检索结果-内蒙古大学图书馆

EDSD: efficient driving scenes detection based on Swin Transformer

Multimedia Tools and Applications 2024年第39期83卷 87179-87198页

作者： Chen, Wei Zheng, Ruihan Jiang, Jiade Tian, Zijian Zhang, Fan Liu, Yi Beijing100083 China Key Laboratory of Intelligent Mining and Robotics Ministry of Emergency Management Beijing100083 China School of Computer Science & amp Technology China University of Mining and Technology Xuzhou221116 China

In the field of autonomous driving, the detection of targets such as vehicles, bicycles, and pedestrians in complex road conditions is of great importance. Through extensive experimentation, we have found that various vehicle targets generally occupy large sizes in the image but are easily occluded, while small targets such as pedestrians usually appear densely. The detection of targets of different sizes is an important challenge for the performance of current detectors. To address this issue, we proposed a novel hierarchical feature pyramid network structure. This structure comprises a series of CNN-Transformer variant layers, each of which is a superposition of CST neural network modules and Swin Transformer modules. In addition, considering that the huge computation of the global self-attention mechanism is difficult to be applied in the field of autonomous driving, we adopted the shifted window method in SwinFM, which effectively accelerates the inference process by replacing the traditional method by using the self-attention mechanism within the window. This study uses the Swin Transformer as a baseline. Compared to the baseline, our EDSD model improves the average accuracy by 1.8% and 3.1% on the BDD100K dataset and the KITTI dataset, respectively. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

CNN coal and rock recognition method based on hyperspectral data

引用

International Journal of Coal Science & Technology 2022年第5期9卷 59-70页

作者： Jianjian Yang Boshen Chang Yuchen Zhang Wenjie Luo Shirong Ge Miao Wu School of Mechatronics and Information Engineering China University of Mining and Technology-BeijingBeijing 100083China Key Laboratory of Intelligent Mining and Robotics Ministry of Emergency ManagementBeijingChina

Aiming at the problem of coal gangue identifcation in the current fully mechanized mining face and coal washing,this article proposed a convolution neural network(CNN)coal and rock identifcation method based on hyperspectral ***,coal and rock spectrum data were collected by a near-infrared spectrometer,and then four methods were used to flter 120 sets of collected data:frst-order diferential(FD),second-order diferential(SD),standard normal variable transformation(SNV),and multi-style *** coal and rock refectance spectrum data were pre-processed to enhance the intensity of spectral refectance and absorption characteristics,as well as efectively remove the spectral curve noise generated by instrument performance and environmental factors.A CNN model was constructed,and its advantages and disadvantages were judged based on the accuracy of the three parameter combinations(i.e.,the learning rate,the number of feature extraction layers,and the dropout rate)to generate the best CNN classifer for the hyperspectral data for rock *** experiments show that the recognition accuracy of the one-dimensional CNN model proposed in this paper reaches 94.6%.Verifcation of the advantages and efectiveness of the method were proposed in this article.

关键词： Hyperspectral data Data pre-processing 1D-CNN Coal gangue identifcation

来源：评论

学校读者我要写书评

暂无评论

Research on Ultra-sensitive Omnidirectional Strain Sensor Array for Human Motion Detections

引用

IEEE Sensors Journal 2025年第11期25卷 18851-18858页

作者： Ning, Zijun Li, Yan Zhou, Huaxu Yuan, Jingfu Zheng, Wei Qu, Yuanyuan Yang, Fuling China University of Mining and Technology-Beijing No.11 College Road of Haidian district Beijing10089 China Ministry of Emergency Management China University of Mining and Technology-Beijing Key Laboratory of Intelligent Mining and Robotics No.11 College Road of Haidian district Beijing10089 China

Human body movements are flexible and multidirectional. To obtain more accurate parameters during human motion monitoring and health managing, ultra-sensitive and omnidirectional sensing equipment is needed. Therefore, an ultra-sensitive omnidirectional strain sensor array which utilizes a bionic spider crack sensing mechanism is proposed and designed to detect the multidirectional mechanical strains over the skin generated by human activities. Holding the mechanism of the bionic crack sensing layer as foundation, the bionic crack strain sensor exhibits an ultra-high strain sensitivity of GF = 1,920,031. The proposed omnidirectional strain sensor array consists of three bionic crack strain sensors spaced 120° apart. This configuration enables the array to acquire and distinguish mechanical strains from any direction within its coverage skin area. The sensor array can be attached to the skin firmly to measure skin deformation caused by joint and muscles’ activity, as well as the angle of the wrist movements. The monitoring data obtained by the fabricated sensor array was used to train and test a badminton swing action recognition model, and eight types of badminton swing actions were successfully identified with an overall accuracy of 98.125%. Which demonstrates that the omnidirectional strain sensor array can provide high-precision human motion posture information for intelligent health management as well as medical limb monitoring. © 2001-2012 IEEE.

