U-shaped Structure Research Articles

Pulmonary CT Registration Network Based on Deformable Cross Attention.

Current Transformer structure utilizes the self-attention mechanism to model global contextual relevance within image, which makes an impact on medical image registration. However, the use of Transformer in handling large deformation lung CT registration is relatively straightforwardly. These models only focus on single image feature representation neglecting to employ attention mechanism to capture the across image correspondence. This hinders further improvement in registration performance. To address the above limitations, we propose a novel registration method in a cascaded manner, Cascaded Swin Deformable Cross Attention Transformer based U-shape structure (SD-CATU), to address the challenge of large deformations in lung CT registration. In SD-CATU, we introduce a Cross Attention-based Transformer (CAT) block that incorporates the Shifted Regions Multihead Cross-attention (SR-MCA) mechanism to flexibly exchange feature information and thus reduce the computational complexity. Besides, a consistency constraint in the loss function is used to ensure the preservation of topology and inverse consistency of the transformations. Experiments with public lung datasets demonstrate that the Cascaded SD-CATU outperforms current state-of-the-art registration methods (Dice Similarity Coefficient of 93.19% and Target registration error of 0.98mm). The results further highlight the potential for obtaining excellent registration accuracy while assuring desirable smoothness and consistency in the deformed images.

Impact of U-shaped waveguide structures on optical comb generation in microresonator systems

In the research of soliton microcombs, the pump energy conversion efficiency (ECE) of the microcomb is an important parameter. However, in traditional microresonator systems, the ECE is only a few percent. Researchers have enhanced it by introducing additional feedback coupling structures or composite cavity structures into the microresonator system to suppress the output of pump light in the soliton generation region. The additional structures may affect the output comb characteristics of the system, but most articles do not delve into this issue, which hinders the practical application of high-efficiency soliton microcombs. In this paper, we establish rate equations describing the nonlinear dynamics of a microresonator with a U-shaped waveguide feedback coupling structure (coupled-cavity system, CCS), and numerically simulate and conduct steady-state analysis of the simulation results to obtain the influence of the U-shaped waveguide on the CCS’s output comb characteristics. Then we optimize the structural parameters of the CCS reducing the parametric oscillation threshold of the pump by more than 50%, while increasing the ECE to over 25% when the optical field in the microresonator propagates in the form of a single soliton.

Poly-Adenine Assisted Signaling Displaced Probe Ratiometric Electrochemical Aptasensor for Accurate Detection of Alzheimer's Disease Aβ Biomarkers.

Alzheimer's disease (AD) is a widely prevalent neurodegenerative condition globally, arousing significant interest in the noninvasive early detection of the disease. The concentration of amyloid β (Aβ) biomarkers in the blood is closely linked to the progression of AD, emphasizing the importance of developing a precise method for detecting these biomarkers in blood samples for early diagnosis. In this study, we developed a ratiometric electrochemical aptamer-based (EAB) biosensor for accurate detection of Aβ42 and Aβ40. The ratiometric biosensor utilizes a gold nanoparticle (AuNPs) modified electrode through an electrochemical deposition assay and a self-assembled poly-A based U-shaped structure to achieve signal changes through competitive binding of target molecules. Importantly, using the calibration of the clipping reference probe, the proposed biosensor displayed accurate detection ability for Aβ42 and Aβ40, with detection limits of 59 and 21 pM, respectively. It also has great stability and anti-interference ability in complex biological samples. Furthermore, it demonstrated satisfactory performance in detecting the Aβ peptides in a 20% FBS sample, with recoveries ranging from 96.7% to 102.2% in serum samples, and relative standard deviations (RSD) ranging from 3.3% to 8.9%. Moreover, the proposed method is expected to directly quantify the Aβ42/Aβ40 ratio by introducing a suitable reference probe, providing a potential and effective tool for the early diagnosis of AD.

