检索结果-内蒙古大学图书馆

2015 ieee International Conference on computational intelligence and Research (ICCIC)

作者： Vanithamani, R. Dhivya, R. Sharmili, S. Avinashilingam Univ Women Dept Biomed Instrumentat Engn Coimbatore Tamil Nadu India

ISBN: (纸本)9781479978496

Ultrasound imaging has features like non-invasive nature, real time image formation capacity and relatively low cost, which makes this diagnostic tool attractive and hence has become an important imaging modality in medical diagnoses. However the usefulness of this imaging is degraded by the presence of speckle noise. Hence, speckle suppression in ultrasound images is essential for improving the image quality. In this paper, a novel hybrid method is proposed for enhancing the visual quality of medical ultrasound images. To preserve edges, a nonlinear filter is used in the pre-processing stage. Further the pre-processed image is denoised in wavelet domain using bivariate shrinkage. The quality of the despeckled images is assessed using objective quality assessment metrics such as Peak signal to Noise Ratio (PSNR), Root Mean Squared Error (RMSE) and Edge Preservation Index (EPI). The quantitative results indicate that the proposed method yields better performance than the existing methods.

关键词： Speckle Noise Bivariate Shrinkage Despeckled image

来源：评论

学校读者我要写书评

暂无评论

Superresolution image Reconstruction: Selective milestones and open problems

引用

ieee signal processing MAGAZINE 2023年第5期40卷 54-66页

作者： Li, Xin Dong, Weisheng Wu, Jinjian Li, Leida Shi, Guangming West Virginia Univ Lane Dept Comp Sci & Elect Engn Morgantown WV 26506 USA Xidian Univ Sch Elect Engn Xian 710071 Peoples R China Xidian Univ Sch Artificial Intelligence Xian 710071 Peoples R China

In multidimensional signal processing, such as image and video processing, superresolution (SR) imaging is a classical problem. Over the past 25 years, academia and industry have been interested in reconstructing high-resolution (HR) images from their low-resolution (LR) counterparts. We review the development of SR technology in this tutorial article based on the evolution of key insights associated with the prior knowledge or regularization method from analytical representations to data-driven deep models. The coevolution of SR with other technical fields, such as autoregressive modeling, sparse coding, and deep learning, will be highlighted in both model-based and learning-based approaches. Model-based SR includes geometry-driven, sparsity-based, and gradient-profile priors;learning-based SR covers three types of neural network (NN) architectures, namely residual networks (ResNet), generative adversarial networks (GANs), and pretrained models (PTMs). Both model-based and learning-based SR are united by highlighting their limitations from the perspective of model-data mismatch. Our new perspective allows us to maintain a healthy skepticism about current practice and advocate for a hybrid approach that combines the strengths of model-based and learning-based SR. We will also discuss several open challenges, including arbitrary-ratio, reference-based, and domain-specific SR.

关键词： Deep learning Analytical models Visualization Uncertainty computational modeling Superresolution Imaging

来源：评论

学校读者我要写书评

暂无评论

Fast Lip Feature Extraction Using Psychologically Motivated Gabor Features 8

Fast Lip Feature Extraction Using Psychologically Motivated ...

引用

8th ieee symposium Series on computational intelligence (ieee SSCI)

作者： Abel, Andrew Gao, Chengxiang Smith, Leslie Watt, Roger Hussain, Amir Xian Jiaotong Liverpool Univ Comp Sci & Software Engn Suzhou Peoples R China Univ Stirling Fac Nat Sci Stirling Scotland

ISBN: (纸本)9781538692769

The extraction of relevant lip features is of continuing interest in the speech domain. Using end-to-end feature extraction can produce good results, but at the cast of the results being difficult for humans to comprehend and relate to. We present a new, lightweight feature extraction approach, motivated by glimpse based psychological research into facial barcodes. This allows for 3D geometric features to be produced using Gabor based image patches. This new approach can successfully extract lip features with a minimum of processing, with parameters that can be quickly adapted and used for detailed analysis, and with preliminary results showing successful feature extraction from a range of different speakers. These features can be generated online without the need for trained models, and are also robust and can recover from errors, making them suitable for real world speech analysis.