关键词： Motion sensors

来源：评论

学校读者我要写书评

暂无评论

Off-Design Performance of 9F Gas Turbine Based on gPROMs and BP Neural Network Model

引用

Journal of Thermal Science 2022年第1期31卷 261-272页

作者： HAO Xuedi SUN Lei CHI Jinling ZHANG Shijie School of Mechanical Electronic&Information Engineering China university of Mining&Technology-BeijingBeijing 100083China Key Laboratory of Intelligent Mining and Robotics Ministry of Emergency ManagementBeijing 100083China Key Laboratory of Advanced Energy and Power Institute of Engineering ThermophysicsChinese Academy of SciencesBeijing 100190China

Gas turbines are increasingly and widely used,whose research and production reflect a country’s industrial capacity and *** to the changeable working environment,gas turbines usually work under the condition of simultaneous changes of ambient temperature,load and ***,the current researches mainly focus on the change in single condition,and do not fully consider the simultaneous change in different *** the basis of single condition,this paper further studies the dual off-design performance of gas turbines under three conditions:temperature-load,fuel-load and ***,the whole machine model of a gas turbine is established,in which the compressor model has the greatest impact on the performance of gas ***,this paper obtains a more accurate compressor model by combining the engineering modeling advantages of gPROMs and the powerful mathematical calculation ability of MATLAB neural ***,according to the established gas turbine model,the dual off-design performance is studied,which is mainly based on the parameter of output and *** result shows that the efficiency and power output of gas turbines will decrease with the increase of ambient *** the decrease of fuel calorific value,power output and efficiency will *** the load decreases,the efficiency of the gas turbines will decrease,and these changes are consistent with the single off-design ***,when the fuel and temperature change simultaneously,only adjusting the IGV angle cannot avoid the surge when the temperature is above 30°*** this time,it is necessary to adjust the extraction rate in order to ensure the safe and stable operation of gas ***,the research on dual off-design performance of gas turbines has an important significance for the peak shaving operation of gas turbines.

关键词： off-design performance gas turbine gPROMs MATLAB neural network peak shaving operation

来源：评论

学校读者我要写书评

暂无评论

Feature Extraction Network SFPN for Edge Computing Scenarios 23

Feature Extraction Network SFPN for Edge Computing Scenarios

引用

6th International Conference on Signal Processing and Machine Learning, SPML 2023

作者： Yao, Minda Liu, Jueting Liu, Xingyu Chen, Wei Xu, Tingting Liu, Yingchun School of Computer Science & Technology China University of Mining and Technology Xuzhou221116 China Beijing100083 China Key Laboratory of Intelligent Mining and Robotics Ministry of Emergency Management China

ISBN: (纸本)9798400707575

Aiming at the problems that the existing networks do not adapt to the image characteristics of edge scenes, do not effectively process the detail information in images and have insufficient feature fusion, this thesis proposes a feature extraction network model SFPN that can better utilize the different depths and detail semantic information of the network. Firstly, the S-soft unit is designed to process the detail information of the feature map. Secondly, this thesis proposes a new secondary fusion strategy that combines the features of the middle layer are fused with the lower and higher layers respectively, and with this, SFPN Block is proposed as an auxiliary classifier, which participates in the final prediction by weighting, fully considering the prediction information of different layers of the network. Finally, a joint loss function is introduced for training, so that the feature extraction ability of the network for targets with large scale changes and small sizes is improved, and it is more adapted to the edge computing scenario of detection requirements and improve the accuracy of feature extraction. © 2023 ACM.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

An Optimal Energy Consumption Scheduling Algorithm for Dump Trucks in Mines 42

An Optimal Energy Consumption Scheduling Algorithm for Dump ...