Automated Grasp Labeling and Detection Framework with Pixel-Level Precision

Due to the differences in shape, material, and color of objects, the detection of planar grasps by robots remains challenging. Traditional methods rely on discrete grasp configurations for annotation, ignoring many possible grasp configurations. This leads to poor network generalization, making it challenging to handle a diverse range of grasped objects. Manual reannotation to continuous labels can effectively address this issue but comes with a significant cost. Therefore, this paper proposes a Pixel-level Grasp framework. Firstly, APGLG can automatically generate pixel-level grasp labels and discrete labels into continuous labels, effectively increasing the information content of individual data and improving the network generalization performance. Then, we propose the Max-Grasp-Net, built on the Multi-axis Vision Transformer and Dynamic Convolution Decomposition, to construct a U-shaped Network structure. A grasp decoder is incorporated, and deep supervision is applied to enhance network generalization. On the Cornell dataset, we achieve the best results, a grasping accuracy of 99.55%, an average success rate of 97.92% in single-object grasping, and 95.83% in multi-object grasping. We verified the effectiveness of our label generation algorithms and network innovation through actual grasp experiments.

Visual detection of drilling robot position for rockburst prevention in mining processing by a new image dehazing method

The positioning of drill pipe based on visual detection is a crucial aspect in achieving unmanned operation of drilling robot for rockburst prevention. However, the images directly collected from mechanized mining face are always polluted. In order to eliminate the interference of dust and haze in the image, an image dehazing method based on improved CycleGAN model is proposed in this paper. Firstly, a pipe image dehazing dataset for the rockburst prevention drilling robot is collected and established. Moreover, a generator architecture with a multi-scale U-shaped network structure is designed to improve the quality and accuracy of image recovery. A new reconstruction block is designed and an SK fusion layer is introduced to improve the feature extraction capability of the model, and a MU-CycleGAN network structure is constructed. Finally, an experimental platform for drill pipe image dehazing of the drilling robot for rockburst prevention is set up in the intelligent mining equipment laboratory. Experimental results showed that the PSNR and SSIM of the image dehazing model achieved 27.04 and 0.946, and the success rate of drill pipe grabbing has increased by 13.75%. Experimental results reveal that the proposed framework achieves superior image enhancement performance than the comparison algorithms.

Simulation analysis and optimization of pilot type electro-hydraulic proportional directional valve based on multi field coupling

The pilot-operated electro-hydraulic proportional directional valve is widely used in complex and precision engineering fields. Its operational stability and reliability determine the overall performance of the hydraulic system. During the working process of the pilot-operated electro-hydraulic proportional directional valve, the spool is deformed due to the influence of pressure and temperature, resulting in problems such as clamping or stagnation. In order to solve this problem, this study uses the fluid-solid-thermal coupling simulation method to analyze the flow field characteristics and thermal deformation of the spool with different structural throttling grooves, and optimizes the U-shaped throttling groove structure. The results indicate that as the valve opening increases, the front section of the throttling groove becomes the primary region where the maximum radial deformation of the spool occurs. In this paper, the steady-state flow force of the valve core is taken as the measurement standard, the constraint range of the structural parameters is determined, and the radial thermal deformation is included in the optimization analysis of the structural size. It's verified that the optimized flow characteristics are improved by about 32 % compared with that before optimization, and the maximum steady-state flow force after considering thermal deformation is reduced by about 4N compared with that before optimization. This provides an important design basis and reference for the structural design and optimization of the pilot electro-hydraulic proportional directional valve.

Impact of SiGe pocket on different shape TFET structures for gas sensing application

Gas sensing requires highly sensitive and selective sensor technologies for environmental monitoring, industrial safety, and public health objectives. Although conventional MOSFET-based gas sensors are widely utilized, their detection of low quantities of gases, such as ammonia, is hindered by a number of severe constraints. TFET is emerged as the better device than MOSFET, particularly for the sensing applications. In this work we discussed the source engineered and shape engineered techniques to improve the performance of TFETs, specifically for ammonia gas detection. Comprehensive simulations on SILVACO TCAD and compared the four TFET structures SiGe-pocket DGTFET, SiGe-pocket vertical TFET SiGe-pocket Z-shape TFET, and SiGe-pocket U-shape TFET structure on the basis of various electrical characterization parameters. Significant improvements in efficiency and sensitivity are obtained by adjusting the work function of the molybdenum (4.40–4.60 eV) catalytic gate metal to find the optimal values. The findings reveal that the SiGe-pocket U-shape TFET structure exhibits superior performance, demonstrating an ION of 8.01 × 10−4 A/μm, an ION/IOFF ratio of 1.25×1013, and an OFF current sensitivity (SOFF) of 1.262. These results highlight the enhanced sensitivity and efficacy of the proposed SiGe-pocket U-shape tunnel FET in ammonia gas sensing applications, making it a promising candidate for practical uses in environmental monitoring and industrial safety.