关键词： image processing barcodes gabor lip-reading word features

来源：评论

学校读者我要写书评

暂无评论

DFT and DCT based discrete sinc-interpolation methods for direct Fourier tomographic reconstruction 3

DFT and DCT based discrete sinc-interpolation methods for di...

引用

3rd International symposium on image and signal processing and Analysis (ISPA)

作者： Yaroslavsky, LP Chernobrodov, Y Tel Aviv Univ Fac Engn Leonid P Yaroslavsky Dept Interdisciplinary Studi IL-69978 Tel Aviv Israel

ISBN: (纸本)953184061X

The paper describes the implementation and use of recently developed DFT and DCT based discrete Sinc-interpolation algorithms for the Direct Fourier Method of reconstructing images from projections. DFT based discrete sinc-interpolation is the only completely reversible discrete interpolation technique. DCT based discrete sinc-interpolation algorithm implements the same interpolation kernel that that of the DFT based method and allows to avoid its boundary affect artifacts. It is also computationally more efficient. Two modifications of the DCT based interpolation for tomographic reconstruction are suggested: interpolation by global zooming of 1-D DFT spectra Of projections and interpolation with variable zooming factor along the angle coordinate. The latter allows to substantially reduce the computational complexity without compromising interpolation accuracy. Results of comparative numerical simulation of suggested algorithms show a good image reconstruction quality with a reduced level of artifacts.

关键词： computational complexity Discrete cosine transforms Discrete Fourier transforms image converters image reconstruction Interpolation Kernel Numerical simulation Reconstruction algorithms Tomography

来源：评论

学校读者我要写书评

暂无评论

MAT2DSP - A MATLAB tool for rapid feedback on the implementation requirements of signal processing algorithms

MAT2DSP - A MATLAB tool for rapid feedback on the implementa...

引用

40th Midwest symposium on Circuits and Systems

作者： Bose, S Friedlander, B Zeira, A Univ Calif Davis Dept Elect & Comp Engn Davis CA 95616 USA

ISBN: (纸本)0780336941

We introduce and describe MAT2DSP, a MATLAB toolbox, whose function is to estimate the computational load of algorithms specified in the form of a MATLAB program (or programs). This toolbox is aimed at providing researches developing advanced signal and image processing algorithms, a quick and convenient way of estimating the implementation requirements of their algorithm on a variety of processors. MAT2DSP analyzes the user program and generates reports on its computational requirements.

关键词： Digital signal processing

来源：评论

学校读者我要写书评

暂无评论

A METHOD TO ENHANCE THE DEEP LEARNING IN AN AERIAL image 25

A METHOD TO ENHANCE THE DEEP LEARNING IN AN AERIAL IMAGE

引用

International symposium on Intelligent signal processing and Communication Systems (ISPACS)

作者： Chou, Kuang-Pen Li, Dong-Lin Lin, Wen-Chieh Prasad, Mukesh Lin, Chin-Teng Natl Chiao Tung Univ Inst Comp Sci & Engn Hsinchu Taiwan Univ Technol Sydney FEIT Sch Software Ctr Artificial Intelligence Sydney NSW Australia Natl Chiao Tung Univ Inst Elect & Control Engn Hsinchu Taiwan Natl Chiao Tung Univ Dept Comp Sci Hsinchu Taiwan

In this paper, we propose a kind of pre-processing method which can be applied to the depth learning method for the characteristics of aerial image. This method combines the color and spatial information to do the qui... 详细信息

ISBN: (纸本)9781538621592

关键词： Drones Aerial image Deep learning Background filter

来源：评论

学校读者我要写书评

暂无评论

Study of unsaturated magnetic image processing in wire rope aplication

Study of unsaturated magnetic image processing in wire rope ...

引用

作者： Tan, Xiaojiang Zhang, Juwei Henan University of Science and Technology College of Electrical Engineering Luoyang China

ISBN: (纸本)9781538636749

Traditional magnetic flux leakage signal need be processed through each channels in magnetic array data, that increased processing system complexity. This paper proposed a processing system for unsaturated magnetic image, the processes included image transformation, filtering and resolution enhancement. This system overcame the shortcomings of interpolation method and digital signal processing. The experimental results showed that every defects could effectively be extracted and a good reconstruction image of multi-frame super-resolution was obtained with 3 simulated sampling intervals. © 2017 ieee.

