检索结果-内蒙古大学图书馆

2019 ieee symposium Series on computational intelligence, SSCI 2019

作者： Tang, Hao Yang Wei Duan, Yi Sun, Lin Li Ran Li, Yi Xi'an University of Posts and Telecommunications Xi'an China

ISBN: (纸本)9781728124858

As a high efficiency video coding (HEVC) extension, screen content coding (SCC) has a significant effect in the compressed screen content, but causes a problem encoder calculates a higher complexity. This paper proposes a fast decision method based on HEVC screen content coding. The image content is divided into a natural coding unit (NCU) and a screen coding unit (SCU). The image gradients of different properties are different, and the gini impurity classification gradient value is used to speed up the encoding speed. The experimental results show that under the frame configuration of HM-15.0 SCM-2.0, the method can save about 41.4% of the encoder time. © 2019 ieee.

关键词： Video signal processing

来源：评论

学校读者我要写书评

暂无评论

A Multiband Structure based on Hammerstein Model for Nonlinear Audio System Identification 11

A Multiband Structure based on Hammerstein Model for Nonline...

引用

11th International symposium on image and signal processing and Analysis (ISPA)

作者： Terenzi, A. Bruschi, V. Cornell, S. Castellani, A. Cecchi, S. Univ Politecn Marche Dept Informat Engn Ancona Italy

ISBN: (纸本)9781728131405

In recent years, the increasing computational power has enabled the large-scale use of techniques for modeling real world systems and their non-linear characteristics. Hammerstein model is a well known structure for non-linear identification of memoryless systems. The proposed work aims at showing two innovative approaches based on multiband structure adaptive filters for Hammerstein adaptive identification process. In particular two multiband structures have been developed using the discrete cosine transform and the wavelet transform for the filterbank construction. The proposed approaches have been compared with a fullband normalized least mean square (NLMS) Hammerstein model, taking into consideration synthetic and real world audio devices.

关键词： Discrete wavelet transforms Discrete cosine transforms Distortion Mathematical model computational modeling

来源：评论

学校读者我要写书评

暂无评论

Robust Black-box Watermarking for Deep Neural Network using Inverse Document Frequency

Robust Black-box Watermarking for Deep Neural Network using ...

引用

ieee International symposium on Dependable, Autonomic and Secure Computing (DASC)

作者： Mohammad Mehdi Yadollahi Farzaneh Shoeleh Sajjad Dadkhah Ali A. Ghorbani Faculty of Computer Science Canadian Institute for Cybersecurity (CIC) UNB Fredericton Canada

ISBN: (纸本)9781665421751

Recently, Deep Neural Networks (DNNs), presented exceptional achievement in implementing human-level capabilities for various predicaments, such as Natural Language processing (NLP), voice recognition, and image processing, etc. Training these models is expensive in terms of computational power and the existence of enough labelled data. Thus, ML-based models such as DNNs establish genuine business value and intellectual property (IP) for their owners. Therefore the trained models need to be protected from any adversary attacks such as illegal redistribution, reproducing, and derivation. Watermarking can be considered as an effective technique for securing a DNN model. However, so far, most of the watermarking algorithms focus on watermarking the DNN by adding noise to an image. To this end, we propose a framework for watermarking a DNN model designed for textual domain. The watermark generation scheme provides a secure watermarking method by combining Term Frequency (TF) and Inverse Document Frequency (IDF) of a particular word. The proposed embedding procedure takes place in the model's training stage, which makes the watermark verification stage straightforward by sending the watermarked document to the trained model. The experimental results show that watermarked models have the same accuracy as the original one, and the proposed framework accurately verifies the ownership of all surrogate models without impairing the performance. The proposed algorithm is robust against well-known attacks such as parameter pruning and brute force attack.

关键词： Training Deep learning image recognition computational modeling Neural networks Force Watermarking

来源：评论

学校读者我要写书评

暂无评论

Characterization of the Subsoil using computational Techniques for Seismic Microzonation Studies 22

Characterization of the Subsoil using Computational Techniqu...

