Remote-sensing images constitute an important means of obtaining geographic information. Image super-resolution reconstruction techniques are effective methods of improving the spatial resolution of remote-sensing images. Super-resolution reconstruction networks mainly improve the model performance by increasing the network depth. However, blindly increasing the network depth can easily lead to gradient disappearance or gradient explosion, increasing the difficulty of training. This report proposes a new pyramidal multi-scale residual network (PMSRN) that uses hierarchical residual-like connections and dilation convolution to form a multi-scale dilation residual block (MSDRB). The MSDRB enhances the ability to detect context information and fuses hierarchical features through the hierarchical feature fusion structure. Finally, a complementary block of global and local features is added to the reconstruction structure to alleviate the problem that useful original information is ignored. The experimental results showed that, compared with a basic multi-scale residual network, the PMSRN increased the peak signal-to-noise ratio by up to 0.44 dB and the structural similarity to 0.9776.

Text classification refers to labelling text with specified labels, and it is widely used in public opinion supervision, spam detection, and other fields. However, due to the complex semantics of natural language and the difficulty of extracting semantic features, users of traditional methods encounter difficulties when trying to achieve better classification results. In response to this problem, a text classification method based on the CBM (Convolutional and Bi-LSTM Model) model, which can extract shallow local semantic features and deep global semantic features, is proposed. First, the text is vectorised using the Glove model in the embedding layer. Then, the vector text is sent to the Multiscale Convolutional Neural Network (MCNN) and the Bidirectional Long Short-Term Memory network (Bi-LSTM) respectively. The Bi-LSTM layer is also designed in the present work with use of mixed attention to extract deeper semantic features. Finally, the MCNN features and Bi-LSTM features are fused and sent to the softmax layer for classification. Experimental results show that the model can significantly improve the accuracy of text classification.

In recent years, deep learning methods have been widely used for road extraction in remote sensing images. However, the existing deep learning semantic segmentation networks generally show poor continuity in road segmentation due to the high-class similarity between roads and buildings surrounding roads in remote sensing images, and the existence of shadows and occlusion. To deal with this problem, this paper proposes strip attention networks (SANet) for extracting roads in remote sensing images. Firstly, a strip attention module (SAM) is designed to extract the contextual information and spatial position information of the roads. Secondly, a channel attention fusion module (CAF) is designed to fuse low-level features and high-level features. The network is trained and tested using the CITY-OSM dataset, DeepGlobe road extraction dataset, and CHN6-CUG dataset. The test results indicate that SANet exhibits excellent road segmentation performance and can better solve the problem of poor road segmentation continuity compared with other networks.

For mobile users, the ability to accurately acquire their own location is critical. By locating, mobile users can get information about their environment and access location-related services. RSS fingerprint-based indoor localization method collects a training database of measurement fingerprints and uses a machine learning classifier to determine a person’s location from a new fingerprint. However, as the environment changes over time due to furniture or other objects being moved, the new fingerprints diverge from those in the original database. Therefore, an RSS-difference based localization system is designed to deal with the above problem. This method combines back-propagation neural network (BPNN) and weighted K-Nearest Neighbor (WKNN) method to improve the fingerprint similarity based indoor location method (FSIL). We train BPNN in the off-line stage to obtain the optimal BPNN parameter settings. In the online stage, K nearest neighbor points are firstly selected based on the improved FSIL algorithm, and then the difference of signal strength values between the K nearest neighbor points and the target user is input into the BPNN network, to obtain the Euclidean distance between the K nearest neighbor points and the target user, and finally the WKNN algorithm is used to obtain the user’s ultimate location. Simulation experiments based on the LDPL model and the Wireless Insite software, as well as the test results based on the indoor localization dataset IPIN2016, show that the localization accuracy in complex indoor scenarios can be improved by at least 11% when using the method proposed in this paper.

This article proposes a model for Chinese text classification based on a feature-enhanced nonequilibrium bidirectional long short-term memory (Bi-LSTM) network that analyzes Chinese text information in depth and improves the accuracy of text classification. First, the bidirectional encoder representations from transformers model was used to vectorize the original Chinese corpus and extract preliminary semantic features. Then, a nonequilibrium Bi-LSTM network was applied to increase the weight of text information containing important semantics and further improve the effects of the key features in Chinese text classification. Simultaneously, a hierarchical attention mechanism was used to widen the gap between the important and unimportant data. Finally, the softmax function was used for classification. By comparing the classification performance of the proposed scheme with those of various other models, it was observed that the model substantially improved the precision of Chinese text classification and had a strong ability to recognize Chinese text features. The model achieved 97% precision on the experimental dataset.