Abstract:Accurately identifying building distribution from remote sensing images with complex background information is challenging. The emergence of diffusion models has prompted the innovative idea of employing the reverse denoising process to distill building distribution from these complex backgrounds. Building on this concept, we propose a novel framework, building extraction diffusion model (BEDiff), which meticulously refines the extraction of building footprints from remote sensing images in a stepwise fashion. Our approach begins with the design of booster guidance, a mechanism that extracts structural and semantic features from remote sensing images to serve as priors, thereby providing targeted guidance for the diffusion process. Additionally, we introduce a cross-feature fusion module (CFM) that bridges the semantic gap between different types of features, facilitating the integration of the attributes extracted by booster guidance into the diffusion process more effectively. Our proposed BEDiff marks the first application of diffusion models to the task of building extraction. Empirical evidence from extensive experiments on the Beijing building dataset demonstrates the superior performance of BEDiff, affirming its effectiveness and potential for enhancing the accuracy of building extraction in complex urban landscapes.

Abstract:Models dedicated to building long-range dependencies often exhibit degraded performance when transferred to remote sensing images. Vision transformer (ViT) is a new paradigm in computer vision that uses multi-head self-attention (MSA) rather than convolution as the main computational module, with global modeling capabilities. However, its performance on small datasets is usually far inferior to that of convolutional neural networks (CNNs). In this work, we propose a small window SWinUNet (2SWUNet) for building extraction from high-resolution remote sensing images. Firstly, the 2SWUNet is trained based on swin transformer by designing a fully symmetric encoder-decoder U-shaped architecture. Secondly, to construct a reasonable U-shaped architecture for building extraction from high-resolution remote sensing images, different forms of patch expansion are explored to simulate up-sampling operations and recover feature map resolution. Then, the small window-based multi-head self-attention (W-MSA) is designed to reduce the computational and memory burden, which is more appropriate for the features of remote sensing images. Meanwhile, the pre-training mechanism is advanced to make up for the lack of decoder parameters. Finally, comparison experiments with other mainstream CNNs and ViTs validate the superiority of the proposed model.

Abstract:In order to solve the problem that blade fixing bolt cannot be detected quickly and conveniently in the field in actual production, this paper proposed a field rapid detection method of wind turbine blade fixing bolt defects based on field programmable gate array (FPGA), and Yolov4-tiny is selected as the basic algorithm. Nonetheless, the original Yolov4-tiny was not suitable for detecting small defects, so this paper improved the Yolov4-tiny to enhance the detection effect. Next, the convolutional operations in the algorithm were encapsulated into intellectual property (IP) cores by high-level synthesis (HLS) and Vivado, and parallel computation was realized using FPGA features. In the end, using Python to call the IP core and the FPGA hardware library, this paper achieved the purpose of rapid detection. Compared with traditional detection methods and other algorithms, the accuracy and speed of this method are significantly improved, which has a good application value.

Abstract:In order to meet the high-resolution and wide spectrum range of the backscattering Raman system, this paper designs and builds a Raman test system based on the echelle spectrometer. In the optical splitting system, compared with the ordinary planar grating spectrometer, the use of the echelle improves the resolution of the system without increasing the volume of the system. The use of intensified charge-coupled device (ICCD) in the detection system improves the signal-to-noise ratio (SNR) and the detection limit of weak spectrum. Finally, the Raman system was spectrally calibrated. The broadband backscattering Raman experimental results are given and discussed. The experimental results show that the system has an excellent application prospect for broadband and high-resolution Raman spectrum measurement.

Abstract:To solve the problems with coronary stent implantation, coronary artery stent surface was directly modified by three-beam laser interference lithography through imitating the water-repellent surface of lotus leaf, and uniform micro-nano structures with the controllable period were fabricated. The morphological properties and contact angle (CA) of the microstructure were measured by scanning electron microscope (SEM) and CA system. The water repellency of stent was also evaluated by the contact and then separation between the water drop and the stent. The results show that the close-packed concave structure with the period of about 12 μm can be fabricated on the stent surface with special parameters (incident angle of 3°, laser energy density of 2.2 J∙cm-2 and exposure time of 80 s) by using the three-beam laser at 1 064 nm, and the structure has good water repellency with CA of 120°.

Abstract:The thermal properties of photonic crystal fiber (PCF) laser with 18 circularly distributed cores are investigated by using full-vector finite element method (FEM). The results show that the 18-core PCF has a more effective thermal dispersion construction compared with the single core PCF and 19-core PCF. In addition, the temperature distribution of 18-core PCF laser with different thermal loads is simulated. The results show that the core temperature approaches the fiber drawing value of 1800 K approximately when the thermal load is above 80 W/m which corresponds to the pumping power of 600 W approximately, while the coating temperature approaches the damage value of about 550 K when the thermal load is above 15 W/m which corresponds to the pumping power of 110 W approximately. Therefore the fiber cooling is necessary to achieve power scaling. Compared with other different cooling systems, the copper cooling scheme is found to be an effective method to reduce the thermal effects.