检索结果-内蒙古大学图书馆

A Fine-Grained Object Detection Model for Aerial Images Based on YOLOv5 Deep Neural Network

Chinese Journal of Electronics 2023年第1期32卷 51-63页

作者： ZHANG Rui XIE Cong DENG Liwei Heilongjiang Provincial Key Laboratory of Complex Intelligent System and Integration School of AutomationHarbin University of Science and Technology

Many advanced object detection algorithms are mainly based on natural scenes object and rarely dedicated to fine-grained objects. This seriously limits the application of these advanced detection algorithms in remote sensing object detection. How to apply horizontal detection in remote sensing images has important research significance. The mainstream remote sensing object detection algorithms achieve this task by angle regression, but the periodicity of angle leads to very large losses in this regression method, which increases the difficulty of model learning. Circular smooth label(CSL)solved this problem well by transforming the regression of angle into a classification form. YOLOv5 combines many excellent modules and methods in recent years, which greatly improves the detection accuracy of small *** use YOLOv5 as a baseline and combine the CSL method to learn the angle of arbitrarily oriented targets,and distinguish the fine-grained between instance classes by adding an attention mechanism module to accomplish the fine-grained target detection task for remote sensing images. Our improved model achieves an average category accuracy of 39.2% on the FAIR1M dataset. Although our method does not achieve satisfactory results,this approach is very efficient and simple, reducing the hardware requirements of the model.

关键词： Deep learning Neural networks Object detection Transformers Hardware Classification algorithms Task analysis

来源：评论

学校读者我要写书评

暂无评论

Multi-modal Bionic Motion Analysis of A Cpg-controlled Pneumatic Soft Robot

引用

Journal of Bionic Engineering 2024年第5期21卷 2247-2257页

作者： Yu Zhang Peiyu Huang Dongjie Li Jiangyu Zhou Yu Li Bo You Yanhe Zhu Heilongjiang Provincial Key Laboratory of Complex Intelligent System and Integration Harbin University of Science and TechnologyHarbin150080China State Key Laboratory of Robotics and System Harbin Institute of Technology(HIT)Harbin150001China

This paper designs a soft robot with a multi-chamber,multi-airbag mimicking soft biological structure,where the airbags of the same chamber are interconnected with each *** upper and lower chambers are separated by an intermediate layer(thin plate),which is extended and widened to achieve robot movement and *** applying pressure to the different chambers of the soft robot,it is possible to produce a variety of bionic movements of the inchworm and *** to the strong nonlinearity and infinite number of degrees of freedom properties of the material,it is impossible to obtain the analytical solution of the bending morphology and pressure of the soft robot ***,a method to establish a mathematical model of soft robot deformation based on the classical stacked plate theory is proposed,and a chain composite model of soft robot bending motion is established based on the large-deflection modeling *** paper proposes a method to generate a multi-mode soft robot motion control based on the Central Pattern Generator(CPG)using a single controller,which achieves the switching of sine wave-like patterns,half-wave-like patterns,and dragging patterns by adjust-ing frequency,amplitude and period of ***,a pneumatic control platform is built for the validation of the theoretical model and different experimental models of the motion of the *** comparation of the different motion modes of the soft robot under similar non-load and load conditions.

关键词： Soft robot Bionic multi-modal movement Laminated plate theory CPG control

来源：评论

学校读者我要写书评

暂无评论

Contact Force Optimization to Enhance Fault-tolerant Motion Stability of a Hexapod Robot

引用

Journal of Bionic Engineering 2024年第5期21卷 2199-2214页

作者： Bo You Shangdong Shi Chen Chen Jiayu Li Nan Li Liang Ding The Key Laboratory of Intelligent Technology for Cutting and Manufacturing Ministry of Education and the Heilongjiang Provincial Key Laboratory of Complex Intelligent System and Integration Harbin University of Science and TechnologyHarbin150080China The State Key Laboratory of Robotics and System Harbin Institute of TechnologyHarbin150001China

This paper presents a novel method for optimizing the contact force of a hexapod robot to enhance its dynamic motion stability when one of its legs *** proposed approach aims to improve the Force Angle Stability Margin(FASM)and adapt the foot trajectory through contact force optimization to achieve safe and stable motion on various *** foot force optimization approach is designed to optimize the FASM,a factor rarely considered in existing contact force optimization *** formulating the problem of enhancing the motion stability of the hexapod robot as a Quadratic Programming(QP)optimization problem,this approach can effectively address this *** of a hexapod robot using a fault-tolerant gait,along with real field experiments,were conducted to validate the effectiveness and feasibility of the contact force optimization *** results demonstrate that this approach can be used to design a motion controller for a hexapod robot with a significantly improved motion *** summary,the proposed contact force optimization method offers a promising solution for enhancing the motion stability of hexapod robots with single leg failures and has the potential to significantly advance the development of fault-tolerant hexapod robots for various applications.