Multi-scale input layers and dense decoder aggregation network for COVID-19 lesion segmentation from CT scans

Accurate segmentation of COVID-19 lesions from medical images is essential for achieving precise diagnosis and developing effective treatment strategies. Unfortunately, this task presents significant challenges, owing to the complex and diverse characteristics of opaque areas, subtle differences between infected and healthy tissue, and the presence of noise in CT images. To address these difficulties, this paper designs a new deep-learning architecture (named MD-Net) based on multi-scale input layers and dense decoder aggregation network for COVID-19 lesion segmentation. In our framework, the U-shaped structure serves as the cornerstone to facilitate complex hierarchical representations essential for accurate segmentation. Then, by introducing the multi-scale input layers (MIL), the network can effectively analyze both fine-grained details and contextual information in the original image. Furthermore, we introduce an SE-Conv module in the encoder network, which can enhance the ability to identify relevant information while simultaneously suppressing the transmission of extraneous or non-lesion information. Additionally, we design a dense decoder aggregation (DDA) module to integrate feature distributions and important COVID-19 lesion information from adjacent encoder layers. Finally, we conducted a comprehensive quantitative analysis and comparison between two publicly available datasets, namely Vid-QU-EX and QaTa-COV19-v2, to assess the robustness and versatility of MD-Net in segmenting COVID-19 lesions. The experimental results show that the proposed MD-Net has superior performance compared to its competitors, and it exhibits higher scores on the Dice value, Matthews correlation coefficient (Mcc), and Jaccard index. In addition, we also conducted ablation studies on the Vid-QU-EX dataset to evaluate the contributions of each key component within the proposed architecture.

LapUNet: a novel approach to monocular depth estimation using dynamic laplacian residual U-shape networks

Monocular depth estimation is an important but challenging task. Although the performance has been improved by adopting various encoder-decoder architectures, the estimated depth maps lack structure details and clear edges due to simple repeated upsampling. To solve this problem, this paper presents the novel LapUNet (Laplacian U-shape networks), in which the encoder adopts ResNeXt101, and the decoder is constructed with the novel DLRU (dynamic Laplacian residual U-shape) module. The DLRU module based on the U-shape structure can supplement high-frequency features by fusing dynamic Laplacian residual into the process of upsampling, and the residual is dynamically learnable due to the addition of convolutional operation. Also, the ASPP (atrous spatial pyramid pooling) module is introduced to capture image context at multiple scales though multiple parallel atrous convolutional layers, and the depth map fusion module is used for combining high and low frequency features from depth maps with different spatial resolution. Experiments demonstrate that the proposed model with moderate model size is superior to other previous competitors on the KITTI and NYU Depth V2 datasets. Furthermore, 3D reconstruction and target ranging by utilizing the estimated depth maps prove the effectiveness of our proposed method.

Fish behavior recognition based on an audio-visual multimodal interactive fusion network

In light of the challenges imposed by fish behavior recognition, which arise from environmental noise and dim lighting in aquaculture environments and adversely affect the effectiveness of unimodal recognition methods based on either sound or visual cues, this paper proposes a fish behavior recognition model, Mul-SEResNet50, based on the fusion of audio and visual information. To address issues such as image blurring and indistinct sounds in aquaculture environments, which hinder the effectiveness of multimodal fusion and complementary modalities, a multimodal interaction fusion (MIF) module is introduced. This module integrates audio-visual modalities at multiple stages to achieve a more comprehensive joint feature representation. To enhance complementarity during the fusion process, we designed a U-shaped bilinear fusion structure to fully utilize multimodal information, capture cross-modal associations, and extract high-level features. Furthermore, to address the potential loss of key features, a temporal aggregation and pooling (TAP) layer is introduced to preserve more fine-grained features by extracting both the maximum and average values within pooling regions. To validate the effectiveness of the proposed model, both ablation experiments and comparative experiments are conducted. The results demonstrate that Mul-SEResNet50 achieves a 5.04 % accuracy improvement over SEResNet50 without sacrificing detection speed. Compared to the state-of-the-art U-FusionNet-ResNet50 +SENet model, Mul-SEResNet50 achieves accuracy and F1 score improvements of 0.47 % and 1.32 %, respectively. These findings confirm the efficacy of the proposed model in terms of accurately recognizing fish behavior, facilitating the precise monitoring of fish behavior.

