Visual depth guided color image rain streaks removal using sparse coding

The embodiment of the invention provides a positioning method and device, and the method comprises the steps: obtaining a first arrival moment and a transmission duration when a detection signal transmitted by a first base station in a positioning system arrives at a target, and all second arrival moments when detection signals transmitted by other base stations in the positioning system arrive atthe target; according to the first arrival time and each second arrival time, determining the time difference of arrival of the detection signals sent by the first base station and other base stations to the target; determining the distance from the first base station to the target and the distance difference between the first base station and other base stations to the target according to the transmission duration and the time difference; and determining the position information of the target according to the distances and the distance differences. In the embodiment of the invention, the plurality of position points with the same distance from the first base station can be determined according to the distance, and the individual position points are screened out from the plurality of position points with the same distance from the first base station for positioning on the basis of the distance difference between the first base station and the other base stations and the target, so that the target positioning accuracy is improved.

Positioning method and device

Fast RF-UIC: A fast unsupervised image captioning model

Most of the existing deraining methods cannot preserve the details of the image while removing the rain streaks. To solve this problem, we propose a single image de-raining method with dual U-Net generative adversarial network (DU-GAN). By using two U-Net with stronger learning ability as our generator DU-GAN can not only accurately remove more rain streaks but also preserve image details. The network can make full use of image information and extract complete image features. The adversarial loss function using the proposed dual U-Net generator is utilized to generate de-rained images which are close to the ground truth. Furthermore, to obtain the better visual effects of the generated image, The L1 and structure similarity loss functions which are consistent with the human visual effect are applied to generate the final output. The synthetic rainy image datasets and real rainy image datasets are used to evaluate the effectiveness of the proposed network in the experiments. The quantitative and visual experimental results show that the proposed single image deraining method achieves state-of-the-art compared with the other single image deraining methods. The source code can be found at https://github.com/LuBei-design/DU-GAN
 .

Single image deraining with dual U-Net generative adversarial network

As an indispensable part of underwater intelligent operations, underwater unmanned system is an important research direction for consolidating China’s ocean security, national defense security, and helping the country’s great strategy of “ marine power “. Its environmental awareness and cognitive capabilities will become an understanding Fundamental and key issues of the ocean and the economic ocean. Existing systems have great deficiencies in underwater perception and detection capabilities, and they cannot all meet the requirements of operations in complex underwater changing environments. This article conducts technical research on the generation of underwater high-definition panoramic 3D data. Therefore, the distributed sonar and binocular vision hybrid system is used as the perception source, and the efficient data access strategy is adopted to update the weight of the corresponding voxel. The integrated low noise high dynamic range chip is used as the imaging terminal, and the underwater optical communication is used to complete many High-precision stereo vision measurement of scenes with point coordination and the generation of refined panoramic 3D data.

Underwater high-precision panoramic 3D image generation

Multi-scale single image rain removal using a squeeze-and-excitation residual network

The removal of rain streaks in a single image is an extremely challenging task due to the uneven rainfall density in the image. Methods based on deep learning have boosted the performance of rain removal significantly in recent years. However, most of these methods have a certain demand for different density of rain marks in the training data, which prevent them to further improve the performance in some outdoor scenarios. In this paper, we present a novel Rain-Density Squeeze-and-Excitation residual network (RDSER-NET), which adopts the squeeze-and-excitation blocks into the ResNet framework. The proposed network remove rain streaks based on single density of rain marks in the training data, reducing the limitation of multi-density proposals and achieving better results. Extensive experiments on synthetic and real datasets demonstrate that the proposed network outperform the recent state-of-the-art methods greatly.

Rain-Density Squeeze-and-Excitation Residual Network for Single Image Rain-removal

The invention discloses a BDS satellite and ultra wide band based dual-mode positioning method. The method comprises the following steps that the velocity V1 and position information L1 of a monitoring object are obtained on the basis of ultra-wide-band short-range wireless positioning; a BDS is used to obtain the velocity V2 and position information L2 of the monitoring object via particle filterfused positioning; and Kalman filtering is used to carry out fused positioning on the positioning information and velocities (L1, V1), (L2, V2), and accurate positioning information is obtained. Themethod can be used to back feed the accurate position information of the monitoring object, and the monitoring object can be treated timely under an emergent condition or other conditions.

Xipu Hu

Papers

Multi-scale single image rain removal using a squeeze-and-excitation residual network

Rain-Density Squeeze-and-Excitation Residual Network for Single Image Rain-removal

BDS satellite and ultra wide band based dual-mode positioning method