检索结果-内蒙古大学图书馆

conference on Sensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace systems

作者： Alamdari, Ali Ghadimzadeh Ebrahimkhanlou, Arvin Drexel Univ Dept Mech Engn & Mech 3141 Chestnut St Philadelphia PA 19104 USA

ISBN: (纸本)9781510660793;9781510660809

Structural health monitoring is crucial for ensuring the safety of civil infrastructure, and crack detection is an essential component of this process. Cameras provide high-resolution images of the structure's surface, which can be analyzed to detect and locate cracks. LiDAR sensors use laser beams to scan the surface of the structure and produce detailed 3D point clouds that can be used to detect cracks and measure their dimensions. The proposed approach aims to improve the accuracy and efficiency of crack detection in SHM by integrating the complementary strengths of cameras and LiDARs in a simulation environment. The approach involves the use of an intelligent algorithm that can automatically fuse the data from the cameras and LiDARs to produce a more comprehensive and accurate representation of surface cracks. The algorithm uses a machine learning-based crack detection technique that can accurately identify and locate cracks in real-time. Furthermore, a depth camera is used to provide a denser point cloud than LiDAR of the crack. The integration of cameras and LiDARs for crack detection in SHM offers several advantages, such as improved accuracy, faster data acquisition, and reduced costs compared to traditional methods. The proposed approach addresses the challenges of data fusion, image processing, and intelligent algorithm development by offering a novel solution that leverages the strengths of both cameras and LiDARs. The findings of this study suggest that the proposed approach can significantly enhance the capabilities of SHM for crack detection. The approach offers a more accurate and efficient way of detecting cracks in real-time, which can help prevent further damage and ensure the safety of civil infrastructure.

关键词： Structural health monitoring cameras LiDAR crack detection data fusion image processing algorithms intelligent systems

来源：评论

学校读者我要写书评

暂无评论

Attention-Based Grayscale image Colorization 20

Attention-Based Grayscale Image Colorization

引用

20th IEEE International conference on Mechatronics and Automation, ICMA 2023

作者： Sun, Lei Ju, Dongjie Ke, Shi School of Electrical Engineering and Automation Binshui Xidao Extension 391 Tianjin300384 China Tianjin University of Technology Tianjin Key Laboratory for Control Theory and Complicated Industry Systems Binshui Xidao Extension 391 Tianjin300384 China

ISBN: (纸本)9798350320831

image colorization has always been a hot topic in computer vision. Since the emergence of deep learning and its excellent performance in many image-processing tasks, image colorization methods based on convolutional neural networks (CNN) emerge in an endless stream, including automatic colorization methods and A coloring method based on a reference image. In the coloring method based on the reference image, an existing problem is that an algorithm for calculating the similarity between the grayscale image and the reference image needs to be designed to map the color blocks in the reference image to the grayscale image. And we use the attention mechanism to spread color implicitly, and propose an attention-guided image colorization network, which extracts features and corresponding color information from the reference image through a reference encoding sub-network, and simultaneously passes a target encoding sub-network The network extracts grayscale features from grayscale images, uses attention to calculate the correlation between the two types of features, and then restores the features and reconstructs colors through the decoding network. Our method can quickly capture the color of the reference feature, accurately transfer it to the target feature, and effectively solve the problem of small target loss during downsampling. By validating on multiple datasets, the results show that our method is effective, outperforming multiple state-of-the-art image colorization algorithms. © 2023 IEEE.

关键词： Color

来源：评论

学校读者我要写书评

暂无评论

Reconstruction quality evaluation of Compressed Sensing image Mapping Spectrometer

Reconstruction quality evaluation of Compressed Sensing Imag...