关键词： Hexapod robot Fault-tolerant Stability margin Contact force optimization

来源：评论

学校读者我要写书评

暂无评论

Observer-based event-triggered impulsive synchronization control of heterogeneous complex networks with bit-rate constraints

引用

Information Sciences 2025年 717卷

作者： Guo, Jing Huang, Ling Shi, Peng School of Measurement and Communication Harbin University of Science and Technology Heilongjiang Harbin150080 China School of Automation Harbin University of Science and Technology Heilongjiang Harbin150080 China Heilongjiang Provincial Key Laboratory of Complex Intelligent System and Integration Heilongjiang Harbin150080 China The University of Adelaide SA5005 Australia

This paper proposes an observer-based event-triggered impulsive control synchronization for heterogeneous complex networks under bit-rate constraints. An isolated target node and observer systems are introduced to derive synchronization and observation error dynamics. A dynamic quantization-based encoding–decoding strategy is designed for wireless communication with bit-rate constraints, and the relationship between encoded error and bit rate is established. To conserve resources, a new event-triggered mechanism based on the sampling period is proposed. This mechanism reduces unnecessary triggers while ensuring effective impulsive control through a maximum triggering interval. Identical decoder models on both sensor and controller sides eliminate additional communication for control signal transmission. An impulsive controller is designed using decoded signals at triggering moments, and a new framework for closed-loop augmented error systems is constructed. Sufficient conditions for bounded synchronization are derived using hybrid systems theory and Lyapunov stability theory. The relationship among heterogeneity, bit rate, and synchronization error is demonstrated. An optimization algorithm is proposed to minimize a weighted sum of the upper bound of the augmented errors and the maximum triggering interval. Numerical simulations verify the effectiveness of the proposed approaches. © 2025

关键词： Network coding

来源：评论

学校读者我要写书评

暂无评论

Two Stages Segmentation Algorithm of Breast Tumor in DCE-MRI Based on Multi-Scale Feature and Boundary Attention Mechanism

引用

Computers, Materials & Continua 2024年第7期80卷 1543-1561页

作者： Bing Li Liangyu Wang Xia Liu Hongbin Fan Bo Wang Shoudi Tong School of Automation Harbin University of Science and TechnologyHarbin150080China Heilongjiang Provincial Key Laboratory of Complex Intelligent System and Integration Harbin150080China School of Computer Engineering Technology Guangdong Institute of Science and TechnologyZhuhai519090China

Nuclearmagnetic resonance imaging of breasts often presents complex *** tumors exhibit varying sizes,uneven intensity,and indistinct *** characteristics can lead to challenges such as low accuracy and incorrect segmentation during tumor ***,we propose a two-stage breast tumor segmentation method leveraging multi-scale features and boundary attention ***,the breast region of interest is extracted to isolate the breast area from surrounding tissues and ***,we devise a fusion network incorporatingmulti-scale features and boundary attentionmechanisms for breast tumor *** incorporate multi-scale parallel dilated convolution modules into the network,enhancing its capability to segment tumors of various sizes through multi-scale convolution and novel fusion ***,attention and boundary detection modules are included to augment the network’s capacity to locate tumors by capturing nonlocal dependencies in both spatial and channel ***,a hybrid loss function with boundary weight is employed to address sample class imbalance issues and enhance the network’s boundary maintenance capability through additional *** was evaluated using breast data from 207 patients at RuijinHospital,resulting in a 6.64%increase in Dice similarity coefficient compared to the *** results demonstrate the superiority of the method over other segmentation techniques,with fewer model parameters.

关键词： Dynamic contrast-enhanced magnetic resonance imaging(DCE-MRI) breast tumor segmentation multi-scale dilated convolution boundary attention the hybrid loss function with boundary weight

来源：评论

学校读者我要写书评

暂无评论

Design, Analysis, and Experiment of a Novel Ultrasonic Printing system

Design, Analysis, and Experiment of a Novel Ultrasonic Print...

