In recent years, high-resolution remotesensing images have been increasingly used in the field of geoscience. Therefore, how to interpret high-resolution remotesensing images with high quality is an urgent problem t...
详细信息
ISBN:
(数字)9798350360660
ISBN:
(纸本)9798350360677
In recent years, high-resolution remotesensing images have been increasingly used in the field of geoscience. Therefore, how to interpret high-resolution remotesensing images with high quality is an urgent problem to be solved. This paper takes high-resolution remotesensing images as the research object, uses deep convolutional neural network (CNN) technology, conducts remotesensing image segmentation and feature extraction based on deep convolutional neural network, establishes a high-precision remotesensing image interpretation model, and improves the accuracy and computing speed of intelligent recognition of remotesensing images. This paper uses residual network (ResNet) as a learning sample to compare its performance with the classical algorithm on the same set of samples. Preliminary studies of this project have shown that the accuracy of convolutional neural networks in image analysis can reach 97.1%, and it shows obvious superiority at multiple spatial scales, and can effectively improve the efficiency of image interpretation.
As the deep development of spatial information sharing service,it brings forward high request to usability and expansibility of supporting system. Based on large-scale scalable server cluster,cloud computing brings ho...
详细信息
As the deep development of spatial information sharing service,it brings forward high request to usability and expansibility of supporting system. Based on large-scale scalable server cluster,cloud computing brings hopes to resolves the existing difficult problems in the domain of geospatial information service. In this paper,we imported cloud computing technology including MapReduce model and Hadoop platform into the domain of geographic information system (GIS). Those key technology problems in the application of GIS such as spatial data storage,spatial index and spatial operation were described and studied in detail. We evaluated the performance and efficiency of spatial operation in Hadoop experiment environment with the real world data set. It demonstrates the applicability of cloud computing technology in computing-intensive spatial applications.
We consider 4-loop Feynman diagrams with 11 internal lines. The associated 10-dimensional loop integrals are calculated for four diagrams with massive internal lines, and we further handle the massless case of the dia...
详细信息
ISBN:
(纸本)9783030869762;9783030869755
We consider 4-loop Feynman diagrams with 11 internal lines. The associated 10-dimensional loop integrals are calculated for four diagrams with massive internal lines, and we further handle the massless case of the diagram referenced in the literature as M61. The computations are performed with double exponential (DE), Quasi-Monte Carlo (lattice and embedded lattice rules) and adaptive integration algorithms, which do not require any user input regarding the integrand behavior. The lattice rule methods are combined with a transformation to help alleviate boundary singularities. The embedded lattice rules are implemented in CUDA C and their execution is accelerated using an NVIDIA Quadro GV100 GPU, whereas DE is parallelized over MPI and executed on an AMD cluster. Adaptive integration is performed with the ParInt multivariate integration package, which is also layered over MPI. For the massless M61 diagram we use a dimensional regularization approach and extrapolation. The results will be compared with respect to accuracy and efficiency, and verified with pySecDec.
Using CNN and GAN improves road safety, says study. Modern machine learning techniques for mobile crowdsourcing may identify driver fatigue early. Our convolutional neural network (CNN) model detects drivers' fati...
详细信息
The authors delineate the specific roles of the research partner institutions from Turkey, Egypt and the USA, in planning and implementing the North Atlantic Treaty Organization (NATO) Science for Peace sponsored Kama...
详细信息
The authors delineate the specific roles of the research partner institutions from Turkey, Egypt and the USA, in planning and implementing the North Atlantic Treaty Organization (NATO) Science for Peace sponsored Kamal Ewida Earth Observatory (KEEO), a network of real-time satellite remotesensing ground stations, being established over the next three years in Egypt, with a tracking station for polar orbiting satellites at Cairo University, and a networked geostationary receiving station for the European Space Agency's Meteosat being deployed at Al Azhar University. The primary objective of the project is to facilitate early warning and mitigation of a wide range of biogenic and anthropogenic disasters. The project will also address mitigation of epidemics and epizootics, through identification and monitoring of infectious disease vector and reservoir habitat. Some examples of common concern among participating countries are climate change and its impacts, the land use problems in agriculture, air pollution problems in major cities such as Cairo and Istanbul, recent epidemics such as the bird flu, swine flu and oil spills along the seashores. Archival and real-time remotesensing and generation of near-real-time spatial data products, utilizing highperformancecomputing clusters, are planned throughout the life cycle of disaster management, including vulnerability assessment, infrastructure safeguards, early warning, emergency response, humanitarian relief, as well as post-disaster damage assessment, reconstruction and societal recovery.
暂无评论