引用

conference on AOPC - Computing Imaging Technology

作者： Yang, Shuya Ding, Xiaoming Yuan, Hao Lu, Dunqiang Yan, Qiangqiang Tianjin Normal Univ Tianjin Key Lab Wireless Mobile Commun & Power Tr Tianjin 300387 Peoples R China Shaanxi Key Lab Opt Remote Sensing & Intelligent Xian 710119 Peoples R China Xian Inst Opt & Precis Mech CAS Key Lab Spectral Imaging Technol Xian 710119 Peoples R China

ISBN: (纸本)9781510672413;9781510672406

This paper uses traditional algorithms and deep learning algorithms to recover datacube obtained by CASSI and CSIMS in order to verify that CSIMS outperforms CASSI by comparing the Peak Signal to Noise Ratio (PSNR), Structural Similarity (SSIM) and Relative spectral Quadratic Error (RQE) of the reconstructed datacube. The experimental results show that the datacube of CASSI and CSIMS can be both reconstructed by ADMM- TV algorithm which is the most effective among the traditional algorithms. PSNR of the reconstructed datacube of CASSI is 32.50 dB, while that of CSIMS is 35.53 dB, with an increase of 3.03 dB. By using deep learning algorithm, both systems improve substantially under the PnP-HSI network, with PSNR of CASSI growing to 38.85 dB and that of CSIMS growing to 41.97 dB, which can be seen that CSIMS is still 3.12 dB higher than CASSI.

关键词： Snapshot spectral imaging Compression sensing image mapper Deep learning

来源：评论

学校读者我要写书评

暂无评论

Multistatic cylindrical fast backprojection 3D image reconstruction 28

Multistatic cylindrical fast backprojection 3D image reconst...

引用

conference on Radar Sensor Technology XXViii

作者： Sheen, David M. Clark, R. Trevor Pacific Northwest Natl Lab US DOE Battelle Mem Inst POB 999 Richland WA 99352 USA

ISBN: (纸本)9781510674158;9781510674141

The Pacific Northwest National Laboratory (PNNL) has recently developed a next-generation cylindrical millimeter-wave imaging system. This system is based on linear sparse multistatic imaging arrays. Data from this system can be focused using 3D FFT-based reconstruction algorithms, which are reasonably efficient and can be performed in near real time, or by back-projection methods that are versatile and more accurate but are computationally intensive and require lengthy post-processing. Cylindrical Fast Backprojection (CFBP) is a novel image reconstruction algorithm developed at PNNL that radically increases the efficiency of backprojection and is ideally suited to microwave and millimeter-wave imaging systems based on scanned linear arrays such as body scanners in common use for aviation security screening. This method achieves its gains in efficiency by separating a full backprojection into a sequence of three steps, range focusing, vertical focusing, and lateral focusing, with intermediate results used to avoid repetitive multidimensional computation. The method is called cylindrical fast backprojection due to the use of two-dimensional stored results, or look-up tables, that have cylindrical symmetry about the linear array. The method is well suited to cylindrically scanned linear arrays but is equally valid for linear arrays scanned to form planar or arbitrary apertures. This paper describes the CFBP algorithm and validates its performance using simulated data.

关键词： Millimeter-wave imaging microwave imaging concealed-weapon detection security screening

来源：评论

学校读者我要写书评

暂无评论

Comparison of Parallel Implementation Strategies in GPU-Accelerated System-on-Chip Under Proton Irradiation

引用

IEEE TRANSACTIONS ON NUCLEAR SCIENCE 2022年第3期69卷 444-452页

作者： Badia, Jose M. Leon, German Belloch, Jose A. Garcia-Valderas, Mario Lindoso, Almudena Entrena, Luis Univ Jaume 1 Dept Ingn & Ciencias Comp Castellon de La Plana 12071 Castellon Spain Univ Carlos III Madrid Dept Tecnol Elect Leganes 28911 Spain

Commercial off-the-shelf (COTS) system-on-chip (SoC) are becoming widespread in embedded systems. Many of them include a multicore central processing unit (CPU) and a high-end graphics processing unit (GPU). They combine high computational performance with low power consumption and flexible multilevel parallelism. This kind of device is also being considered for radiation environments where large amounts of data must be processed or compute-intensive applications must be executed. In this article, we compare three different strategies to perform matrix multiplication in the GPU of a Tegra TK1 SoC. Our aim is to analyze how the different use of the resources of the GPU influences not only the computational performance of the algorithm, but also its radiation sensitivity. Radiation experiments with protons were performed to compare the behavior of the three strategies. Experimental results show that most of the errors force a reboot of the platform. The number of errors is directly related with how the algorithms use the internal memories of the GPU and increases with the matrix size. It is also related with the number of transactions with the global memory, which in our experiments is not affected by the radiation. Results show that the smallest cross section is obtained with the fastest algorithm, even if it uses the cores of the GPU more intensively.