引用

2022 IEEE International Conference on Manipulation, Manufacturing and Measurement on the Nanoscale, 3M-NANO 2022

作者： Pei, Zhichao Zhao, Haoxiang Li, Dongjie Wang, Lefeng Rong, Weibin Sun, Lining Harbin Institute of Technology State Key Laboratory of Robotics and Systems Harbin China Harbin University of Science and Technology Heilongjiang Provincial Key Laboratory of Complex Intelligent System and Integration Harbin China

ISBN: (数字)9781665475433

ISBN: (纸本)9781665475433

Ultrasonic-driven printing technology is widely used in bioprinting because of its high precision, good biocompatibility, and low cost. However, the conventional ultrasonic printing technology still has problems such as cross-contamination of bio-ink during storage, transportation, and processing, and cumbersome ink replacement process. Therefore, a new ultrasonic printing device is developed in this paper, which uses PZT plates as the driving element. And a fluid sub-system has been fabricated based on the self-made glass micro-nozzle, which is connected to the drive element using ultrasonic couplant gel. The separation of the fluid system and the drive system has been realized through the modular design, thereby realizing the rapid replacement of different inks and this relatively independent fluid system is of great benefit to better sealing and aseptic operation. In addition, a digital model is built to analyze its working mechanism. And an experimental system is established to characterize the performance of the device. The results show that the system can generate droplets stably and accurately, and the droplet volume can be controlled by the excitation voltage amplitude and pulse width. © 2022 IEEE.

关键词： Biocompatibility

来源：评论

学校读者我要写书评

暂无评论

MDRN: Multi-distillation residual network for efficient MR image super-resolution

引用

Mathematical Biosciences and Engineering 2024年第10期21卷 7421-7434页

作者： Deng, Liwei Chen, Jingyi Yang, Xin Huang, Sijuan Heilongjiang Provincial Key Laboratory of Complex Intelligent System and Integration School of Automation Harbin University of Science and Technology Heilongjiang Harbin150080 China State Key Laboratory of Oncology in South China Guangdong Key Laboratory of Nasopharyngeal Carcinoma Diagnosis and Therapy Guangdong Provincial Clinical Research Center for Cancer Sun Yat-sen University Cancer Center Guangdong Guangzhou510060 China Guangdong Esophageal Cancer Institute Guangdong Guangzhou510060 China

Super-resolution (SR) of magnetic resonance imaging (MRI) is gaining increasing attention for being able to provide detailed anatomical information. However, current SR methods often use the complex convolutional network for feature extraction, which is difficult to train and not suitable for limited computation resources in the medical scenario. To tackle these bottlenecks, we propose a multidistillation residual network (MDRN) for more differential feature refinement, which has a superior trade-off between reconstruction accuracy and computation cost. Specifically, a novel feature multidistillation residual block with a contrast-aware channel attention module was designed to make the residual features more focused on low-vision information, which maximizes the power of MDRN. Comprehensive experiments demonstrate the superiority of our MDRN over state-of-the-art methods in reconstruction quality and efficiency. Our method outperforms other existing methods in peak signal-noise ratio by up to 0.44-1.82 dB in 4× scale when GPU memory and runtime are lower than in other SR methods. The source code will be available at https://***/Jennieyy/MDRN. © 2024 the Author(s).

关键词： Magnetic resonance imaging

来源：评论

学校读者我要写书评

暂无评论

PDS-UKAN: Subdivision hopping connected to the U-KAN network for medical image segmentation

引用

Computerized Medical Imaging and Graphics 2025年 124卷

作者： Liwei Deng Wenbo Wang Songyu Chen Xin Yang Sijuan Huang Jing Wang Heilongjiang Provincial Key Laboratory of Complex Intelligent System and Integration School of Automation Harbin University of Science and Technology Harbin Heilongjiang 150080 China School of Computer Science and Technology Harbin University of Science and Technology Harbin Heilongjiang 150080 China State Key Laboratory of Oncology in South China Guangdong Key Laboratory of Nasopharyngeal Carcinoma Diagnosis and Therapy Guangdong Provincial Clinical Research Center for Cancer Sun Yat-sen University Cancer Center Guangzhou Guangdong 510060 China Guangdong Esophageal Cancer Institute Guangzhou Guangdong 510060 China Institute for Brain Research and Rehabilitation South China Normal University Guangzhou 510631 China

Accurate and efficient segmentation of medical images plays a vital role in clinical tasks, such as diagnostic procedures and planning treatments. Traditional U-shaped encoder-decoder architectures, built on convolutional and transformer-based networks, have shown strong performance in medical image processing. However, the simple skip connections commonly used in these networks face limitations, such as insufficient nonlinear modeling capacity, weak global multiscale context modeling, and limited interpretability. To address these challenges, this study proposes the PDS-UKAN network, an innovative subdivision-based U-KAN architecture, designed to improve segmentation accuracy. The PDS-UKAN incorporates a PKAN module—comprising partial convolutions and Kolmogorov - Arnold network layers—into the encoder bottleneck, enhancing the network's nonlinear modeling and interpretability. Additionally, the proposed Dual-Branch Convolutional Boundary Enhancement Module (DBE) focuses on pixel-level boundary refinement, improving edge detail preservation in shallow skip connections. Meanwhile, the Skip Connection Channel Spatial Attention Module (SCCSA) mechanism is applied in the deeper skip connections to strengthen cross-dimensional interactions between channels and spatial features, mitigating the loss of spatial information due to downsampling. Extensive experiments across multiple medical imaging datasets demonstrate that PDS-UKAN consistently achieves superior performance compared to state-of-the-art (SOTA) methods.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Meta-Learning Multi-Scale Radiology Medical Image Super-Resolution

