检索结果-内蒙古大学图书馆

International Symposium on parallel and distributed processing with Applications, ISPA

作者： Shuang Yang Yaobin Wang Ling Li Jiawei Qin Guotang Bi School of Computer Science and Technology Southwest University of Science and Technology Mianyang China

ISBN: (数字)9798331509712

ISBN: (纸本)9798331509729

Discrete cosine transform for 8×8 block(DCT8×8) is widely used in image compression due to its high signal decorrelation rate. Current research for DCT is mainly focused on CPU and single GPU platforms, and the exploration of multi-GPU architectures is still insufficient. With the rapid increase of image data volume, the DCT8×8 algorithm on CPU and single GPU architectures can no longer meet the demand of efficient computation. Therefore, optimizing DCT8×8 algorithms on multi-GPU architectures becomes particularly important. To address this challenge, this paper explores the porting and optimization strategies of DCT algorithms on the new MGPUSim ***, we extend the instruction implementation of MG-PUSim to accomplish the porting of the original DCT8×8 algorithm (O-DCT), and then propose a kernel implementation strategy for optimizing the DCT8×8 algorithm by setting the size of the workgroup appropriately so that the kernel function can take advantage of the symmetry of the transformation matrix to reduce the redundant computation, which we refer to as the M-DCT. We conducted experiments on multiple grayscale images with different resolutions. The experimental results show that O-DCT performs well on MGPUSim, achieving an average 1.93× speedup ratio compared to the original DCT8×8 kernel (kernel1) in the CUDA SDK. Further experimental results show that M-DCT improves the execution efficiency on MGPUSim by 1.76× relative to O-DCT, and also achieves an average of 1.3× acceleration ratio compared to the optimized DCT8×8 kernel (kernel2) in the CUDA SDK, which improves the computational performance of the DCT8×8 algorithm on multi-GPU architectures. It provides new ideas and methods for future research on image compression on multi-GPU architectures.

关键词： image resolution image coding Instruction sets Graphics processing units Computer architecture Gray-scale Computational efficiency Discrete cosine transforms Kernel Optimization

来源：评论

学校读者我要写书评

暂无评论

RAMMM: A Rapid Attention-Based Multimodal Modification Model for Massive image Generation

RAMMM: A Rapid Attention-Based Multimodal Modification Model...

引用

International Conference on parallel and distributed Systems (ICPADS)

作者： Zhenyuan Xu Xuewen Chen Taotao Wang Wei Quan Yi Zhu Nanjing Audit University Nanjing China

Nowadays, deep learning (DL) techniques have found extensive applications in the Internet of Things (IoT) community, such as autonomous driving and medical diagnosis. Despite these successful implementations, elevated data collection expenses, data confidentiality concerns, and privacy issues all contribute to the complexity and costliness of data acquisition in DL-based image processing. image modification methods based on diffusion models can subtly modify images to generate new images with similar content to the original image but with unique differences, providing a powerful tool for data augmentation. Nonetheless, diffusion models exhibit constraints in cross-modal image modification, such as sensitivity to prompt, model complexity, and user-friendliness. To address these issues, this paper proposes a rapid attention-based multimodal modification model (RAMMM) to facilitate straightforward and efficient text-guided image modification. Our proposed RAMMM primarily enhances the text processing and image generation procedures via an attention mechanism, thus improving the ability in capturing semantic and contextual information within the text. Consequently, RAMMM excels in generating high-quality sample images aligned with the description provided for the modified text. In addition, by utilizing text-guided image modification, RAMMM is capable of generating batches of image samples systematically to augment the dataset size. Evaluation results demonstrate that RAMMM can improve the performance and generalization capabilities of diffusion models by enhancing both the quality and quantity of the dataset. When employed for data augmentation on the CIFAR10 dataset, RAMMM achieves a 1.5% increase in target recognition accuracy.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Scaling Single-image Super-Resolution Training on Modern HPC Clusters: Early Experiences

Scaling Single-Image Super-Resolution Training on Modern HPC...