关键词： Graphics processing units Performance evaluation Protons Radiation effects Sensitivity Computer architecture Instruction sets Embedded systems graphics processing unit (GPU) parallelization proton irradiation

来源：评论

学校读者我要写书评

暂无评论

Localization and Distance Measurement of Intruding Objects in Power Transmission Lines Based on image and Point Cloud Fusion 43

Localization and Distance Measurement of Intruding Objects i...

引用

43rd Chinese Control conference, CCC 2024

作者： Zhang, Tianhao Qian, Kun Deng, Feng Southeast University School of Automation Nanjing210096 China Nari Technology Co. Ltd Nanjing211102 China

ISBN: (纸本)9789887581581

To safeguard the operation of ultra-high voltage transmission lines, this study introduces a technique for dynamically monitoring potential security threats to these lines. This technique integrates images and point cloud data to pinpoint and measure the distance of objects intruding into the vicinity of the transmission lines. The methodology focuses on achieving high-precision localization and distance measurement of such objects. Key innovations include advanced image object detection combined with efficient processing of point cloud data. This allows for precise location and reliable distance assessment of objects at significant distances. The process begins with the creation of detailed scene point clouds using unmanned aerial vehicle (UAV) SLAM mapping technology. These point clouds are then aligned with the monitoring LIDAR coordinate system through registration algorithms to use the catenary line model to reconstruct transmission lines. The intrusion's highest point is precisely located by correlating two-dimensional image detection results with the image-to-point cloud mapping and extracting the intruding object's point cloud using Euclidean clustering. The method concludes by calculating the minimum distance from the top of the intrusion to the nearest transmission line, using the catenary line equation. Experimental results indicate that this approach can detect intruders and gauge their distances efficiently and accurately across various ranges and environmental conditions, showcasing exceptional accuracy and robustness, particularly in long-range object detection scenarios. © 2024 Technical Committee on Control Theory, Chinese Association of Automation.

关键词： Unmanned aerial vehicles (UAV)

来源：评论

学校读者我要写书评

暂无评论

Automatic Segmentation and Extraction of Skin Lesion in Dermoscopic images using image processing

Automatic Segmentation and Extraction of Skin Lesion in Derm...

引用

2023 International conference on Innovative Computing, Intelligent Communication and Smart Electrical systems, ICSES 2023

作者： Angeline Valentina Sweety, A. Maria Delphiya, A. Rejoline Vincima, S. Kavya, N. Jeresha, J.M. Heltin Genitha, C. St.Joseph's Institute of Technology Department of CSE Chennai India St.Joseph's College of Engineering Department of IT Chennai India

ISBN: (纸本)9798350319200

Dermoscopy, an epiluminescence light microscope that magnifies lesions and enables investigation down to the dermo-epidermal interface, is a non-invasive method that doctors may use to help with the diagnosis of melanoma. The research on melanoma skin cancers and the many subtypes of each is included in our paper. The four steps of the approach are preprocessing, segmentation, extraction of features, and classification. The artifact-filled image is preprocessed during the preprocessing stage using noise filtering and contrast sharpening. The processed image (Fuzzy C-Means method) immediately initiates a segmentation method. K-means clustering outperforms fuzzy C-Means in terms of outcomes for overlapping data sets. Gabor then does feature extraction. The Gabor filter and Glcm perform feature extraction in the next stage. By computing the pair of pixels with defined values and appearing in a given spatial image in an image, Glcm is used to characterise the image. The Gabor filter determines if a picture contains any certain frequency content. Last but not least, for classification, the bulk of the algorithms described in the literature rely on a feature extraction stage that is followed by a classifier that takes the retrieved features as inputs. However, Support Vector Machine is utilised for classification in our method. SVM produces spaces with many dimensions. The picture is categorised according to the type of melanoma malignancies in the classification output. © 2023 IEEE.