引用

42nd Chinese Control Conference, CCC 2023

作者： Zhu, Desheng Chen, Yunchao Huang, Zhipeng Ke, Xiaotian Yang, Kehu School of Mechanical Electronic and Information Engineering China University of Mining and Technology Beijing China Ministry of Emergency Management of People's Republic of China Key Laboratory of Intelligent Mining and Robotics Beijing China Inner Mongolia Research Institute of CUMTB Ordos Inner Mongolia China

ISBN: (纸本)9789887581543

In open-pit coal mining, energy utilization is mainly focused on vehicular transportation, with dump trucks contributing to approximately 32.5% of energy consumption. Therefore, adopting fuel-saving scheduling can effectively reduce overall energy consumption. However, the lack of a practical fuel model and a rational mine scheduling model hampers the achievement of optimal energy consumption scheduling. In this paper, we present a novel algorithm for optimal energy consumption scheduling, which integrates a fuel consumption model that employs the Comprehensive Modal Emission Model (CMEM) method and a scheduling approach that considers the acceleration and deceleration processes. The effectiveness of the suggested truck scheduling model and real-time fuel consumption model approach was validated using actual operational data from dump trucks. The mine transport system is estimated to have an average specific fuel consumption (SFC) of 99.9 g/t, and the results of the case indicate a possible fuel savings of 5.5%. © 2023 Technical Committee on Control Theory, Chinese Association of Automation.

关键词： Energy utilization

来源：评论

学校读者我要写书评

暂无评论

Multi-Quadrotor Cooperative Encirclement and Capture Approach in Obstacle Environments

Multi-Quadrotor Cooperative Encirclement and Capture Approac...

引用

2025 IEEE International Conference on Mechatronics, ICM 2025

作者： Li, Zhenyu Wu, Ruixin Wang, Jin Hu, Chuanqi Zheng, Zhi Meng, Jun Lu, Guodong Zhejiang University State Key Laboratory of Fluid Power and Mechatronic Systems Hangzhou310058 China Zhejiang University Zhejiang Key Laboratory of Industrial Big Data and Robot Intelligent Systems Hangzhou310058 China Robotics Research Center of Yuyao City Ningbo315400 China College of Electrical Engineering Zhejiang University Hangzhou310027 China Center for Data Mining and Systems Biology College of Electrical Engineering Zhejiang University Hangzhou310027 China

ISBN: (纸本)9798331533892

This article proposes an algorithm designed for autonomous quadrotors to capture target quadrotors in complex obstacle environments, enabling multiple pursuers to perform distributed capture while simultaneously avoiding obstacles and collisions. We first introduce a 3D obstacle space representation method based on safety-priority reachable region (SPRR) and implement real-time construction for general point cloud maps. Subsequently, leveraging SPRR, we employ an area minimization approach to ensure the effective capture of the evader within a limited time. The outcomes are then integrated into a MINCO-based trajectory planning framework to ensure their practical execution by real quadrotors. Relying on SPRR, the optimization problem is simplified and the solution efficiency is improved. In the scenario of simulating the dynamics of real quadrotors, we compare the proposed algorithm with state-of-the-art evasion and pursuit methods, validating its effectiveness and advancements. This work presents a practical solution for multi-quadrotor collaborative capture. © 2025 IEEE.

关键词： Motion planning

来源：评论

学校读者我要写书评

暂无评论

Restoration of Dunhuang Murals on Large-scale pretraining 23

Restoration of Dunhuang Murals on Large-scale pretraining

引用

6th International Conference on Signal Processing and Machine Learning, SPML 2023

作者： Xu, Zishan Yao, Minda Chen, Wei Zhu, Min Tian, Zijian Zhang, Fan Zhang, Xiaofeng Zhang, Chuanlei China University of Mining and Technology China School of Computer Science & Technology China University of Mining and Technology Xuzhou221116 China Beijing100083 China Key Laboratory of Intelligent Mining and Robotics Ministry of Emergency Management China Department of Intensive Care Unit Xuzhou First People's Hospital Jiangsu221000 China China Shanghai Jiaotong University China Tianjin University of Science and Technology China