引用

22nd symposium on image, signal processing and Artificial Vision (STSIVA)

作者： Atehortua-Sanchez, Nelcy N. Echeverry-Correa, Julian D. Lopez Varona, Ricardo Univ Tecnol Pereira Grp Invest GIADSc Pereira Colombia Univ Tecnol Pereira Grp Invest Electrofisiol Pereira Colombia

ISBN: (纸本)9781728114910

The role of superficial geology of the subsoil in studies of seismic microzonation is of great interest because it is a latent threat before an earthquake. With non-invasive techniques, the behavior of seismic waves can be inferred by studying the ways they propagate. The Nakamura technique (H/V) is the most used in order to identify the natural frequency of the site in the seismic microzonation studies. This technique allows to know the surface geology through experimental in-site data and implementing computational techniques. In this paper, records of environmental seismic noise were used for the extraction of characteristics of each of the signal components. The H/V curves allowed to identify the peaks in the frequency range from 0,1 to 10 Hz. A clustering technique has been used on the feature vector to identify different subsets related to similar seismic behavior for the characterization of the subsoil.

关键词： signal characterization environmental seismic noise cluster analysis seismic microzonation

来源：评论

学校读者我要写书评

暂无评论

Efficient Integer-Arithmetic-Only Convolutional Networks with Bounded ReLU

Efficient Integer-Arithmetic-Only Convolutional Networks wit...

引用

ieee International symposium on Circuits and Systems (ISCAS)

作者： Hengrui Zhao Dong Liu Houqiang Li CAS Key Laboratory of Technology in Geo-Spatial Information Processing and Application System University of Science and Technology of China Hefei China

To facilitate large-scale deployment of convolutional networks, integer-arithmetic-only inference has been demonstrated effective, which not only reduces computational cost but also ensures cross-platform consistency. However, previous studies on integer networks usually report a decline in the inference accuracy, given the same number of parameters as floating-point-number (FPN) networks. In this paper, we propose to finetune and quantize a well-trained FPN convolutional network to obtain an integer convolutional network. Our key idea is to adjust the upper bound of a bounded rectified linear unit (ReLU), which replaces the normal ReLU and effectively controls the dynamic range of activations. Based on the tradeoff between learning ability and quantization error of networks, we managed to preserve full accuracy after quantization and obtain efficient integer networks. Our experiments on ResNet for image classification demonstrate that our 8-bit integer networks achieve state-of-the-art performance compared with Google's TensorFlow and NVIDIA's TensorRT. Moreover, we experiment on VDSR for image super-resolution and on VRCNN for compression artifact reduction, both of which serve regression tasks that natively require high inference accuracy. Besides ensuring the equivalent performance as the corresponding FPN networks, our integer networks have only 1/4 memory cost and run 2× faster on GPUs.

关键词： Upper bound Quantization (signal) image coding Superresolution Dynamic range Task analysis image classification

来源：评论

学校读者我要写书评

暂无评论

A Survey on the Identification of Causal Relation in Texts

A Survey on the Identification of Causal Relation in Texts

引用

symposium of image, signal processing, and Artificial Vision (STSIVA)

作者： Mingyue Han Yinglin Wang School of Information Management and Engineering Shanghai University of Finance and Economics Shanghai China

ISBN: (纸本)9781665426565

Causality is the basis for humans to make rational decisions and is widely mentioned in different fields. In the natural language processing (NLP) community, the problem of causality is complex and challenging. This paper serves as an effort to briefly discuss the causal relation identification in texts, from the existing causal resources, research methodology, and the robustness problems. First, we introduce relevant causal datasets and resources. Second, the existing typical approaches that have been used in causal relation identification are categorized into unsupervised and supervised methods. In addition, the robustness of causality identification models is discussed succinctly. Finally, we try to list the research challenges at present and raise the future research directions in this field.

关键词： Conferences computational modeling Robustness Natural language processing Informatics

来源：评论

学校读者我要写书评

暂无评论

Person Footprint of Uncertainty-Based CWW Model for Power Optimization in Handheld Devices

引用

ieee TRANSACTIONS ON FUZZY SYSTEMS 2020年第3期28卷 558-568页

作者： Muhuri, Pranab K. Gupta, Prashant K. Mendel, Jerry M. South Asian Univ Dept Comp Sci New Delhi 110021 India Univ Southern Calif Signal & Image Proc Inst Los Angeles CA 90089 USA Tianjin Normal Univ Coll Artificial Intelligence Tianjin 300384 Peoples R China