High sensitivity fiber optic SPR sensor for selenium ion concentration detection

Balance of selenium content is of great significance to human health, and there is a need for rapid and sensitive detection of selenium ion concentrations in the human body and food. This article proposes a highly sensitive fiber surface plasmon resonance (SPR) sensor for selenium ion concentration detection. By fabricating a fiber U-shaped ring composite structure and modifying Cu3VSe4 material, the sensitivity of the fiber SPR sensor is improved to 6826.82 nm/RIU. Using 2,4-diamino-6phenyl-1,3,5-triazine to modify the surface of the sensor to achieve specific detection of selenium ion concentration. The experiment results indicate that the detection sensitivity of the sensor is 1.858 nm/(lg(pg/ml)) with range of 100 pg/ml to 1 μg/ml, and the LOD is 124.06 pg/ml. The highly sensitive SPR sensor proposed in this article, without labeling, can achieve high sensitivity and rapid detection of selenium ions, providing a new approach for trace element content detection.

Temperature compensation method for impedance signals of bolt loosening under temperature varying conditions using TransUnet

Abstract Piezoelectric transducer based electro-mechanical impedance technology is an effective and reliable detection method for bolt structure loosening under external load. However, temperature change usually causes some non-loosening factors to change the impedance spectrum of bolt structure. In order to reduce the misjudgment of loosening damage caused by temperature change, it is necessary to construct a temperature compensation model for impedance spectrum. In this paper, the convolutional structure in U-net and Transformer are effectively combined to form a TransUnet deep neural network structure suitable for input and output of one-dimensional data. Using impedance data between 10 °C and 50 °C and temperature as input to the network. After convolution operation, the convolutional block attention module is embedded in the U-net to optimize the encoder transmission characteristics and enhance the performance of the skip connection in the traditional U-shaped structure. The temperature compensation rate (TCR) is defined to measure the effect of temperature compensation. Then the lightweight convolutional neural network structure is used to recognize the bolt loosening damage of the compensated impedance signal. The generalization ability of the TransUnet was tested using impedance data that is not in the training dataset. The results show that the TransUnet proposed in the paper can realize the temperature compensation of multi-peak impedance signals. The TCR and recognition accuracy of bolt loose damage reaches reach 0.003 and 95.7%, respectively, which is 11% higher than that without temperature compensation. At 60 °C, the TCR and the identification accuracy of loosening damage can still reach 0.0055 and 90.2%, respectively, which show that the TransUnet has strong generalization ability.

UCFTNet: multi-class dtection of city critical disaster information based on U-shaped cross fusion transformer

ABSTRACT With the continuous enhancement of remote sensing data acquisition capabilities, the demand for efficient extraction of multi-class critical urban disaster information has become increasingly urgent. However, current deep learning-based models primarily focus on single disaster objects and struggle with accuracy in complex post-disaster scenarios. To address this, we propose UCFTNet, a semantic segmentation model for urban post-disaster damaged roads and buildings. UCFTNet employs a U-shaped encoder-decoder structure and ConvNext modules, integrating spatial and multi-scale information of damaged buildings and roads from different scales and channels. Additionally, the decoder uses the CCT module and CFF module to merge multi-scale features and eliminate irrelevant background interference. Finally, a weighted loss function is designed to enhance the model's focus on disaster category pixels. Experimental results demonstrate that UCFTNet improves the accuracy of extracting damaged buildings and roads by at least 5% across three urban disaster datasets, confirming its effectiveness in post-disaster damage information extraction.