引用

35th IEEE International parallel and distributed processing Symposium (IPDPS)

作者： Anthony, Quentin Xu, Lang Subramoni, Hari Panda, Dhabaleswar K. Dk Ohio State Univ Dept Comp Sci & Engn Columbus OH 43210 USA

ISBN: (纸本)9781665435772

Deep Learning (DL) models for super-resolution (DLSR) are an emerging trend in response to the growth of ML/DL applications requiring high-resolution images. DLSR methods have also shown promise in domains such as medical imaging, surveillance, and microscopy. However, DLSR models are extremely computationally demanding, and require unreasonably long training times on modern Volta GPUs. In our experiments, we observed only 10.3 images/second on a single Volta GPU for training EDSR, a state-of-the-art DLSR model for single-image super-resolution. In comparison, a Volta GPU can process 360 images/second while training ResNet-50, a state-of-the-art model for image classification. Therefore, we believe supercomputers provide a good candidate to speed up DLSR model training. In this paper, we select EDSR as the representative DLSR PyTorch model. Further, we introduce Horovod-based distributed EDSR training. However, we observed poor default EDSR scaling performance on the Lassen HPC system at Lawrence Livermore National Laboratory. To investigate the performance degradations, we perform exhaustive communication profiling. These profiling insights are then used to optimize CUDA-Aware MPI for DLSR models by ensuring advanced MPI designs involving CUDA IPC and registration caching are properly applied by DL frameworks. We present a comprehensive scaling study of EDSR with MVAPICH2-GDR and NCCL up to 512 GPUs on Lassen. We demonstrate an improvement in scaling efficiency by 15.6% over default Horovod training, which translates to a 1.26x speedup in training performance.

关键词： DNN Training Performance Characterization MVAPICH2 MPI PyTorch Horovod Super Resolution

来源：评论

学校读者我要写书评

暂无评论

Industrial CT Truncated Projection Data Reconstruction System Based on Artificial Intelligence

Industrial CT Truncated Projection Data Reconstruction Syste...

引用

Signal processing and Intelligent Computing (SPIC), International Conference on

作者： Yang Yang Songjie Wu Liaoning Institute of Science and Technology Shenyang China

ISBN: (数字)9798350368888

ISBN: (纸本)9798350368895

In industry, Computer Tomography (CT) is a key technology used for non-destructive testing and evaluation of internal structures, but significant challenges remain in the processing of truncated projection data. The aim of this study is to investigate an artificial intelligence-based reconstruction system for industrial CT truncated projection data. We investigate the common data truncation problem in industrial CT technology and propose a new data reconstruction method. The method uses the ideas of full variance minimization and segmentation constants to achieve accurate reconstruction of truncated projection data by means of parallel computation. Our study validates the effectiveness and superiority of the method through comparative experiments. Compared with conventional methods, our system excels in both accuracy and stability of reconstruction results, and is better able to handle the challenges posed by truncated data. In addition, we explore the practical application value of the system, demonstrating its potential use in industrial CT. From a single-core to a dual-core processor, the processing time is reduced from 120 to 60 seconds, showing a doubling of the processing speed, which indicates that tasks are efficiently distributed and executed among multiple cores. The research results in this paper will have a positive impact on the development and application of industrial CT technology and promote its wide application in industrial production and quality control.

关键词： Technological innovation image segmentation Accuracy TV Computed tomography parallel processing Minimization Stability analysis image reconstruction Testing

来源：评论

学校读者我要写书评

暂无评论

Research on parallel Detection Technology of Remote Sensing Object Based on Deep Learning 4

Research on Parallel Detection Technology of Remote Sensing ...

引用

4th International Conference on Intelligent Autonomous Systems (ICoIAS)

作者： Zhang, Chengguang Zhang, Xuebo Jiang, Min Space Engn Univ Sch Grad Beijing Peoples R China Space Engn Univ Sch Space Informat Beijing Peoples R China

ISBN: (纸本)9781665441957

In recent years, methods based on deep learning have become a hot spot and trend in the field of remote sensing object detection, and a series of encouraging results have been achieved. With the development of science and technology, people's technology and ability to acquire remote sensing data have been comprehensively improved, and high-resolution large-scale remote sensing image data has increased dramatically. However, the current mainstream object detection models cannot directly input large-scale high-resolution images for prediction. This paper proposes a remote sensing object parallel detection algorithm, which uses the mpi4py module to realize multi-CPU+GPU distributed parallel processing. Based on the yolov4 object detection model, it can detect remote sensing images of any scale without reducing the prediction accuracy. And shorten the object detection time according to the number of distributed nodes. The experimental results show that the parallel algorithm has a high speedup ratio, and the parallel detection technology has a good development prospect in the field of remote sensing image object detection.