关键词： Contrast Stretching Feature Extraction Fuzzy C-Means Gabor Filter Glcm Noise Filtering Preprocessing Segmentation Svm

来源：评论

学校读者我要写书评

暂无评论

An image Encryption Algorithm Suitable for Medical images 16

An Image Encryption Algorithm Suitable for Medical Images

引用

16th IEEE International Scientific conference on Informatics, Informatics 2022

作者： Oravec, Jakub Ovsenik, Lubos Lapcak, Maros Zdravecky, Norbert Andrejcik, Samuel Technical University of Košice Feei Dept. of Electronics and Multimedia Communications Košice Slovakia

ISBN: (纸本)9798350310344

Security of data becomes more and more important each day. The importance is even more pronounced if the data contain sensitive information, such as those shown in medical images. There are already several solutions that can be used in order to protect medical images during their storage or transmission. The most popular of those is probably encryption by conventional encryption algorithms such as Advanced Encryption Standard. This proposal tries to investigate possibility of using dedicated image encryption algorithms that have applications in fields of steganography and biometric systems for encryption of medical images. The dedicated image encryption algorithms can be easily adjusted for processing medical images, what can be verified by several experiments. Main advantage of this proposal lies in an addition of a substitution block, which should increase the security of whole algorithm. The performance of our solution will be evaluated by several numerical parameters and those will be compared with results reported in similar works. © 2022 IEEE.

关键词： Cryptography

来源：评论

学校读者我要写书评

暂无评论

image Captioning Using Xception-Long Short-Term Memory 1

引用

9th International conference on Mining Intelligence and Knowledge Exploration, MIKE 2023

作者： Panchal, Nisha Garg, Dweepna Gujarat Changa Anand India Gujarat Changa Anand India

ISBN: (数字)9783031440847

ISBN: (纸本)9783031440830

Deep Learning has shown great potential in developing applications capable of automatically generating captions or descriptions for images and video frames. The critical components of this process are image processing and natural language processing, which play a crucial role in captioning images and videos. These applications can be used in several areas, including robotic vision systems, assisting people with visual impairments, generating metadata for search engines, answering visual questions, visual grounding, and more. This paper discusses various working algorithms such as a combination of CNN-RNN, encoder-decoder, attention mechanisms, and transformation models with evaluation matrices, datasets, and limitations of existing models. Xception-LSTM shows great potential compared to the traditional encoding-decoding model using BLEU and METEOR evaluation matrics. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Long short-term memory

来源：评论

学校读者我要写书评

暂无评论

Real-Time image and Video processing Applications Using Raspberry Pi 1

Real-Time Image and Video Processing Applications Using Rasp...

引用

1st IEEE Industrial Electronics Society Annual On-Line conference, ONCON 2022

作者： Yildirim, Merve Karaduman, Ozgur Kurum, Hasan University of Firat Department of Electrical and Electronics Engineering Elazig Turkey University of Firat Department of Software Engineering Elazig Turkey

ISBN: (纸本)9798350398069

image and video processing applications have significant importance in many areas which are the industrial and medical applications, especially the vehicular technology. To provide safe driving, Driving Assistance systems (DAS) including detection and tracking of a road situation, objects and behavior of the driver should be considered by using the image processing algorithms. Since these applications need the processors having high speed and power, a Raspberry Pi platform meeting these specifications is used in this study. A Canny edge detector which is one of the important edge detection methods, face and eye detection, the road line tracking, motion detection, and object detection applications are handled. Conversely the studies in the literature which are separately examined these techniques, all of them are realized as a real-time based on the Rapberry Pi according to a video image taken by a camera. For this reason, OpenCv and TensorFlow libraries having widely computer vision algorithms and Python software language are used. It is observed that the results of all image and video processing applications are successful and satisfying. As a result, a safe and comfortable driving can be provided thanks to the proposed methods. © 2022 IEEE.

关键词： Python

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：