检索结果-内蒙古大学图书馆

IAPR International conference on Biometrics (ICB)

作者： B H Shekar R K Bharathi Josef Kittler Yu. v. vizilter Leonid Mestestskiy Department of Computer Science Mangalore University Mangalore India Department of Master of Computer Applications S J College of Engineering Mysore India Centre for Vision Speech and Signal Processing University of Surrey Guildford UK The Federal State Unitary Enterprise State Research Institute of Aviation System Moscow Russia Department of Computational Mathematics Moscow State University Moscow Russia

In this paper, we present a grid structured morphological pattern spectrum based approach for off-line signature verification. The proposed approach has three major phases: preprocessing, feature extraction and verification. In the feature extraction phase, the signature image is partitioned into eight equally sized vertical grids and grid structured morphological pattern spectra for each grid is obtained. The grid structured morphological spectrum is represented in the form of 10-bin histogram and normalised to overcome the problem of scaling. The eighty dimensional feature vector is obtained by concatenating all the eight vertical morphological spectrum based normalised histogram. For verification purpose, we have considered two well known classifiers, namely SvM and MLP and conducted experiments on standard signature datasets namely CEDAR, GPDS-160 and MUKOS, a regional language (Kannada) dataset. The comparative study is also provided with the well known approaches to exhibit the performance of the proposed approach.

关键词： Feature extraction Support vector machines Forgery Accuracy Training Histograms Shape

来源：评论

学校读者我要写书评

暂无评论

Early Human Actions Detection using BK Sub-triangle Product

Early Human Actions Detection using BK Sub-triangle Product

引用

IEEE International conference on Fuzzy Systems

作者： Ekta vats Chee Kau Lim Chee Seng Chan Centre of Image and Signal Processing Faculty of Computer Science and Information Technology University of Malaya 50603 Kuala Lumpur Malaysia

ISBN: (纸本)9781467374293

Humans have the natural capabilities to perceive and anticipate actions of objects they interact with, including incidents happen within their neighborhood. These days, this important aspect of human perception has been widely incorporated in the computer vision framework to perform human action detection task. However, little attention is paid to the problem of detecting ongoing human actions as early as possible, which is crucial in a number of important applications ranging from video surveillance to health-care. In this paper, we propose a framework for detecting ongoing human actions as early as possible, i.e. detecting an action as soon as it begins, but before it completes. This is make possible with the used of Fuzzy Bandler and Kohout's sub-triangle product (BK subproduct) inference mechanism, utilizing the fuzzy capabilities in handling the arisen uncertainties during the human action recognition stage for a reliable decision making. Experimental results on publicly available dataset illustrate the effectiveness of the proposed method.

关键词： Early Human Actions Detection human perception proposed method Human images computer vision Action video monitoring INCIDENTS Inference mechanisms Frameworks

来源：评论

学校读者我要写书评

暂无评论

Multi-energy X-ray imaging with linear CZT pixel arrays and integrated electronics

Multi-energy X-ray imaging with linear CZT pixel arrays and ...

引用

IEEE Symposium on Nuclear Science (NSS/MIC)

作者： v.B. Cajipe R.F. Calderwood M. Clajus S. Hayakawa R. Jayaraman T.O. Tumer B. Grattan O. Yossifor NOVA Research and Development Inc. Riverside CA USA Grattan Engineering Loma Linda CA USA Ocean Side Consulting San Pedro CA USA

We have developed an X-ray image scanning detector system named NEXIS/sup TM/ with multi-energy capability that represents significant improvements over a previous version developed for automated baggage screening. The system is built around a 1 mm pitch CZT pixel detector array and a new 32-channel signal processor chip called the XENA/sup TM/ IC The CZT array and XENA IC together enable X-ray detection at photon fluxes exceeding one million events per second per channel, at up to five energy levels with medium energy resolution. In the NEXIS system, the detectors are abutted to form a seamless 128 mm long linear array on each detector module board. Up to sixteen boards can be tiled together to form scanning arms of various lengths. The board can also accommodate the use of 0.5 mm pixel pitch linear arrays. Multi-energy-bin images of compositionally and geometrically diverse objects have been acquired to demonstrate materials recognition via energy discrimination and image enhancement by suppression of scattered lower-energy radiation. applications of this multi-energy x-ray imaging system include security screening, medical imaging and industrial inspection.