ISBN: (纸本)9798400707575

Dunhuang murals are a precious cultural heritage and their restoration is of vital importance. Traditional image restoration methods and methods based on generative adversarial networks (GANs) have limitations in the mural restoration task. In this paper, we propose a diffusion model-based method for restoring Dunhuang murals using the RePaint model for image restoration. We first collected and produced 5135 Dunhuang mural image data that had undergone data augmentation methods such as cropping and panning, and then pre-trained the RePaint model on a large public dataset and fine-tunedthe model on the Dunhuang mural data. The experimental results show that the fine-tuned RePaint model has significantly improved the evaluation metrics such as PSNR and SSIM in the Dunhuang mural restoration task compared to other recent excellent image restoration models. This suggests that the RePaint model has superior performance in mural restoration tasks, especially in the areas of texture generation and structure retention. This study provides a novel and effective method for the field of Dunhuang mural restoration, which is expected to provide more support for heritage conservation. © 2023 ACM.

关键词： Image reconstruction

来源：评论

学校读者我要写书评

暂无评论

Towards Efficient Path Finding and Moving Stability-Focused Trajectory Planning for Mobile Robots

Towards Efficient Path Finding and Moving Stability-Focused ...

引用

2024 IEEE International Conference on robotics and Biomimetics, ROBIO 2024

作者： Hou, Muyang Lei, Zhenya Hu, Haidong Wang, Zhong Wang, Yong Xie, Le Wang, Hesheng Shanghai Jiao Tong University Department of Automation Shanghai200240 China School of Information and Control Engineering China University of Mining and Technology Xuzhou221116 China Beijing Institute of Control Engineering Beijing100190 China Institute of Medical Robotics Shanghai Jiao Tong University Shanghai200240 China Shanghai Engineering Research Center of Intelligent Control and Management Shanghai Jiao Tong University Key Laboratory of System Control and Information Processing of Ministry of Education Key Laboratory of Marine Intelligent Equipment and System of Ministry of Education Department of Automation Shanghai200240 China

ISBN: (纸本)9781665481090

Nowadays, mobile robots play an important role in a variety of service scenarios. They need to plan and track trajectories to accomplish tasks such as delivery or guided tours. In such tasks, there are two remaining challenges: on one hand, the time and memory consumption of present path planners are still significant;on the other hand, service robots, which have a higher center of gravity, demand greater moving stability than that of current trajectory planners provide. To address these two challenges. Firstly, we propose a safety boundary first A∗ which fully utilizes environmental obstacle information to create a safety boundary and searches in it to quickly find an initial path. Secondly, we formulate the trajectory optimization problem as a nonlinear optimization problem, where the smoothness, safety, feasibility, and moving stability of the robot are taken into account. Furthermore, to constrain the robot's lateral acceleration, we design a time allocation algorithm based on non-uniform B-spline, enhancing the quality of the resulting trajectory. Simulations and real-world experiments demonstrate that our algorithms significantly improve path search efficiency, enhance moving stability, reduce the difficulty of tracking, and improve the quality of task completion. © 2024 IEEE.

关键词： Trajectories

来源：评论

学校读者我要写书评

暂无评论

An optimal energy consumption scheduling algorithm for dump trucks in mines

An optimal energy consumption scheduling algorithm for dump ...

引用

第42届中国控制会议

作者： Desheng Zhu Yunchao Chen Zhipeng Huang Xiaotian Ke Kehu Yang School of Mechanical Electronic and Information Engineering China University of Mining and Technology Key Laboratory of Intelligent Mining and Robotics Ministry of Emergency Management of People's Republic of China Inner Mongolia Research Institute of CUMTB

In open-pit coal mining, energy utilization is mainly focused on vehicular transportation, with dump trucks contributing to approximately 32.5% of energy consumption. Therefore, adopting fuel-saving scheduling can effectively reduce overall energy consumption. However, the lack of a practical fuel model and a rational mine scheduling model hampers the achievement of optimal energy consumption scheduling. In this paper, we present a novel algorithm for optimal energy consumption scheduling, which integrates a fuel consumption model that employs the Comprehensive Modal Emission Model(CMEM) method and a scheduling approach that considers the acceleration and deceleration processes. The effectiveness of the suggested truck scheduling model and real-time fuel consumption model approach was validated using actual operational data from dump trucks. The mine transport system is estimated to have an average specific fuel consumption(SFC) of 99.9 g/t, and the results of the case indicate a possible fuel savings of 5.5%.

关键词：

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：