关键词： remote sensing object detention deep learning parallel processing

来源：评论

学校读者我要写书评

暂无评论

Efficiently Classify Synthesized Facial images Generated By Different Synthesis methods 12

Efficiently Classify Synthesized Facial Images Generated By ...

引用

12th International Conference on Graphics and image processing (ICGIP)

作者： Li, Wei Qiao, Peng Dou, Yong Natl Univ Def Technol Sci & Technol Parallel & Distributed Lab Changsha Peoples R China

ISBN: (纸本)9781510642782;9781510642775

With image generation and manipulation as one of impressive progress of convolutional neural networks (CNNs), facial image synthesis methods, e.g., DeepFakes, pose serious challenges to social and personal security. Specifically, we find that (1) CNN-based synthesized facial image detection methods generally fail to identify synthesized images generated by other synthesis methods;(2) classical detection methods exploiting one-class support vector machines (SVMs) and traditional features of video clips fail when only one image is available. In view of the above challenges, we propose and experimentally verify a method combining CNNs features and one-class SVMs, which not only effectively detects synthesized facial images generated by different methods, but also has good robustness to the variances of the scene content.

关键词： synthesized facial image detection convolutional neural networks one-class SVMs robustness

来源：评论

学校读者我要写书评

暂无评论

Exploration on Facial image Recognition and processing Algorithms under Hadoop

Exploration on Facial Image Recognition and Processing Algor...

引用

Integrated Intelligence and Communication Systems (ICIICS), International Conference on

作者： Meifang Gu Hailin Tang Lili Li Institutional Research Center Guangdong Baiyun University Guangzhou Guangdong China Faculty of Intelligent Manufacturing Engineering Guangdong Baiyun University Guangzhou Guangdong China Faculty of Megadata and Computing Guangdong Baiyun University Guangzhou Guangdong China

Face recognition is an important technology in the field of digital image processing, which has a crucial impact on subsequent work. This article introduced an example of facial recognition based on the combination of Hadoop algorithm and facial feature extraction. The experimental results were validated in the MATLAB environment and trained using neural networks to improve system performance and adaptability. The test results indicated that the processing speed was between 0.84 and 0.97; the highest resource utilization rate was 0.06, and the lowest was 0.02; the throughput range of facial data was between 13598 and 15479, and the throughput range of user data was between 5037 and 5879. This indicated that the system had fast recognition speed and high-precision image output, and exhibited good fault tolerance and robustness compared to other methods. Exploring facial image recognition and processing algorithms in the Hadoop environment can utilize Hadoop’s distributed computing capabilities to accelerate the processing of large-scale facial image data, achieve fast and efficient processing of large-scale facial image data, accelerate algorithm operation, and possess distributed computing capabilities. Multiple computing nodes can be used for parallel processing to improve algorithm performance.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Accelerating Hyperspectral Anomaly Detection With Enhanced Multivariate Gaussianization Based on FPGA

引用

IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING 2024年 62卷 1页

作者： Yu, Kun Wu, Zebin Sun, Jin Zhang, Yi Xu, Yang Wei, Zhihui Zheng, Shangdong Nanjing Univ Sci & Technol Sch Comp Sci & Engn Nanjing 210094 Peoples R China Nanjing Univ Posts & Telecommun Sch Comp Sci Nanjing 210023 Peoples R China

Hyperspectral anomaly detection (AD), as a frontier research topic in the field of remotely sensed data processing, aims to identify targets of interest from complex and vast images. Existing AD methods typically involve complex models and many parameters, posing challenges in meeting the requirements of computational efficiency in hyperspectral AD. To address this issue, this article presents an AD acceleration algorithm based on the multivariate Gaussian model as well as its field programmable gate array (FPGA) implementation. By exploiting the parallel processing capabilities of FPGA, we introduce an innovative spectral dimensionality reduction method in which the data processing flow can be accomplished in a distributed manner. Then, we employ an improved linear rotation strategy based on correlation coefficients to accelerate the convergence rate of the proposed AD algorithm. The rotation of Gaussianization in the improved strategy is independent of eigenvalue decomposition, thereby substantially reducing the computational complexity involved during the rotation procedure. Furthermore, we apply a pipeline parallel mechanism to facilitate the FPGA implementation of the AD algorithm and to significantly enhance the computational efficiency. Experimental results on an embedded FPGA platform demonstrate that the FPGA implementation of the hyperspectral AD algorithm proposed in this article achieves a significant acceleration rate with guaranteed high detection accuracy.