关键词： X-ray imaging Pixel X-ray detection Sensor arrays Photonic integrated circuits X-ray detectors Event detection signal processing Energy states Energy resolution

来源：评论

学校读者我要写书评

暂无评论

Classification of various Colon Diseases in Colonoscopy video using Cross-wavelet Features

Classification of Various Colon Diseases in Colonoscopy Vide...

引用

International conference on Wireless Communications, signal processing and Networking

作者： Mainak Biswas Arghya Bhattacharya Debangshu Dey Electrical Engineering Department Camellia Institute of Technology Electrical Engineering Department Jadavpur University

ISBN: (纸本)9781467393393

One of the leading cause of deaths from cancer being colon cancer, which is observed in 10% of men and 9.2% in case of women [1], prevention technique has been approached where the major chance lies in detection in early stage during Optical Colonoscopy (OC) of the subject. This paper proposes a novel scheme for detecting various colon diseases through Colonoscopy videos using Cross-wavelet Transform (XWT) [2]. The datasets consist of five different colonoscopy videos which are split into five sets of image datasets i.e., Colon Polyp, Colon Ulcer, Chron's Diseases, Sigmoidal Colon and Normal Colon collected from American College of Gastroenterology [3], Medico Gastroenterologo [4] and Gastrolab [5]. A novel cross-wavelet transform, is opted for extracting the features from the datasets. This is followed by a multiclass classification using Multiclass Support vector Machine (MSvM). The recognition rate using this improved technique (XWT_MSvM) is 98.46% which is far better than any other well-known and benchmark procedures.

关键词： Colon cancer Optical Colonoscopy Colonoscopy Cross-wavelet Transform Support vector Machine (SvM)

来源：评论

学校读者我要写书评

暂无评论

Multi-temporal Fusion of Abdominal CT images for Effective Liver Cancer Diagnosis

Multi-temporal Fusion of Abdominal CT Images for Effective L...

引用

International conference on Wireless Communications, signal processing and Networking

作者： B. Lakshmi Priya S. Joshi Adaikalamarie K. Jayanthi Department of ECE Pondicherry Engineering College Puducherry India

ISBN: (纸本)9781467393393

Medical imaging has encountered exuberant developments obligated to its capacity to win data regarding human advantage for the motive of diagnosis. The main goal of liver perfusion imaging is to enliven the legitimacy in the characterization of liver disorders. Integration of information from the acquired multitemporal CT images can be achieved by means of image fusion, which is the process of integrating the relevant evidence from a set of source images directed towards a single composite image, where the resultant conception will be greater informative and meticulous than entire of the input images. To fuse multiple images, there should be one-to-one pixel correspondence which is achieved through image registration. In this paper, an fusion in NSCT domain using Phase Congruency and Log-Gabor fusion rule is proposed. The combination of these two fusion rules can preserve more information from source images and thus improves the quality of the fused image in a substantial manner. The statistical assessment metrics such as Spatial Frequency, SSIM, Cross Correlation, Cross Entropy and PSNR for the proposed framework provides an accurate analysis and comparison is carried out with the existing wavelet and Contourlet transforms.

关键词： Computed Tomography (CT) wavelet Transform Contourlet Transform Nonsubsampled Contourlet Transform (NSCT) Phase Congruency Log-Gabor Filter

来源：评论

学校读者我要写书评

暂无评论

Pipeline and Parallel Processor Architecture for Fast Computation of 3D-DWT using Modified Lifting Scheme

Pipeline and Parallel Processor Architecture for Fast Comput...

