PSIVT 2017: 8th Pacific-Rim Symposium on Image and Video Technology

计算机视觉

需积分: 10 87 浏览量更新于2024-07-17 收藏 67.31MB PDF 举报

身份认证购VIP最低享 7 折!

30元优惠券

"Image and Video Technology: 8th Pacific-Rim Symposium, PSIVT 2017, Wuhan" 本书是2017年11月在中国武汉举行的第8届太平洋沿岸图像与视频技术研讨会（PSIVT 2017）的经同行评审的会后论文集。PSIVT是一个高质量的系列研讨会，旨在为研究者和实践者提供一个平台，讨论和贡献图像与视频技术的理论进步和实际应用。该书包含了39篇经过修订的论文，这些论文是从91篇提交的论文中精心挑选出来的。每篇论文都反映了计算机视觉领域的最新研究成果和技术创新，涵盖了图像处理、视频分析、模式识别、机器学习等多个方面。计算机视觉作为现代信息技术的重要组成部分，对于图像和视频的理解、解析和处理有着深远的影响。编辑团队包括Manoranjan Paul、Carlos Hitoshi和Qingming Huang等人，他们都是该领域的知名专家。《Lecture Notes in Computer Science》系列出版了这本书，该系列自1973年开始出版，由一批顶尖的计算机科学家担任顾问委员会成员，包括来自世界顶级大学和研究机构的专家。研讨会涵盖了广泛的议题，如图像恢复与增强、目标检测与跟踪、深度学习在视觉任务中的应用、视频编码与压缩、视觉内容理解以及基于内容的检索等。这些技术不仅推动了学术研究的进步，也为实际应用如自动驾驶、安防监控、虚拟现实、社交媒体分析等领域提供了强大的技术支持。例如，深度学习在计算机视觉中的应用已经成为近年来的热点。通过使用深度神经网络，研究人员能够训练模型来自动识别图像和视频中的对象，甚至理解和预测场景的行为。这种技术已经在人脸识别、图像分类、物体检测等方面取得了显著成果，并且正在不断进化，以解决更复杂的视觉问题。此外，图像和视频处理技术的发展还涉及到了视频内容分析，包括行为识别、事件检测和异常检测。这些技术对于智能监控系统、安全防范和视频大数据分析至关重要。它们可以帮助系统自动检测潜在的安全威胁，或者在海量视频数据中找到特定的事件或行为模式。 "Image and Video Technology: 8th Pacific-Rim Symposium, PSIVT 2017, Wuhan"这一资源集合了全球顶尖的研究成果，展示了计算机视觉领域的前沿进展，对了解和学习图像与视频处理技术的专业人士来说，是一份极具价值的参考资料。它不仅促进了跨学科的合作，也为未来的技术创新奠定了坚实的基础。

资源详情

资源推荐

Age Estimation with Local Ternary Directional Patterns. . . . . . . . . . . . . . . . 421

Raphael Angulu, Jules R. Tapamo, and Aderemi O. Adewumi

Computer Vision and Applications

Predicting Turn-Taking by Compact Gazing Transition Patterns

in Multiparty Conversation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 437

Li Tian, Qi Jia, and Zhen Zhu

Intelligent Assistant for People with Low Vision Abilities . . . . . . . . . . . . . . 448

Oleksandr Bogdan, Oleg Yurchenko, Oleksandr Bailo,

Francois Rameau, Donggeun Yoo, and In So Kweon

Using Facial Expression Recognition for Crowd Monitoring. . . . . . . . . . . . . 463

Ross Philip Holder and Jules-Raymond Tapamo

Block-Wise Gaze Estimation Based on Binocular Images. . . . . . . . . . . . . . . 477

Xuemei Wu, Jing Li, Qiang Wu, Jiande Sun, and Hua Yan

Integrated Multi-scale Event Verification in an Augmented

Foreground Motion Space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 488

Qin Gu, Jianyu Yang, Wei Qi Yan, and Reinhard Klette

Author Index ............................................ 501

XVI Contents

Multiset Canonical Correlation Analysis:

Texture Feature Level Fusion of Multiple

Descriptors for Intra-modal Palmprint

Biometric Recognition

Raouia Mokni

)

, Anis Mezghani

, Hassen Drira

, and Monji Kherallah

Faculty of Economics and Management of Sfax, University of Sfax, Sfax, Tunisia

raouia.mokni@gmail.com

University of Sfax, Sfax, Tunisia

anis.mezghani@gmail.com

Institut Mines-T´el´ecom/T´el´ecom Lille, CRIStAL (UMR CNRS 9189),

Lille, France

hassen.drira@telecom-lille.fr

Faculty of Sciences of Sfax, University of Sfax,

Road Soukra Km 3, 3038 Sfax, Tunisia

monji.kherallah@fss.usf.tn

Abstract. This paper describes a novel intra-modal feature fusion for

palmprint recognition based on fusing multiple descriptors to analyze the

complex texture pattern. The main contribution lies in the combination

of several texture features extracted by the Multi-descriptors, namely:

Gabor Filters, Fractal Dimension and Gray Level Concurrence Matrix.

This means to their eﬀectiveness to confront the various challenges in

terms of scales, position, direction and texture deformation of palmprint

in unconstrained environments. The extracted Gabor ﬁlter-based texture

features from the preprocessed palmprint images to be fused with the

Fractal dimension-based-texture features and Gray Level Concurrence

Matrix-based texture features using the Multiset Canonical Correlation

Analysis method (MCCA). Realized experiments on three benchmark

datasets prove that the proposed method surpasses other well-known

state of the art methods and produces encouraging recognition rates

by reaching 97.45% and 96.93% for the PolyU and IIT-Delhi Palmprint

datasets.

