資訊工程系智慧型系統實驗室 iLab 南台科技大學 1 A new social and momentum component adaptive PSO algorithm for image segmentation Expert Systems with Applications 38 (2011)

Slides:



Advertisements
Similar presentations
Particle swarm optimization for parameter determination and feature selection of support vector machines Shih-Wei Lin, Kuo-Ching Ying, Shih-Chieh Chen,
Advertisements

Modeling and Analysis of Random Walk Search Algorithms in P2P Networks Nabhendra Bisnik, Alhussein Abouzeid ECSE, Rensselaer Polytechnic Institute.
图像处理技术讲座(2) Digital Image Processing (2) 灰度直方图和二值化操作 Gray-level Histogram and Threshold 顾 力栩 上海交通大学 计算机系
電腦視覺 Computer and Robot Vision I Chapter2: Binary Machine Vision: Thresholding and Segmentation Instructor: Shih-Shinh Huang 1.
主講人:虞台文 大同大學資工所 智慧型多媒體研究室
Oriented Wavelet 國立交通大學電子工程學系 陳奕安 Outline Background Background Beyond Wavelet Beyond Wavelet Simulation Result Simulation Result Conclusion.
Information Theoretic Image Thresholding Laura Frost Supervisors: Dr Peter Tischer Dr Sid Ray.
A Comprehensive Study of Wavelet Transforms for SPIHT 台北科技大學資工所指導教授:楊士萱學生:廖武傑 2003/03/27.
1 高等演算法 Homework One 暨南大學資訊工程學系 黃光璿 2004/11/11. 2 Problem 1.
Adaptive Web Caching: Towards a New Caching Architecture Authors and Institutions: Scott Michel, Khoi Nguyen, Adam Rosenstein and Lixia Zhang UCLA Computer.
張 燕 光 資訊工程學系 Dept. of Computer Science & Information Engineering,
Elastic Registration in the Presence of Intensity Variations Source: IEEE Transactions on Medical Imaging, Vol. 22, No. 7, July 2003 Authors: Senthil Periaswamy,
A new predictive search area approach for fast block motion estimation Kuo-Liang Chung ( 鍾國亮 ) Lung-Chun Chang ( 張隆君 ) 國立台灣科技大學資訊工程系暨研究所 IEEE TRANSACTIONS.
Proxy Caching the Estimates Page Load Delays Roland P. Wooster and Marc Abrams Network Research Group, Computer Science Department, Virginia Tech 元智大學.
Managing key hierarchies for access control enforcement: Heuristic approaches Author: Carlo Blundo, Stelvio Cimato, Sabrina De Capitani di Vimercati, Alfredo.
南台科技大學 資訊工程系 Posture Monitoring System for Context Awareness in Mobile Computing Authors: Jonghun Baek and Byoung-Ju Yun Adviser: Yu-Chiang Li Speaker:
資訊工程系智慧型系統實驗室 iLab 南台科技大學 1 Optimizing Cloud MapReduce for Processing Stream Data using Pipelining 出處 : 2011 UKSim 5th European Symposium on Computer Modeling.
A Genetic Algorithms Approach to Feature Subset Selection Problem by Hasan Doğu TAŞKIRAN CS 550 – Machine Learning Workshop Department of Computer Engineering.
Design of double- and triple-sampling X-bar control charts using genetic algorithms 指導教授: 童超塵 作者: D. HE, A. GRIGORYAN and M. SIGH 主講人:張怡笳.
Copyright © 2012, A Minimum Cost Resource Allocation Approach for Cloud Data Centers 指導教授:王國禎 學生:連懷恩 國立交通大學資訊工程系 行動計算與寬頻網路實驗室 1.
Hierarchical Distributed Genetic Algorithm for Image Segmentation Hanchuan Peng, Fuhui Long*, Zheru Chi, and Wanshi Siu {fhlong, phc,
南台科技大學 資訊工程系 Automatic Website Summarization by Image Content: A Case Study with Logo and Trademark Images Evdoxios Baratis, Euripides G.M. Petrakis, Member,
南台科技大學 資訊工程系 A web page usage prediction scheme using sequence indexing and clustering techniques Adviser: Yu-Chiang Li Speaker: Gung-Shian Lin Date:2010/10/15.
智慧型系統實驗室 iLab 南台資訊工程 1 Evaluation for the Test Quality of Dynamic Question Generation by Particle Swarm Optimization for Adaptive Testing Department of.
南台科技大學 資訊工程系 Through-Walls Collaboration Adviser: Yu-Chiang Li Speaker: Gung-Shian Lin Date: 2010/04/08 Pervasive Computing, IEEE Volume 8, Issue 3, July-Sept.
資訊工程系智慧型系統實驗室 iLab 南台科技大學 1 A Static Hand Gesture Recognition Algorithm Using K- Mean Based Radial Basis Function Neural Network 作者 :Dipak Kumar Ghosh,
南台科技大學 精密製造實驗室, 102 年 8 月 8 日 班級 : 碩研機械一甲 姓名 : 鄧文斌 指導老師 : 曾信智.
數量方法 課程名稱數量方法 課程編碼 60M01701 系所代碼 / 名稱 06 / 國企系 開課班級碩研國企一甲 開課教師林士琪 學分 3.0 時數 3 必選修系定選修 南台科技大學 課程資訊.
CS654: Digital Image Analysis
Smooth Side-Match Classified Vector Quantizer with Variable Block Size IEEE Transaction on image processing, VOL. 10, NO. 5, MAY 2001 Department of Applied.
數位影像處理概論 課程名稱數位影像處理概論 課程編碼 30N06701 系所代碼 / 名稱 03 / 電子系 開課班級夜四技電子四甲 夜四技電子四乙 開課教師賴培淋 學分 3.0 時數 3 必選修選修 南台科技大學 課程資訊.
