检索结果-内蒙古大学图书馆

2020 3rd international conference on Sensors, signal and image processing

作者： Magdalena Garvanova Vladimir Ivanov University of Library Studies and Information Technologies Bulgaria Institute of Information and Communication Technologies-BAS Bulgaria

ISBN: (纸本)9781450388283

Camera shake during video recording is a major source of image defocus. Restoring these defocuses is one of the most important tasks in image processing. This article examines and evaluates the most commonly used algorithms in practice, namely: Wiener filtration, Richardson-Lucy algorithm, Tikhonov regularization, and Blind deconvolution. The comparisons were made using structural similarity and root mean square error.

关键词： camera shake image processing defocuses

来源：评论

学校读者我要写书评

暂无评论

Self-Adaptive Visual Background Extraction with Ghost Regions Elimination

Self-Adaptive Visual Background Extraction with Ghost Region...

引用

IEEE international conference on signal and image processing (ICSIP)

作者： Ling Liu Shuai Liu Zhong Qu Dongyang Zhou School of Big Data and Software Engineering College of Mobile Telecommunications of Chongqing University of Posts and Telecommunications Chongqing China College of Computer Science and Technology Chongqing University of Posts and Telecommunications Chongqing China College of Mobile Telecommunications of Chongqing University of Posts and Telecommunications Chongqing China

ISBN: (纸本)9781665446006

The ViBe (Visual Background Extraction) algorithm uses the first frame to initialize the pixel background model for each pixel, which may introduce the ghost regions in the following video frames. Although ViBe can use its background model update mechanism to eliminate the ghost regions, it is time-consuming. Firstly, a ghost elimination algorithm is proposed in this paper in order to reduce the elimination time for the ghost regions based on its properties. Secondly, a fixed global segmentation decision threshold is used to segment pixels in ViBe algorithm. However, because it cannot adapt to the dynamic background, a self-adaptive segmentation decision threshold based on single pixel is used to adjust to the different backgrounds. Finally, a self-adaptive update factor is presented, which makes the background model more reliable and adjusts to the background changes through the self-adaptive processing. The experimental result shows that this improved method can quickly eliminate the ghost regions, and adapt to the different kinds of backgrounds, ensuring more accurate results.

关键词： Visualization Adaptation models Heuristic algorithms image processing conferences Lighting Reliability

来源：评论

学校读者我要写书评

暂无评论

A Graph Learning Algorithm Based On Gaussian Markov Random Fields And Minimax Concave Penalty

A Graph Learning Algorithm Based On Gaussian Markov Random F...

引用

IEEE international conference on Acoustics, Speech and signal processing

作者： Tatsuya Koyakumaru Masahiro Yukawa Eduardo Pavez Antonio Ortega Keio University Japan University of Southern California Los Angeles California USA

This paper presents a graph learning framework to produce sparse and accurate graphs from network data. While our formulation is inspired by the graphical lasso, a key difference is the use of a nonconvex alternative of the l_1 norm as well as a quadratic term to ensure overall convexity. Specifically, the weakly-convex minimax concave penalty (MCP) is used, which is given by subtracting the Huber function from the l_1 norm, inducing a less-biased sparse solution than l_1. In our framework, the graph Laplacian is represented by a linear transform of the vector corresponding to its upper triangular part. Via a reformulation relying on the Moreau decomposition, the problem can be solved by the primal-dual splitting method. An admissible choice of parameters for provable convergence is presented. Numerical examples show that the proposed method significantly outperforms its l_1-based counterpart for sparse grid graphs.

关键词： Learning systems Laplace equations image color analysis conferences signal processing algorithms Transforms signal processing

来源：评论

学校读者我要写书评

暂无评论

image compression and enlargement algorithms

引用

international Journal of Circuits, Systems and signal processing 2021年 15卷 836-846页

作者： Burova, I.G. Dem’yanovich, Yu.K. Terekhov, A.N. Altynova, A.Yu. Satanovskiy, A.D. Babushkin, A.A. St. Petersburg State University 7/9 Universitetskaya nab St.Petersburg199034 Russia

In some cases, there are problems associated with the compression and enlargement of images. The use of splines is quite effective in some cases. In this paper, a new image compression algorithm is presented. The features of increasing the size of an image when using local polynomial or non-polynomial splines are considered. The main method for enlarging an image is based on the use of splines of the second and third order of approximation. Polynomial and trigonometric splines are considered. To speed up the process of enlarging the image, we used the parallelization techniques. © 2021, North Atlantic University Union NAUN. All rights reserved.

关键词： image compression

来源：评论

学校读者我要写书评

暂无评论

Fast: Feature Aggregation for Detecting Salient Object in Real-Time

Fast: Feature Aggregation for Detecting Salient Object in Re...

