
Patrizia Donato, Agnes Lamacz, Ben Schweizer
Let be an interval. Since is supported on , Proposition 3.1 is proved as soon as we can show that the limit measure satisfies We will use the following function with large gradients: We want to use a localization function . As a test function we then consider . The proof of (22) consists in calculating the quantity in two different ways.
Solar-driven zinc-doped graphitic carbon nitride photocatalytic fibre for simultaneous removal of hexavalent chromium and pharmaceuticals
Published in Environmental Technology, 2022
Zhexin Zhu, Yongquan Miao, Gangqiang Wang, Wenxing Chen, Wangyang Lu
The structure information and element valence states of the photocatalyst were further explored by XPS analysis [31,41]. As shown in Figure 2, the elements N, Zn and O containing no other elements were showed in the XPS spectrum. For the N 1s spectrum, four peaks were observed at 398.5, 399.6, 401.02, and 404.28 eV. The major signal near 398.5 eV due to the presence of sp2 bonded N(C–N = C), while the signal at 399.6 eV is generally considered to be a N-(C) 3 group. The signal at 401.02 eV is attributed to C–N-H. The weak signal at 404.28 eV can be explained as the positive charge localization in the heptazine ring. N 1s XPS spectra of g-C3N4/PAN and Zn-g-C3N4/PAN due to loading content can be fitted to two main peaks near 399.6 and 400 eV, which is explained to sp2-hybridized nitrogen atom bonded to two carbon atoms (C–N-C) and tertiary nitrogen N-(C)3 group in Zn-g-C3N4 powder catalyst. The Zn 2p spectrum of Zn-g-C3N4/PAN catalyst fibre shows double peaks near 1021.4 and 1044.7 eV, which explain Zn 2p3/2 and 2p1/2 of Zn2 +. It is proved that the zinc was successfully loaded with g-C3N4. For the O 1s spectrum, the signal of the Zn-g-C3N4/PAN catalyst does not change much compared to other catalysts.