关键词： image processing Magnetic imageing Super-resolution Wire rope

来源：评论

学校读者我要写书评

暂无评论

Parallel image Segmentation using Multi-Threading and K-Means Algorithm

Parallel Image Segmentation using Multi-Threading and K-Mean...

引用

ieee International Conference on computational intelligence and Computing Research (ICCIC)

作者： Bose, Soumyo Mukherjee, Aniruddha Madhulika Chakraborty, Sayan Samanta, Sourav Dey, Nilanjan JIS Coll Engn Dept CSE Kalyani W Bengal India Univ Inst Technol CSE Dept Burdwan W Bengal India JIS Coll Engn Dept IT Kalyani W Bengal India

ISBN: (纸本)9781479915941;9781479915958

In the present era of mechanization, image segmentation has brought revolutionary changes in the field of image processing. There are many image segmentation methods in use but most of the prevalent techniques are not very efficient. In our current work, a method is proposed to segment images using multi-threaded programming and k-means clustering. Parts of images are used as a thread and k-means clustering is used to segment them on every thread. An efficient image segmentation technique using the multi-threaded program is done in this paper. These results are compared with segmentation results without multi-threading. The obtained results are very much encouraging.

关键词： image segmentation Clustering k-Means Multi-threaded programming

来源：评论

学校读者我要写书评

暂无评论

Regularizing Visual Semantic Embedding With Contrastive Learning for image-Text Matching

引用

ieee signal processing LETTERS 2022年 29卷 1332-1336页

作者： Liu, Yang Liu, Hong Wang, Huaqiu Liu, Mengyuan Chongqing Univ Technol Sch Artificial Intelligence Chongqing 401135 Peoples R China Peking Univ Key Lab Machine Percept Beijing 100871 Peoples R China SunYat Sen Univ Sch Intelligent Syst Engn Guangzhou 510275 Peoples R China

Learning visual semantic embedding for image-text matching has achieved high success by using triplet loss to pull positive image-text pairs which share similar semantic meaning and to push negative image-text pairs which share different semantic meaning. Without modeling constraints from image-image or text-text pairs, the generated visual semantic embedding inevitably faces the problem of semantic misalignments among similar images or among similar texts. To solve this problem, we present a contrastive visual semantic embedding framework, named ConVSE, which achieves intra-modal semantic alignment by contrastive learning from augmented image-image (or text-text) pairs and achieves inter-modal semantic alignment by applying hardest-negative-enhanced triplet loss on image-text pairs. To the best of our knowledge, we are the first to find that contrastive learning benefits visual semantic embedding. Extensive experiments on large-scale MSCOCO and Flickr30 K datasets verify the effectiveness of our proposed ConVSE by outperforming visual semantic embedding-based methods and achieving new state-of-the-art.

关键词： Semantics Visualization Integrated circuit modeling Feature extraction image retrieval Task analysis computational modeling Cross-modal retrieval image-text matching visual semantic embedding

来源：评论

学校读者我要写书评

暂无评论

Blind image Quality Assessment For Color images With Additive Gaussian White Noise Using Standard Deviation

Blind Image Quality Assessment For Color Images With Additiv...

引用

International symposium on Intelligent signal processing and Communication Systems (ISPACS)

作者： Lim, Chern-Loon Paramesran, Raveendran Univ Malaya Dept Elect Engn Fac Engn Kuala Lumpur 50603 Malaysia

ISBN: (纸本)9781479961207

This paper presents a no-reference (NR) quality assessment method for color images contaminated with additive Gaussian white noise (AGWN). The proposed metric operates on the test image and the scaled down version of the test image. Standard deviations for each of the RGB components for the test image and its scaled down version are computed. The standard deviation ratios of the scaled down version image to the original test image are weighted summed to obtain the quality score for noise. The performance of the proposed metric is compared with existing full-reference (FR) and NR metrics on LIVE database. Pearson correlation coefficient (CC), mean absolute error (MAE) and root mean square error (RMSE) are used to evaluate the performance of the proposed metric. A CC score of 0.9785 shows that the proposed metric has high correlation with the subjective scores for the LIVE database.

关键词： Mean square error

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：