关键词： Field programmable gate arrays Hyperspectral imaging Computational modeling Detectors Convergence Computational efficiency Computational complexity Classification algorithms Accuracy Reliability Anomaly detection (AD) field programmable gate array (FPGA) hyperspectral image (HSI) linear rotation

来源：评论

学校读者我要写书评

暂无评论

RGBT Tracking via Multi-Adapter Network with Hierarchical Divergence Loss

引用

IEEE TRANSACTIONS ON image processing 2021年 30卷 5613-5625页

作者： Lu, Andong Li, Chenglong Yan, Yuqing Tang, Jin Luo, Bin Anhui Univ Sch Artificial Intelligence Informat Mat & Intelligent Sensing Lab Anhui Prov Anhui Prov Key Lab Multimodal Cognit Computat Hefei 230601 Peoples R China

RGBT tracking has attracted increasing attention since RGB and thermal infrared data have strong complementary advantages, which could make trackers all-day and all-weather work. Existing works usually focus on extracting modality-shared or modality-specific information, but the potentials of these two cues are not well explored and exploited in RGBT tracking. In this paper, we propose a novel multi-adapter network to jointly perform modality-shared, modality-specific and instance-aware target representation learning for RGBT tracking. To this end, we design three kinds of adapters within an end-to-end deep learning framework. In specific, we use the modified VGG-M as the generality adapter to extract the modality-shared target representations. To extract the modality-specific features while reducing the computational complexity, we design a modality adapter, which adds a small block to the generality adapter in each layer and each modality in a parallel manner. Such a design could learn multilevel modality-specific representations with a modest number of parameters as the vast majority of parameters are shared with the generality adapter. We also design instance adapter to capture the appearance properties and temporal variations of a certain target. Moreover, to enhance the shared and specific features, we employ the loss of multiple kernel maximum mean discrepancy to measure the distribution divergence of different modal features and integrate it into each layer for more robust representation learning. Extensive experiments on two RGBT tracking benchmark datasets demonstrate the outstanding performance of the proposed tracker against the state-of-the-art methods.

关键词： Feature extraction Target tracking Data mining Mobile computing Adaptation models Ad hoc networks Kernel RGBT tracking multiple adapters parallel design hierarchical divergence loss representation learning

来源：评论

学校读者我要写书评

暂无评论

parallel vs distributed Edge Detection for Large Medical image Datasets

Parallel vs Distributed Edge Detection for Large Medical Ima...

引用

E-Health and Bioengineering Conference (EHB)

作者： Renato Constantin ivanescu Rodica Nagy Anca Istrate Ofiteru Cristina Comanescu Alexandru Manda Department of Computers and Information Technology University of Craiova Craiova Romania University of Medicine and Pharmacy of Craiova University of Craiova Craiova Romania

ISBN: (数字)9781665485579

ISBN: (纸本)9781665485586

The medical routine of the future is strongly influenced by medical information technology. The quality and efficiency of medicine are at higher standards due to the image-based methods and the increase in computation power. parallel and distributed computing are attractive alternatives to improve processing performance in many areas of study and this paper investigates such solutions for increasing the throughput on large datasets in medical imaging. In this research, different edge detection techniques are used for image processing, as it is the most common and challenging task in the medical domain. The results are relevant and present the effectiveness of the two processing approaches compared to a single CPU for the acceleration of medical image processing. The CUDA processing platform using a hybrid streams-based programming model is the parallel approach. The distributed system is a Linux-based cluster, which uses message passing for communication and efficient load balancing between nodes for high performance. This paper highlights parallel computing as a very good solution for processing large image datasets in the medical domain. Good results were also obtained for the distributed system.

关键词： image processing image edge detection Graphics processing units Streaming media Programming Load management Task analysis

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：