引用

International conference on Wireless Communications, signal processing and Networking

作者： C. Ashok Kumar B. K. Madhavi K. Lalkishore Department of ECE CMR Engineering College Department of ECE Sridevi Women Engineering College Department of ECE JNTU Anantapur

ISBN: (纸本)9781467393393

A pipelined parallel processing 3D DWT architecture is designed in this paper based on lifting scheme algorithm with 9/7 wavelet filters. The 3D DWT architecture process a 512×512 image with 8 groups of frames sequentially with improved throughput. The first stage computes 1D-DWT along the rows with 4 parallel processors, the memory interface with FIFO reduces latency between first stage and second stage that computes 2D DWT computation. 3D DWT computes wavelet coefficients in the temporal direction and designed to operate from 512*4 clock cycles in sequence along with 1D and 2D DWT computation. The FIFO designed synchronizes the data movement. Memories at every stage are designed to store the parallel processed data. The proposed architecture has high performance and is suitable for high speed, low power and portable applications. With utilization of 51% of slice registers 3D-DWT architecture implemented on virtex-5 FPGA and frequency of operation is 373 MHz. The designed DWT-IDWT can be used as IP Core.

关键词： Lifting scheme Pipelined architecture 3D DWT Parallel processing FPGAs

来源：评论

学校读者我要写书评

暂无评论

Siracusa: A Low-Power On-Sensor RISC-v SoC for Extended Reality visual processing in 16nm CMOS

Siracusa: A Low-Power On-Sensor RISC-V SoC for Extended Real...

引用

European conference on Solid-State Circuits (ESSCIRC)

作者： Moritz Scherer Manuel Eggimann Alfio Di Mauro Arpan Suravi Prasad Francesco Conti Davide Rossi Jorge Tomás Gómez Ziyun Li Syed Shakib Sarwar Zhao Wang Barbara De Salvo Luca Benini Dept. of Information Technology and Electrical Engineering ETH Zürich Switzerland Dept. of Electrical Electronic and Information Engineering University of Bologna Italy Meta Reality Labs Research USA

Extended Reality (XR) has become increasingly popular in recent years, with applications in entertainment, education, healthcare, and more. However, mass adoption of XR technology still faces several challenges in meeting stringent latency and power consumption requirements. On-sensor computing, where a capable XR processor is tightly packaged with an image sensor, is a promising technology that can help address these challenges as it provides several benefits, including reduced data analysis latency, low power consumption, small form factor, and greater privacy. This work introduces Siracusa, an on-camera computing platform for next-generation XR devices. Siracusa features a flexible mixed-precision Machine Learning (ML) accelerator and a cluster of application-tuned RISC-v cores, sharing a highly configurable on-chip memory hierarchy designed to minimize expensive data copies. As a result, Siracusa achieves a peak energy efficiency of 9.9 $\mathrm{T}\mathrm{O}\mathrm{p}/\mathrm{J}$ for deep neural network (DNN) inference, an increase of 1.2 x compared to similar designs, while supporting complex, heterogeneous application workloads, which combine ML with conventional signal processing and control.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Pseudoinverse Incremental Algorithm for Fast Training Deep Neural Networks with Application to Spectra Pattern Recognition

A Pseudoinverse Incremental Algorithm for Fast Training Deep...

引用

International Joint conference on Neural Networks

作者： Ke Wang Ping Guo Qian Yin A-Li Luo Xin Xin School of Computer Science and Technology Beijing Institute of Technology Image Processing and Pattern Recognition Laboratory Beijing Normal University Key Laboratory of Optical Astronomy National Astronomical Observatories Chinese Academy of Sciences