Keywords: Palmprint

· Texture analysis · Gabor Filters

Fractal Dimension

· Gray Level Concurrence Matrix

Information fusion

· Multiset Canonical Correlation Analysis (MCCA)

1 Introduction and related works

Human biometrics has become an area of tremendous importance and poten-

tial. By human biometrics we mean the use of physiological characteristics,

 Springer International Publishing AG, part of Springer Nature 2018

M. Paul et al. (Eds.): PSIVT 2017, LNCS 10749, pp. 3–16, 2018.

https://doi.org/10.1007/978-3-319-75786-5

4 R. Mokni et al.

of human body parts and their appearances, to identify individual human beings

in the course of their daily activities. The appearances of body parts, especially in

imaged data, have a large variability and are inﬂuenced by their texture patterns,

colors, lighting environment, and so on. Therefore, the biometrics researchers

have focused on body parts and images that try to minimize this variability

within class (subjects) and maximize it across classes. This paper investigates

the identiﬁcation of the persons from one of the physical body parts, namely:

Palmprint. Till now, more attention by many researchers has been paid to palm-

print recognition systems thanks to its low cost, aﬀordable materials, high recog-

nition accuracy, etc. These systems have been grouped into two major ways: (1)

Line-based structural approaches and (2) Texture-based global approaches.

Line-based structural approaches group essentially the principal lines [20], the

wrinkles [4] and the ridges [5]. However, the major drawback of these approaches

is the presence of the computation cost to extract its features, which means to

a signiﬁcant computational time. Unluckily, those features alone cannot provide

satisfactory information to identify the person eﬃciently. Otherwise, texture-

based global approaches incorporate the global information of images involving

the texture. Various descriptors have been applied for analyzing the palmprint

texture pattern, such as the Local Binary Pattern (LBP) [11], Fractal Dimension

(FD) [7], Discrete cosine Transform (DCT) [2], Gabor Wavelets [28], Gabor ﬁlter

(GF) [16], Gray Level Co-occurrence Matrix (GLCM) [19], etc.

Many researchers claim that they spend less eﬀort on analyzing the texture

information contrast the methods that are relied on analyzing the principal line

shape information. Therefore, the accuracy of their recognition commonly relies

on the reliability of the features.

In this paper, we propose a novel intra-modal feature fusion approach for

palmprint identiﬁcation based on fusing multiple-texture features extracted by

diﬀerent descriptors using the feature fusion level. For this end, Gabor ﬁlter,

Fractal Dimension and Gray Level Co-occurrence Matrix descriptors are used to

analyze the palmprint texture. Subsequently, we apply the dimensionality reduc-

tion technique based on the Principal Component Analysis in order to select the

more discriminant feature. Then, we fuse these obtained features at feature fusion

level relying on the Multiset Canonical Correlation Analysis method.

The reminder of this paper is structured as follows: In Sect. 2, the pipeline of

the proposed intra-modal feature fusion approach and its modules are detailed.

Experimental evaluation and comparison of the accuracy of our proposed app-

roach with other state-of-the-art methods are presented in Sect. 3. Lastly, a con-

clusion and some perspectives are drawn in Sect. 4

2 Proposed Intra-modal Texture-Feature Fusion

for Palmprint Identiﬁcation

The pipeline of our suggested intra-modal feature fusion approach for palm-

print identiﬁcation summarizes the following modules: (1) Hand Acquisition

剩余514页未读，继续阅读

wang1062807258

粉丝: 13
资源: 272

PSIVT 2017: 8th Pacific-Rim Symposium on Image and Video Technol...

最新资源

PSIVT 2017: 8th Pacific-Rim Symposium on Image and Video Technol...

023-05-31 16:14:46.672 ERROR 27104 --- [eate-1159694255] com.alibaba.druid.pool.DruidDataSource : create connection SQLException, url: jdbc:mysql://rm-bp143516c39alb8th0o.mysql.rds.aliyuncs.com:3306/sys_db, errorCode 0, state 08S01怎么理解

关于ai绘画的参考文献

8th多项式拟合法MATLAB算法

discrete mathematics and its applications, 8th edition

formel-q new 8th 试卷

请给出C语言学习的书籍，同时列出书的链接。

BP神经网络 PI控制的文献

fluid mechanics-frank m white solution 8th

关于ssm框架的外国参考文献

谢希仁计算机网络第八版 参考文献格式

systems analysis and design, 8th edition, dennis, wixom, roth, wiley

led <= {led[6:0],led[7]};

computer networking top-down approach 8th edition pdf

三次谐波注入PWM参考文献

calculus-early transcendentals 8th edition james stewart pdf

利用8阶RK积分公式计算开普勒卫星积分轨道的MATLAB程序

P0口的2个扩展口作为段控口和位控口，通过使用P2.6和P2.7对SN74HC573芯片的使能。P3.2（INT0）作为按钮输入口构成一个“0#～#7”的8个按钮和8个LED的显示按钮电路，系统复位时，显示“HELLO-- ”。请写出基于at89c51的汇编代码

A Robust Rectangular Object Detection Method using Multiple Features", 2019 8th International Conference on Computer and Communication Engineering (ICCCE) 链接

最新资源

谢希仁计算机网络第八版参考文献格式