引用

IEEE international conference on Acoustics, Speech and signal processing

作者： Lv Tang Bo Li Yanliang Wu Bo Xiao Shouhong Ding Nanjing University China Tencent Shanghai China School of Information Science and Technology Southwest Jiaotong University China

This paper introduces a method named FAST for real-time salient object detection with an extremely efficient CNN architecture. Our proposed network starts from a single lightweight backbone and aggregates discriminative features through network-level and phase-level respectively. Based on the multi-scale feature propagation, FAST substantially reduces the number of parameters, but still obtains sufficient receptive field and enhances the model learning ability, which strikes a balance between the speed and performance. To better preserve object boundaries, we also explore the complementary between salient object information and edge information within our lightweight architecture. Extensive evaluations and analysis demonstrate that the proposed algorithm achieves the leading accuracy performance with real-time speed (186fps) which is significantly faster than the existing state-of-the-art methods.

关键词： Convolution image edge detection conferences Aggregates signal processing algorithms Object detection Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Color Intensity: A Study of RPPG Algorithm for Heart Rate Estimation

Color Intensity: A Study of RPPG Algorithm for Heart Rate Es...

引用

Computational Performance Evaluation (ComPE), international conference on

作者： Ruchika Sinhal Kavita R. Singh K. O. Gupta DMIETR Wardha Maharashtra India YCCE Nagpur Maharashtra India SVPCET Nagpur Maharashtra India

ISBN: (纸本)9781665436571

With the growing advancements and development in the field of digital image processing and computer vision, an individual’s heart pulse can be extracted from the human skin surfaces. This method is termed as remote photoplethysmography (rPPG). The method can be applied from the video recorded from the consumer-based mobile camera also. In this paper, the work presented has mainly twofold goals. Firstly to develop a fruitful yet simple rPPG algorithm that should be simple for any individual to understand and implement that will increase the understanding of the rPPG subject. Secondly, to compare the algorithm designed for the RGB color model with the state-of-art rPPG algorithms developed and presented in the literature. And finally, we present the comparative analysis of rPPG algorithms reported in the literature with our proposed rPPG algorithm which is simple and has demonstrated comparably high performance for the green channel as compared to other algorithms.

关键词： Heart rate Performance evaluation image color analysis Tracking signal processing algorithms Lighting Photoplethysmography

来源：评论

学校读者我要写书评

暂无评论

Data Augmentation with signal Companding for Detection of Logical Access Attacks

Data Augmentation with Signal Companding for Detection of Lo...

引用

IEEE international conference on Acoustics, Speech and signal processing

作者： Rohan Kumar Das Jichen Yang Haizhou Li National University of Singapore Singapore

The recent advances in voice conversion (VC) and text-to-speech (TTS) make it possible to produce natural sounding speech that poses threat to automatic speaker verification (ASV) systems. To this end, research on spoofing countermeasures has gained attention to protect ASV systems from such attacks. While the advanced spoofing countermeasures are able to detect known nature of spoofing attacks, they are not that effective under unknown attacks. In this work, we propose a novel data augmentation technique using a-law and mu-law based signal companding. We believe that the proposed method has an edge over traditional data augmentation by adding small perturbation or quantization noise. The studies are conducted on ASVspoof 2019 logical access corpus using light convolutional neural network based system. We find that the proposed data augmentation technique based on signal companding outperforms the state-of-the-art spoofing countermeasures showing ability to handle unknown nature of attacks.

关键词： Voice activity detection Quantization (signal) Perturbation methods image edge detection conferences Acoustics Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Research on Salt and Pepper Noise Removal Method Based on Adaptive Fuzzy Median Filter

Research on Salt and Pepper Noise Removal Method Based on Ad...

引用

IEEE advanced Information Technology, Electronic and Automation Control conference (IAEAC)

作者： Ping Luo Xiaoxiao Zhang Zheng Chang Wang Liu College of Automation Chongqing University of Posts and Telecommunications Chongqing China

ISBN: (数字)9781728180281

ISBN: (纸本)9781728180298

Aiming at the problem that the traditional median filtering algorithm can not well retain the image details while eliminating image noise, this paper proposes an adaptive fuzzy median filtering algorithm based on the mean of absolute average gradient difference. The whole filtering process is divided into three parts: coarse noise detection, fine noise detection and noise filtering. First, the extreme value method is used to perform rough noise detection on the image to be tested, which is divided into suspected noise points and signal points; then use this algorithm to construct the fuzzy membership function of the mean of absolute average gradient difference between the detected suspected noise points and the signal points in the window (including the filtered noise points), and use this function to perform secondary detection on the suspected noise points to determine Whether it is noise and fuzzy classification processing; finally, perform different filtering processing on the noise fuzzy classification results. Experimental results show that the algorithm proposed in this paper has significant improvements in PSNR, MSE, and SSIM compared to the traditional median filtering algorithm. Therefore, the algorithm in this paper can effectively remove the salt and pepper noise of the image while protecting the image texture details.