ISBN: (纸本)9781509006212

Deep learning scheme has received significant attention during these years, particularly as a way of building hierarchical representations from unlabeled data for a variety of signal and information processing tasks. However, deep neural networks suffer from slow learning speed since most used training algorithms are based on variations of the gradient descent algorithms which require iterative optimization and thus are time-consuming. In addition, a series of control parameters need to be specified empirically which lacks of the theoretical guidance, and current learning algorithms for deep networks are not very suitable to incremental learning scenario. To address these issues, we propose a fast learning scheme in this paper. The basic idea of our approach is to pre-train basic units such as auto-encoders of the deep architecture in an analytical way without any iterative optimization procedure. This scheme is also extended to an incremental learning version. The experimental result shows the superiority of our approach over the state-of-the-art gradient descent based algorithms. To demonstrate the impact of our algorithm on complicated real world applications, we give an example of its performance in astronomical spectra pattern recognition.

关键词： astronomical spectra Spectra Pattern recognition learning (artificial intelligence) Recognition (Psychology) Neural network algorithms control parameter Optimization Procedure descent Fasting Functional training Road

来源：评论

学校读者我要写书评

暂无评论

Artificial Intelligence Design on Embedded Board with Edge Computing for vehicle applications

Artificial Intelligence Design on Embedded Board with Edge C...

引用

Artificial Intelligence and Knowledge Engineering (AIKE)

作者： Ching-Lung Su Wen-Cheng Lai Yu-Kai Zhang Ting-Jia Guo Yi-Jiun Hung Hui-Chiao Chen National Yunlin University of Science and Technology Taiwan R.O.C

ISBN: (数字)9781728187082

ISBN: (纸本)9781728187099

This article proposes advanced driver assistance system (ADAS) from neural network by YOLO v3-tiny on vehicle platform of NXP S32v234 with edge computing to detect pedestrians and knights. The implemented embedded board has limitation to perform a lot of convolution. As proposed design need to reduce the amount of operation, the considered problem of reduced precision at the same time. The proposed architecture uses method of Squeeze Net and quantization to reduce the amount of operation about 46% and the precision has only slightly reduced. The proposed methods of image to column (Im2col) and memory efficient convolution (MEC) rearranges continuous matrix space to access. The proposed hardware of APEX uses to accelerate operations can reduce execution time and increase detection speed by ten multiples compared with YOLO v3-tiny architecture.

关键词： Quantization (signal) Convolution Neural networks Memory management Parallel processing Network architecture Edge computing

来源：评论

学校读者我要写书评

暂无评论

Electrode Selection for Noninvasive Fetal Electrocardiogram Extraction using Mutual Information Criteria

引用

AIP conference Proceedings 2006年第1期872卷 97-104页

作者： R. Sameni F. vrins F. Parmentier C. Hérail v. vigneron M. verleysen C. Jutten M. B. Shamsollahi *Laboratoire des Images et des Signaux (LIS) — CNRS UMR 5083 INPG UJF Grenoble France †Biomedical Signal and Image Processing Laboratory (BiSIPL) School of Electrical Engineering Sharif University of Technology Tehran Iran **Machine Learning Group (MLG) Microelectronics Laboratory Université Catholique de Louvain (UCL) Louvain‐La‐Neuve Belgium ‡Laboratoire Systèmes Complexes (LSC) — CNRS FRE 2494 Evry France

Blind source separation (BSS) techniques have revealed to be promising approaches for the noninvasive extraction of fetal cardiac signals from maternal abdominal recordings. From previous studies, it is now believed that a carefully selected array of electrodes well‐placed over the abdomen of a pregnant woman contains the required ‘information’ for BSS, to extract the complete fetal components. Based on this idea, previous works have involved array recording systems and sensor selection strategies based on the Mutual Information (MI) criterion. In this paper the previous works have been extended, by considering the 3‐dimensional aspects of the cardiac electrical activity. The proposed method has been tested on simulated and real maternal abdominal recordings. The results show that the new sensor selection strategy together with the MI criterion, can be effectively used to select the channels containing the most ‘information’ concerning the fetal ECG components from an array of 72 recordings. The method is hence believed to be useful for the selection of the most informative channels in online applications, considering the different fetal positions and movements.

关键词：

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：