Intelligent Space 國立台灣大學資訊工程研究所 智慧型空間實驗室 Service Behavior Consistency in the OSGi Platform Authors Y.Qin, H.Hao,L.Jun, G.Jidong and L.Jian Proceedings.
A Study on the Exponential Decay for the Bearing Defect Diagnosis 沈毓泰 南台科技大學 機械系 教授.
Intelligent Database Systems Lab 國立雲林科技大學 National Yunlin University of Science and Technology Advisor : Dr. Hsu Graduate : Yu Cheng Chen Author: Manoranjan.
南台科技大學 資訊管理研究所 LEARNING SEQUENCES CONSTRUCTION USING VAN HIELE MODEL AND BAYESIAN NETWORK J. Wey Chen, Professor Department of Information Management Southern.
Boosted Particle Filter: Multitarget Detection and Tracking Fayin Li.
CFTP - A Caching FTP Server Mark Russell and Tim Hopkins Computing Laboratory University of Kent Canterbury, CT2 7NF Kent, UK 元智大學 資訊工程研究所 系統實驗室 陳桂慧.
Image Segmentation by Histogram Thresholding Venugopal Rajagopal CIS 581 Instructor: Longin Jan Latecki.
國立清華大學高速通訊與計算實驗室 NTHU High-Speed Communication & Computing Laboratory Optimal Provisioning for Elastic Service Oriented Virtual Network Request in Cloud.
LUT Method For Inverse Halftone 資工四 林丞蔚 林耿賢. Outline Introduction Methods for Halftoning LUT Inverse Halftone Tree Structured LUT Conclusion.
Applying the Resonance Frequencies of Mechanical System in the Analysis of Bearing Vibration 沈毓泰 南台科技大學 機械系 副教授.
Intelligent Database Systems Lab 國立雲林科技大學 National Yunlin University of Science and Technology 1 Adaptive FIR Neural Model for Centroid Learning in Self-Organizing.
Digital Image Processing
Intelligent Space 國立台灣大學資訊工程研究所 智慧型空間實驗室 Brainstorming Principles Reporter Chun-Feng Liao Sep 12,2005 Source D.Bellin and S.S.Simone, ”Brainstorming: A.
A Flexible Interleaved Memory Design for Generalized Low Conflict Memory Access Laurence S.Kaplan BBN Advanced Computers Inc. Cambridge,MA Distributed.
Evaluation of Image Segmentation algorithms By Dr. Rajeev Srivastava.
無線網路安全性設計實務 課程名稱無線網路安全性設計實務 課程編碼 G0M04001 系所代碼 / 名稱 0G / 資工系 開課班級碩研資工二甲 開課教師許子衡 學分 3.0 時數 3 必選修選修 南台科技大學 課程資訊.
DEVELOPMENT OF EDUCATIONAL CAMERA GAMES FOR CHILDREN XIE Fei, CAI Shan, CHENG Ben, CHEN Chao College of Information System & Management, National University.
南台科技大學 資訊工程系 Data hiding based on the similarity between neighboring pixels with reversibility Author:Y.-C. Li, C.-M. Yeh, C.-C. Chang. Date:
Byzantine Agreement in the Presence of Mixed Faults on Processor and Links Hin-Sing Siu, Yeh-Hao Chin, Wei-Pang Yang Senior Member, IEEE Computer Society,
南台科技大學 資訊工程系 An effective solution for trademark image retrieval by combining shape description and feature matching 指導教授:李育強 報告者 :楊智雁 日期 : 2010/08/27.
1 Problems of Perfect Multi- Secret Sharing Schemes Advisor: 阮夙姿教授 Presenter: 蔡惠嬋 Date: 2008/08/11 國立暨南國際大學資訊工程學系.
Intelligent Database Systems Lab Advisor : Dr. Hsu Graduate : Jian-Lin Kuo Author : Aristidis Likas Nikos Vlassis Jakob J.Verbeek 國立雲林科技大學 National Yunlin.
EM Algorithm 主講人:虞台文 大同大學資工所 智慧型多媒體研究室. Contents Introduction Example  Missing Data Example  Mixed Attributes Example  Mixture Main Body Mixture Model.
Improving the WWW: Caching or Multicast? Pablo RodriguezErnst W. BiersackKeith W. Ross Institut EURECOM 2229, route des Cretes. BP , Sophia Antipolis.
SemiBoost : Boosting for Semi-supervised Learning Pavan Kumar Mallapragada, Student Member, IEEE, Rong Jin, Member, IEEE, Anil K. Jain, Fellow, IEEE, and.
A distributed PSO – SVM hybrid system with feature selection and parameter optimization Cheng-Lung Huang & Jian-Fan Dun Soft Computing 2008.
Zhaoxia Fu, Yan Han Measurement Volume 45, Issue 4, May 2012, Pages 650–655 Reporter: Jing-Siang, Chen.
Intelligent Database Systems Lab 國立雲林科技大學 National Yunlin University of Science and Technology 1 Intelligent Exploration for Genetic Algorithms Using Self-Organizing.
義守大學資訊工程學系 作者:郭東黌, 張佑康 報告人:徐碩利 Date: 2006/11/01
Discrete Fourier Transform (DFT)
利用iBeacon設計感知教室學習活動歷程雲端服務系統應用於翻轉教室教學 Using iBeacon to Develop a Cloud-based Awareness Classroom Learning Activity Portfolio System Applied in Flipped Classroom.
IMAGE SEGMENTATION USING THRESHOLDING
Shih-Wei Lin, Kuo-Ching Ying, Shih-Chieh Chen, Zne-Jung Lee
Source: Pattern Recognition Letters 29 (2008)
Longest Common Subsequence (LCS)
Simulated Annealing & Boltzmann Machines
Edit Distance 張智星 (Roger Jang)
海軍軍官學校 「迎向海洋,逐夢啟航」學術研討會
Presentation transcript:

資訊工程系智慧型系統實驗室 iLab 南台科技大學 1 A new social and momentum component adaptive PSO algorithm for image segmentation Expert Systems with Applications 38 (2011) 4998–5004 Akhilesh Chander, Amitava Chatterjee, Patrick Siarry, Reporter : Yu Chih Lin

資訊工程系 Outline ﻪIntroduction ﻪLiterature ﻪMethods ﻪExperiments ﻪConclusion 2

資訊工程系 Introduction(1/4) ﻪImage segmentation is useful in separating Background Discriminating objects (gray-levels) 3

資訊工程系 Introduction(2/4) 4 ﻪUsually image segmentation can be classified as Bi-level thresholding Multilevel thresholding

資訊工程系 Introduction(3/4) 5 ﻪProposed an iterative procedure Determinies the number of thresholds Positions of these thresholds

資訊工程系 Introduction(4/4) 6 ﻪProposed a variant of PSO based stochatic algorithm Utilized for gray image segmentation purpose Used thresholds as initial values

資訊工程系 Literature(1/3) 7 ﻪThresholding techniques can be classified into two types Optimal thresholding methods Property based thresholding methods

資訊工程系 Literature(2/3) 8 ﻪBi-level thresholding has a problem Only one gray value to be found Get more and more complex by employing Multilevel thresholoding

資訊工程系 Literature(3/3) 9 ﻪAll of these methods have a common problem Conputational complexity rises exponentially ﻪUse bi-level Otsu thresholding method Multilevel thresholding with less computational complexity