引用

Computers, Materials & Continua 2023年第5期75卷 2671-2684页

作者： Liwei Deng Yuanzhi Zhang Xin Yang Sijuan Huang Jing Wang Heilongjiang Provincial Key Laboratory of Complex Intelligent System and Integration School of AutomationHarbin University of Science and TechnologyHarbin150080HeilongjiangChina Department of Radiation Oncology Sun Yat-sen University Cancer CenterState Key Laboratory of Oncology in South ChinaCollaborative Innovation Center for Cancer MedicineGuangdong Key Laboratory of Nasopharyngeal Carcinoma Diagnosis and TherapyGuangzhou510060GuangdongChina Faculty of Rehabilitation Medicine Biofeedback LaboratoryGuangzhou Xinhua UniversityGuangzhou510520GuangdongChina

High-resolution medical images have important medical value,but are difficult to obtain *** by hardware equipment and patient’s physical condition,the resolution of directly acquired medical images is often not ***,many researchers have thought of using super-resolution algorithms for secondary processing to obtain high-resolution medical ***,current super-resolution algorithms only work on a single scale,and multiple networks need to be trained when super-resolution images of different scales are *** definitely raises the cost of acquiring high-resolution medical ***,we propose a multi-scale superresolution algorithm using *** algorithm combines a metalearning approach with an enhanced depth of residual super-resolution network to design a meta-upscale *** meta-upscale module utilizes the weight prediction property of meta-learning and is able to perform the super-resolution task of medical images at any ***,we design a non-integer mapping relation for super-resolution,which allows the network to be trained under non-integer magnification *** to the state-of-the-art single-image super-resolution algorithm on computed tomography images of the pelvic *** meta-learning multiscale superresolution algorithm obtained a surpassing of about 2%at a smaller model *** on different parts proves the high generalizability of our ***-scale super-resolution algorithms using meta-learning can compensate for hardware device defects and reduce secondary harm to patients while obtaining high-resolution medical *** can be of great use in imaging related fields.

关键词： Super resolution deep learning meta learning computed tomography

来源：评论

学校读者我要写书评

暂无评论

A Novel Unsupervised MRI Synthetic CT Image Generation Framework with Registration Network

引用

Computers, Materials & Continua 2023年第11期77卷 2271-2287页

作者： Liwei Deng Henan Sun Jing Wang Sijuan Huang Xin Yang Heilongjiang Provincial Key Laboratory of Complex Intelligent System and Integration School of AutomationHarbin University of Science and TechnologyHarbin150080China Institute for Brain Research and Rehabilitation South China Normal UniversityGuangzhou510631China Department of Radiation Oncology Sun Yat-sen University Cancer CenterState Key Laboratory of Oncology in South ChinaCollaborative Innovation Center for Cancer MedicineGuangdong Key Laboratory of Nasopharyngeal Carcinoma Diagnosis and TherapyGuangzhou510060China

In recent years,radiotherapy based only on Magnetic Resonance(MR)images has become a hot spot for radiotherapy planning research in the current medical ***,functional computed tomography(CT)is still needed for dose calculation in the *** deep-learning approaches to synthesized CT images from MR images have raised much research interest,making radiotherapy based only on MR images *** this paper,we proposed a novel unsupervised image synthesis framework with registration *** paper aims to enforce the constraints between the reconstructed image and the input image by registering the reconstructed image with the input image and registering the cycle-consistent image with the input ***,this paper added ConvNeXt blocks to the network and used large kernel convolutional layers to improve the network’s ability to extract *** research used the collected head and neck data of 180 patients with nasopharyngeal carcinoma to experiment and evaluate the training model with four evaluation *** the same time,this research made a quantitative comparison of several commonly used model *** evaluate the model performance in four evaluation metrics which achieve Mean Absolute Error(MAE),Root Mean Square Error(RMSE),Peak Signal-to-Noise Ratio(PSNR),and Structural Similarity(SSIM)are 18.55±1.44,86.91±4.31,33.45±0.74 and 0.960±0.005,*** with other methods,MAE decreased by 2.17,RMSE decreased by 7.82,PSNR increased by 0.76,and SSIM increased by *** results show that the model proposed in this paper outperforms other methods in the quality of image *** work in this paper is of guiding significance to the study of MR-only radiotherapy planning.

关键词： MRI-CT image synthesis variational auto-encoder medical image translation MRI-only based radiotherapy

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：