Present-day handheld battery-enabled devices such as smartphones and tablets attract rich user experience but are often criticized for their short battery lives. Battery life is a subjective term and depends on a user's perceptions. A novel work to achieve power optimization for these devices, according to users' perceptions, was the design of user-satisfaction-aware power management approach, perceptual computer power management approach (Per-C PMA). But we have found that the design of Per-C PMA requires collection of data intervals from a group of subjects. This limits the practical viability of Per-C PMA for highly personal handheld battery-enabled devices such as smartphones and tablets. So, here we propose a user-satisfaction-aware PMA called Per-C for Personalized Power Management Approach or "Per-C PPMA," one that achieves significant reductions in power consumption compared to existing PMAs and noticeable improvements in the overall user satisfaction. Per-C PPMA uses the mathematical technique of person footprint of uncertainty (FOU) to process users' linguistic opinions. Person FOU can either use an interval approach (IA) or Hao-Mendel approach (HMA) for data processing. The recommendations generated using IA and HMA are the same. However, IA takes a much higher computational time than HMA, even though both have the same asymptotic complexity of $\boldsymbol{O}({\boldsymbol{w}*\boldsymbol{n}})$. We strongly believe that Per-C PPMA is a novel technique and our work is the first such application of Person FOU on any hardware platform. An important outcome of this study is a ready-to-use mobile app "Per-C PPMA" (currently freely available on the website http://***/similar to cilab/).

关键词： Computing with words (CWWs) interval approach (IA) interval type-2 fuzzy sets per-C power management approach (PMA) person footprint of uncertainty (FOU) power management Hao-Mendel approach (HMA)

来源：评论

学校读者我要写书评

暂无评论

Parking system with image processing 17

Parking system with image processing

引用

17th ieee World symposium on Applied Machine intelligence and Informatics (SAMI)

作者： Kanalikova, Alzbeta Bubenikova, Emilia Univ Zilina Dept Control & Informat Syst Zilina Slovakia

ISBN: (纸本)9781728102504

image processing is a method to perform some operations on an image, to either enhance the image or to extract some useful information from it. It is a type of signal processing with an image as input and output in the form of an image or characteristics/features associated with that image. Nowadays, image processing, and digital image processing at that is among rapidly growing technologies. The article deals with image processing, specifically focuses on recognizing objects - cars in 2D image frames through IP cameras. The practical part deals with the implementation of algorithms from the Open CV library designed for vehicle recognition on a large parking lot in order to find the actual occupancy of the parking lot. The occupancy results of the parking rate converted to percentages are displayed on the web portal.

关键词： image processing library Open CV Parking system the method of recognition objects

来源：评论

学校读者我要写书评

暂无评论

A MODIFIED STRUCTURAL SIMILARITY INDEX WITH LOW computational COMPLEXITY

A MODIFIED STRUCTURAL SIMILARITY INDEX WITH LOW COMPUTATIONA...

引用

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

作者： Loh, Woei-Tan Bong, David Boon Liang Univ Malaysia Sarawak Fac Engn Kota Samrahan Malaysia

ISBN: (纸本)9781728130385

Structural Similarity Index (SSIM) has been a benchmark method for image quality assessment (IQA). This is due to its simplicity and good performance. In this paper, we propose a modified SSIM method that reduces the computational complexity with comparable performance. Instead of computing similarities on local windows, the proposed method computes global information similarities. The proposed method omits the luminance part similarities in SSIM due to its less crucial role in assessing image quality. From the presented results, the proposed method has a much lower computational time and comparable performance compared to SSIM.

关键词： image quality computational complexity SSIM

来源：评论

学校读者我要写书评

暂无评论

GPU Usage trends in Medical image processing 13

GPU Usage trends in Medical Image processing

引用

ieee 13th International symposium on Applied computational intelligence and Informatics (SACI)

作者： Benhamida, Abdallah Kozlovszky, Miklos Szenasi, Sandor Obuda Univ BioTech Res Ctr EKIK Budapest Hungary Obuda Univ John von Neumann Fac Informat Budapest Hungary

ISBN: (纸本)9781728106854

Graphics processing units (GPUs), mainly named as general-purpose computing on graphics processing units (GPGPUs) became a widely used technology for many computer-aided procedures in different research or industry areas that require high algorithmic computation capabilities for parallel computing tasks. In the field of bioinformatics and especially in medical image processing, GPUs play an essential role to enable high performance and computation capabilities in order to facilitate the tasks related to image processing. This paper highlights the usage of the graphics processing units in medical image processing and three-dimensional medical image processing related applications (computer-aided image processing, volume rendering, and segmentation), such as image segmentation, noise reduction or denoising, and image registration. It gives a deep understanding of the influence of GPU cards on general purpose programming GPGPU in general and on medical image processing specifically. It presents a good starting point for further applications of GPU programming in the field of biomedical applications and mainly for medical image processing that requires high-performance capabilities for medical image treatment shortly after the medical imaging activity such as in the ICU or the emergency services.

关键词： GPU GPGPU medical image processing biomedical

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：