資訊工程系 Methods(1/13) 10 ﻪProposed three methods Iterative scheme Variant of PSO Entropy criterion based fitness measure

資訊工程系 Method - Iterative scheme 11

資訊工程系 Methods(2/13) 12

資訊工程系 Methods(3/13) ﻪIterative scheme 13

資訊工程系 Methods(4/13) ﻪSelf-iterative scheme 14

資訊工程系 Methods(5/13) 15 Fig1. Time complexities of the iterative scheme

資訊工程系 Methods(6/13) 16 Fig2. Uniformity values of the proposed iterative scheme

資訊工程系 Method -Variant of PSO 17

資訊工程系 Methods(7/13) 18

資訊工程系 Methods(8/13) 19

資訊工程系 Methods(9/13) 20

資訊工程系 Methods(10/13) ﻪThe thresholds are utilized as initial thresholds Randomly distributed around the initial thresholds Find the optimal solution more efficiently 21

資訊工程系 Method -Entropy criterion based fitness measure 22

資訊工程系 Methods(11/13) ﻪUse Otsu’s multi-threshold entropy measure ﻪUse as objective function in PSO ﻪGray levels of a image range over [ 0, L-1 ] ﻪh(i) denote the occurrence of gray-level i 23

資訊工程系 Methods(12/13) 24

資訊工程系 Methods(13/13) 25

資訊工程系 Experiments(1/11) ﻪProposed scheme was compared with GA-learning-Otsu algorithm Gaussian-smoothing method Symmetry-duality method ﻪUse two image, Lena and Pepper (512*512) 26

資訊工程系 Experiments(2/11) ﻪCompared with the results of basic PSO employing linearly decreasing inertia weight. ﻪTwo additional images, House and Elaine 27

資訊工程系 Experiments(3/11) 28

資訊工程系 Experiments(4/11) 29 IterationscNo. of particlesLena image Uniformity values House imageElaine imagePepper image Performance of PSO for varied number of particles Table.1 Uniformity values for c=3 and varying iterations for four benchmark images

資訊工程系 Experiments(5/11) 30 No. of particlescIterationsLena image Uniformity values House imageElaine imagePepper image Table.2 Uniformity values for c=3 and varying number of particles, four benchmark images

資訊工程系 Experiments(6/11) 31 Fig.3 (a) uniformity vs. number of particles (b) uniformity vs. number of iterations.

資訊工程系 Experiments(7/11) 32 ImagescThresholds (Proposed method) Proposed Method Uniformity values Gaussian- smoothing method Symmetry- duality GA- method Lena295, ,151, ,137,163, ,114,143,167, Pepper280, ,122, ,120,148, ,98,136,170, Table.3 Compared with those of Gaussian, Symmetry-duality and GA-based methods

資訊工程系 Experiments(8/11) 33 Imagesc Uniformity values Proposed method Basic PSO Threshold values Proposed method Basic PSO House ,178122, ,143, ,155, ,142,179,20188,153,176, ,131,179,206,21582,142,174,213,217 Elaine ,160122, ,166,202126,156, ,148,176,192112,192,166, ,122,149,174,21090,127,180,184,213 Table.4 Results of the proposed method for House and Elaine images

資訊工程系 Experiments(9/11) 34 ImagescUniformity values Threshold values Lena ,111,138,163,185 Pepper ,81,117,152,161,181 House ,143,181 Elaine ,114,13,148,181,213 Table.5 Results of the proposed PSO method employing self-iterative scheme to find number and values of initial thresholds

資訊工程系 Experiments(10/11) 35 Fig.4 Gray images and the corresponding thresholded images of Lena (c = 5 thresholds) and Pepper (c = 6 thresholds) using the proposed method.

資訊工程系 Experiments(11/11) 36 Fig.5 Gray images and the corresponding thresholded images of Elaine (c = 6 thresholds) and House (c = 3 thresholds) using the proposed method.

資訊工程系 Conclusion(1/2) 37 ﻪEmploying a new proposed variant of PSO For an optimal multilevel thresholding algorithm ﻪProposes an iterative scheme to obtain initial thresholds

資訊工程系 Conclusion(2/2) ﻪ Useful for practical situations because the computational complexity grows linearly with the number of thresholds ﻪProposed PSO algorithm makes a new contribution 38

資訊工程系 Thank you for listening 39