Region Matching of SAR Images Using Blocks for Target Recognition

A synthetic aperture radar (SAR) target recognition method based on image blocking and matching is proposed. The test SAR image is first separated into four blocks, which are analyzed and matched separately. For each block, the monogenic signal is employed to describe its time-frequency distribution and local details with a feature vector. The sparse representation-based classification (SRC) is used to classify the four monogenic feature vectors and produce the reconstruction error vectors. Afterwards, a random weight matrix with a rich set of weight vectors is used to linearly fuse the feature vectors and all the results are analyzed in a statistical way. Finally, a decision value is designed based on the statistical analysis to determine the target label. The proposed method is tested on the moving and stationary target acquisition and recognition (MSTAR) dataset and the results confirm the validity of the proposed method.

Reviews of the Regions of the Russian Empire as a Historical Source on the History of Education (on the Example of the Regions of the Steppe Territory and Turkestan)

Reviews of the provinces and regions of the Russian Empire (appendices to the reports of governors) are a comprehensive source on the history of the regions. As the official statistical publication, the reviews are controversially assessed by historians for the reliability of the data. However, comparisons with other sources, as well as critical analysis, allow researchers to view the survey as a representative source. The article analyzes the source potential of reviews as a source on the history of the development of education in the outskirts territories of the Russian Empire. The characteristics of the information contained in the source on the development of education are given. As a result of working with the source, a set of issues in the field of education development that should be considered using the survey data as an independent or auxiliary source (quantitative growth of educational institutions, students, development of the network of educational institutions, its structure, features). The requirements for the structure of gubernatorial reports and their appendices were established at the beginning of the 19th century. However, the structure of reviews and the information presented in them differs significantly depending on the year or region. Matching of the information with other sections of the source and the historical context of the period under review allow us to conclude that the reviews in the education section are highly informative.

Scattering of SH Wave by a Semi-Circle Inclusion Embedded in Bi-Material Half Space Surface

ABSTRACTInvestigation of SH wave scattering by inclusions in bi-material half space is an important issue in engineering. The purpose of this work is to study the dynamic response of a semi-circle inclusion embedded in bi-material half space surface by SH wave. Graf's addition theorem, Green function method and region-matching technique are used to determine the displacement fields in the bi-material half space and the inclusion. The distributions of dynamic stress concentration factor (DSCF) around the semi-circle inclusion are depicted graphically considering different material parameters. The results show that the frequency and the incidence angle of SH wave, the rigidities of the inclusion and bi-material half space, and the distance from the inclusion to the interface have a great effect on the distribution of DSCF around the inclusion.

Cross-City Transfer Learning for Deep Spatio-Temporal Prediction

Spatio-temporal prediction is a key type of tasks in urban computing, e.g., traffic flow and air quality. Adequate data is usually a prerequisite, especially when deep learning is adopted. However, the development levels of different cities are unbalanced, and still many cities suffer from data scarcity. To address the problem, we propose a novel cross-city transfer learning method for deep spatio-temporal prediction tasks, called RegionTrans. RegionTrans aims to effectively transfer knowledge from a data-rich source city to a data-scarce target city. More specifically, we first learn an inter-city region matching function to match each target city region to a similar source city region. A neural network is designed to effectively extract region-level representation for spatio-temporal prediction. Finally, an optimization algorithm is proposed to transfer learned features from the source city to the target city with the region matching function. Using citywide crowd flow prediction as a demonstration experiment, we verify the effectiveness of RegionTrans. Results show that RegionTrans can outperform the state-of-the-art fine-tuning deep spatio-temporal prediction models by reducing up to 10.7% prediction error.