关键词： image texture conferences Adaptive filters Filtering algorithms Information filters Classification algorithms Information technology

来源：评论

学校读者我要写书评

暂无评论

An Effective Model for Detecting Severeness of Disease in Paddy Leaf

An Effective Model for Detecting Severeness of Disease in Pa...

引用

Information Systems and advanced Technologies (ICISAT), international conference on

作者： Amogh Shukla Aryan Animesh Bhatt Nitin Lodha Vellore Institute of Technology. Vellore Tamil Nadu India

ISBN: (纸本)9781665478250

The capacity to recognize and classify diseases based on leaf images acquired from plants is necessary. Using image processing algorithms to detect diseases in rice leaves would reduce reliance on farmers to preserve agricultural goods. The research paper uses image processing to identify and diagnose sickness in paddy leaves. An innovative neural network-based technique for identifying and categorizing rice leaf disease is given in this paper. advanced mechanisms have employed an approach that incorporates photo compression, edge detection, and a neural network to perform sickness diagnosis rapidly and reliably. The proposed model accuracy is greater than accuracy found in previous studies. Furthermore, the recommended model, which employs a neural network and a edge detection algorithm, has greater precision and recall accuracy than the previous model for bacterial blight, brown spot, health, and rot. Finally, compared to the previous technique, the proposed model is more accurate, adaptive, and scalable. Furthermore, research has concentrated on simulating the disease’s intensity. There are three scenarios that are taken into account. One is when there is no sickness present. The second scenario is a situation in which the sickness has been identified but is not severe and is treatable. The third scenario depicts a circumstance in which the identified sickness is severe yet curable.

关键词： Adaptation models Microorganisms image coding image edge detection Neural networks Mathematical models Reliability

来源：评论

学校读者我要写书评

暂无评论

An Improved Epsilon Method with M2M for Solving Imbalanced CMOPs with Simultaneous Convergence-Hard and Diversity-Hard Constraints 11th

An Improved Epsilon Method with M2M for Solving Imbalanced C...

引用

11th international conference on Evolutionary Multi-Criterion Optimization, EMO 2021

作者： Fan, Zhun Yang, Zhi Tang, Yajuan Li, Wenji Xu, Biao Wang, Zhaojun Sun, Fuzan Long, Zhoubin Zhu, Guijie Department of Electronic and Information Engineering Shantou University Shantou Guangdong515063 China Key Lab of Digital Signal and Image Processing of Guangdong Province Shantou University Shantou Guangdong515063 China Ministry of Education Shantou Guangdong515063 China State Key Lab of Digital Manufacturing Equipment and Technology Huazhong University of Science and Technology Wuhan430000 China

ISBN: (纸本)9783030720612

When tackling imbalanced constrained multi-objective optimization problems (CMOPs) with simultaneous convergence-hard and diversity-hard constraints, a critical issue is to balance the diversity and convergence of populations. To address this issue, this paper proposes a hybrid algorithm which combines an improved epsilon constraint-handling method (IEpsilon) with a multi-objective to multi-objective (M2M) decomposition approach, namely M2M-IEpsilon. The M2M decomposition mechanism in M2M-IEpislon has the capability to deal with imbalanced objectives. The IEpsilon constraint-handling method can prevent populations falling into large infeasible regions, thus improves the convergence performance of the proposed algorithm. To verify the performance of the proposed M2M-IEpsilon, a series of imbalanced CMOPs with simultaneous convergence-hard and diversity-hard constraints, namely ICD-CMOPs, is designed by using the DAS-CMOPs framework. Six state-of-the-art constrained multi-objective evolutionary algorithms (CMOEAs), including CM2M, CM2M2, NSGA-II-CDP, MOEA/D-CDP, MOEA/D-IEpsilon and PPS-MOEA/D, are employed to compare with M2M-IEpsilon on the ICD-CMOPs. Through the analysis of experimental results, the proposed M2M-IEpsilon is superior to the other six algorithms in solving ICD-CMOPs, which illustrates the superiority of the proposed M2M-IEpsilon in dealing with ICD-CMOPs with simultaneous convergence-hard and diversity-hard constraints. © 2021, Springer Nature Switzerland AG.

关键词： Constraint handling

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：