An ultra-high-definition multi-exposure image fusion method based on multi-scale feature extraction

Multiple exposure image fusion is a technique used to obtain high dynamic range images. Due to its low cost and high efficiency, it has received a lot of attention from researchers in recent years. Currently, most deep learning-based multiple exposure image fusion methods extract features from different exposure images using a single feature extraction method. Some methods simply rely on two different modules to directly extract features. However, this approach inevitably leads to the loss of some feature information during the feature extraction process, thus further affecting the performance of the model. To minimize the loss of feature information as much as possible, we propose an ultra-high-definition (UHD) multiple exposure image fusion method based on multi-scale feature extraction. The method adopts a U-shaped structure to construct the overall network model, which can fully exploit the feature information at different levels. Additionally, we construct a novel hybrid stacking paradigm to combine convolutional neural networks and Transformer modules. This combined module can extract both local texture features and global color features simultaneously. To more efficiently fuse and extract features, we also design a cross-layer feature fusion module, which can adaptively learn the correlation between features at different layers. Numerous quantitative and qualitative results demonstrate that our proposed method performs well in UHD multiple exposure image fusion.

Snow Detection in Gaofen-1 Multi-Spectral Images Based on Swin-Transformer and U-Shaped Dual-Branch Encoder Structure Network with Geographic Information

Snow detection is imperative in remote sensing for various applications, including climate change monitoring, water resources management, and disaster warning. Recognizing the limitations of current deep learning algorithms in cloud and snow boundary segmentation, as well as issues like detail snow information loss and mountainous snow omission, this paper presents a novel snow detection network based on Swin-Transformer and U-shaped dual-branch encoder structure with geographic information (SD-GeoSTUNet), aiming to address the above issues. Initially, the SD-GeoSTUNet incorporates the CNN branch and Swin-Transformer branch to extract features in parallel and the Feature Aggregation Module (FAM) is designed to facilitate the detail feature aggregation via two branches. Simultaneously, an Edge-enhanced Convolution (EeConv) is introduced to promote snow boundary contour extraction in the CNN branch. In particular, auxiliary geographic information, including altitude, longitude, latitude, slope, and aspect, is encoded in the Swin-Transformer branch to enhance snow detection in mountainous regions. Experiments conducted on Levir_CS, a large-scale cloud and snow dataset originating from Gaofen-1, demonstrate that SD-GeoSTUNet achieves optimal performance with the values of 78.08%, 85.07%, and 92.89% for IoU_s, F1_s, and MPA, respectively, leading to superior cloud and snow boundary segmentation and thin cloud and snow detection. Further, ablation experiments reveal that integrating slope and aspect information effectively alleviates the omission of snow detection in mountainous areas and significantly exhibits the best vision under complex terrain. The proposed model can be used for remote sensing data with geographic information to achieve more accurate snow extraction, which is conducive to promoting the research of hydrology and agriculture with different geospatial characteristics.

Modeling test study on influence mechanism of large-span double-arch tunnels in proximity to existing tracks

This paper aims to study the influence mechanism of the large-span double-arch tunnels in proximity to existing tracks using 1 g model test. A novel non-contact device able to measure settlements automatically is developed to improve the test accuracy. During the test, the strata settlement, surrounding rock stress and track deformation induced by tunnel excavation are measured simultaneously. The results show that: (1) The excavation of the double-arch tunnel induces an asymmetric “W”-shaped horizontal surface settlement. The horizontal surface settlement near the centerline of the first excavation side is the largest. (2) The U-shaped groove structure exhibits an “M”-shaped deformation, while its structure produces bulges near the maximum horizontal surface settlement. (3) After excavation, the longitudinal surface settlement gradually decreases along the excavation direction, which causes the tilting or bottom voids of the U-shaped groove structure. The U-shaped groove structure may experience torsional shear damage under the load imposed by the train.

Unlocking the potential of CYGNSS for pan-tropical inland water mapping through multi-source data and transformer

Cyclone Global Navigation Satellite System (CyGNSS) data are widely recognized for their sensitivity to inland water bodies. However, the detection of water bodies using single CyGNSS data is subject to uncertainties, presenting challenges for large-scale and accurate water system detection. In this study, we employ CyGNSS data for regression estimation to map inland water bodies. In comparison to previous studies, we incorporate additional constraints, including topographic factors, vegetation information, soil moisture, and latitude and longitude data. Leveraging the U-shaped structure, Swin Transformer, and ContextModule, we effectively extract water body distribution information, referred to as CFRT. Through rigorous performance comparison with prevalent deep learning models, our method demonstrates remarkable accuracy. The generated water percent exhibits notable consistency with the reference data, achieving a root mean square error (RMSE) of 7.15% and a mean intersection over union of 0.778 within the reachable area of the CyGNSS data. Our approach emphasizes the significance of utilizing multi-source data to substantially enhance the accuracy of CyGNSS water system detection.

