In recent years, remarkable performance improvements have been produced by deep convolutional neural networks (CNN) for single image super-resolution (SISR). Nevertheless, a high proportion of CNN-based SISR models are with quite a few network parameters and high computational complexity for deep or wide architectures. How to more fully utilize deep features to make a balance between model complexity and reconstruction performance is one of the main challenges in this field. To address this problem, on the basis of the well-known information multi-distillation model, a multi-attention feature distillation network termed as MAFDN is developed for lightweight and accurate SISR. Specifically, an effective multi-attention feature distillation block (MAFDB) is designed and used as the basic feature extraction unit in MAFDN. With the help of multi-attention layers including pixel attention, spatial attention, and channel attention, MAFDB uses multiple information distillation branches to learn more discriminative and representative features. Furthermore, MAFDB introduces the depthwise over-parameterized convolutional layer (DO-Conv)-based residual block (OPCRB) to enhance its ability without incurring any parameter and computation increase in the inference stage. The results on commonly used datasets demonstrate that our MAFDN outperforms existing representative lightweight SISR models when taking both reconstruction performance and model complexity into consideration. For example, for 4 SR on Set5, MAFDN (597K/33.79G) obtains 0.21 dB/0.0037 and 0.10 dB/0.0015 PSNR/SSIM gains over the attention-based SR model AFAN (692K/50.90G) and the feature distillation-based SR model DDistill-SR (675K/32.83G), respectively.
In China, most provinces have established mature Continuous Operational Reference Systems (CORS) base stations, and enterprises such as Qianxun Spatial Intelligence Inc., (hereafter called Qianxun), China Mobile, and Six Points Technology have all deployed CORS sites one after another. Throughout this paper, the problem of differences in the accuracy of differential corrections broadcast by different types of CORS base stations is addressed, affecting Real Time Kinematic (RTK) positioning. As examples, CORS base stations of Qianxun and China Mobile broadcast differential data, and the same type of receivers are used to collect RTCM3 differential data, static raw data of mobile stations, and GGA data with similar baseline distances. After decoding the raw data, the ionospheric delay and tropospheric delay after double-differenced are back-projected through the combination of the ionosphere free, and time series graphs of their different time periods are obtained to compare the differences of RTCM3 differential data broadcast by different types of base stations. As a result of the method presented in this paper, it is possible to assess how ionospheric and tropospheric delay errors, along with other factors, affect real-time positioning accuracy in mapping, unmanned vehicles, and other fields. This is of great benefit for improving real-time positioning accuracy.