DC-Net: Divide-and-conquer for salient object detection

In this paper, to guide the model’s training process to explicitly present a progressive trend, we first introduce the concept of Divide-and-Conquer into Salient Object Detection (SOD) tasks, called DC-Net. Our DC-Net guides multiple encoders to solve different subtasks and then aggregates the feature maps with different semantic information obtained by multiple encoders into the decoder to predict the final saliency map. The decoder of DC-Net consists of newly designed two-level Residual nested-ASPP (ResASPP2) modules, which improve the sparse receptive field existing in ASPP and the disadvantage that the U-shape structure needs downsampling to obtain a large receptive field. Based on the advantage of Divide-and-Conquer’s parallel computing, we parallelize DC-Net through reparameterization, achieving competitive performance on five LR-SOD and five HR-SOD datasets under high efficiency (60 FPS and 55 FPS). Codes and results are available: https://github.com/PiggyJerry/DC-Net.

Enhancing biocementation performance in low permeability clayey soil through sand column strategy

Soil erosion is a common phenomenon which causes lots of geological and engineering disasters. Clayey soil erosion control is a hot research topic and a challenging issue for its low permeability and lack of effective infiltration of treatment solutions. In this study, a coupling microbially induced calcium carbonate precipitation (MICP)-sand column method was proposed as a promising and sustainable technique to mitigate surface clayey soil erosion with adjustable treatment depth. Seven groups of soil samples were prepared, including pure soil, MICP-treated soil, and MICP-sand column method-treated soil. A series of disintegration tests and penetration tests were conducted to investigate the method feasibility and adjusting mechanism of erosion mitigation with varying sand column heights and diameters. Compared to pure soil sample, sample treated by MICP-sand column (6 cm-height and 3 cm-diameter) method could reach a maximum reduction of 50 % in ultimate disintegration rate and a 30 % increase in the highest penetration resistance. The sample has a 7.9 cm effective treatment depth, which is 5.3 times of the MICP-treated sample. The mechanism of erosion mitigation can be attributed to that the sand column serves as a favorable path for the bacteria suspension and cementation solution in low-permeability-clayey soil and improves the effective depth of MICP treatment. Adjusting the height and diameter of sand column can change the calcium carbonate distribution, especially at the locations near the surface and around the sand column. In this study, the optimal height and diameter are 6 cm and 3 cm, respectively, and finally form a U-shape three-layer structure. The structure significantly increases the proportion of the hard crust layer and weak-cemented layer with high hydro-mechanical properties. In conclusion, the coupling MICP-sand column method with reasonable height and diameter of sand column shows the ability to control surface erosion of soil and has better long-term mitigation performance.

Crystal structure of NYN domain of Human KHNYN in complex with single strand RNA

KHNYN protein with a KH-like domain and a NYN endoribonuclease domain interacts with Zinc-finger antiviral protein (ZAP). ZAP isoforms recognize viral or cellular RNAs and recruit KHNYN to form the ZAP: KHNYN complex. Although the structures of several PIN/NYN domains have been determined, the precise substrate RNA binding mode remains poorly understood. This study presents the crystal structure of a complex of the NYN domain of KHNYN and a 7mer RNA from interferon lambda3 (IFNL3). Our structural analysis reveals that NYN domain of human KHNYN shares structural similarities with other NYN domains of ZC3H12Ã C proteins. The RNA is bound in the central groove region of the protein, facilitated by interactions including coordination by two Mg2+ ions, hydrophobic interactions, and hydrogen bonds. In the observed RNA-protein complex, the U5, A6, and U7 bases are stacked on top of one another, while U3 and U4 bases adopt an “open” conformation (as opposed to base-stacked), forming a U-shaped overall structure. Mutagenesis studies underscore the significance of residues involved in RNA binding for RNase activity. Interestingly, NYN domain of human KHNYN forms a head-to-tail dimer in the crystal, a structural feature also observed in other homologous PIN/NYN proteins, with a residue from the symmetry mate contributing to hydrophobic interactions with the bound RNA.