CHAPTER 5 OPTIMUM SPECTRUM MASK BASED MEDICAL IMAGE FUSION USING GRAY WOLF OPTIMIZATION

Size: px
Start display at page:

Download "CHAPTER 5 OPTIMUM SPECTRUM MASK BASED MEDICAL IMAGE FUSION USING GRAY WOLF OPTIMIZATION"

Transcription

1 CHAPTER 5 OPTIMUM SPECTRUM MASK BASED MEDICAL IMAGE FUSION USING GRAY WOLF OPTIZATION Multimodal medical image fusion is an efficient technology to combine clinical features in to a single frame called fused image. The visual quality and edge enhancement are real challenges in clinical diagnosis and computer guided surgery. To overcome the limitations of existing fusion technique, proposed a new medical image fusion technique called Optimum Spectrum Mask Fusion (OSMF). Unlike the conventional masking technique, in OSMF the scale values are selected dynamically using a recently introduced optimization algorithm called Gray Wolf Optimization (GWO). 5.1 OPTIMUM SPECTRUM MASK FUSION This section is divided into 2 subsections as follows, need of optimum value in spectrum mask and proposed optimum spectrum mask fusion Need of Optimum Scale Value in Spectrum Mask In conventional masking techniques, the static scale values are using irrespective of the varying multimodal input images.the scale values, which selected for one set of input images might be ineffective in other input images because of multimodal acquisition devices. The gain factor of masking based techniques are controlled by scale values, the optimum scale value can adaptively control the gain irrespective of input images. 67

2 5.1.2 Proposed Optimum Spectrum Mask Fusion In this section, proposed a new approach in multi modal medical image fusion called, Optimum Spectrum Mask Fusion (OSMF). The objective of OSMF is to obtain a contrast enhanced multimodal fusion using optimum masking approach. The general GWO algorithm is used to adjust the optimum scale values (s 1 opt, s 2 opt ) for mask creation. The block diagram of proposed OSMF is given in Figure 5.1. In which, the input images are transformed into Fourier space using the Equation (5.1). Where F (u,v) denotes the Fourier space of input image I(x, y). The obtained Fourier spectrum of the input images are optimally scaled using scale values [s 1 opt, s 2 opt ] are obtained using GWO algorithm. In formulated masks, the mask 1 provides the structural features and mask 2 contains the soft tissue details for MR-T2 cases and metabolic features in SPECT and PET cases. The respective formulated masks are called spectrum mask1 and spectrum mask 2. The obtained mask images are fused using pixel based averaging fusion rule, the formulated fused image is Fourier domain. The inverse transform is used for getting the spatial domain fused image; the mathematical expression is given in Equation (5.2). M 1 x=0 N 1 y=0 F (u,v) = I(x, y) e j2π(u x M +v y N ) (5.1) M 1 u=0 N 1 v=0 I F (x,y) = F(u, v) e j2π(u x M +v y N ) (5.2) In which, M and N represents the number of rows and columns of input medical image, the considered images with size of M = N =256, I F (x,y) is the resultant spatial domain fused image. 68

3 Figure 5.1 Block diagram of Optimum Spectrum Mask Fusion 5.2 GRAY WOLF OPTIZATION This section explains the application of conventional GWO algorithm in multimodal medical image fusion applications. Unlike conventional masking technique, instead of conventional scaling, OSMF have optimum scale values. The GWO is a swarm intelligence algorithm, based on the hunting mechanism of gray wolves family in nature. The steps of GWO are described below: Initialization of Gray Wolf Positions In OSMF application, the range of scale value [s 1, s 2 ] for GWO is considered in [0, 1]. The scale values are considered as search agents in GWO. In OSMF analysis, considered 50 iterations and 50 gray wolfs in each pack. The pack size and iterations are fixed using trial and error approach Fitness Function The GWO algorithm is applied for optimal selection of scale values in OSMF. Mutual Information () is a well-known performance measure for multi 69

4 modal fusion and it provides the amount of information retained in the fused image frame. In OSMF, is used as fitness function and it provides better fusion assessment. The mathematical expression for is given in Equation (5.3). (x, y) = P(x, y)log P(x,y) x y (5.3) P(x)P(y) In Equation (5.3), P(x, y) is probability distribution function, P(x) and P(y) are marginal probability functions of both the modalities respectively Social Hierarchy of Gray Wolf Family In nature, gray wolfs are known for social leadership and they are living in a pack. The average member of gray wolf pack consists of Hierarchical order of gray wolf pack is divided in four levels, alpha(α), beta (β), delta(δ) and omega(ω). The first level hierarchy, alpha is making decisions regarding hunting, sleeping place, time to wake, and so on. Beta is second level ranking in a gray wolf pack, the best subordinates to alpha in activates. Omega is the lowest level order in grey wolf pack, plays the role of scapegoat. In pack, omega wolves always have to submit to all the other wolves. In gray wolf pack, if a wolf is not an alpha, beta, or omega, which is called delta wolf and they have to submit to alphas and betas, but they dominate the omega in gray wolf social hierarchy Encircling Prey During their hunting, before the attack gray wolf encircles the prey. The mathematical expressions are given in Equations (5.4) - (5.7). D = C. X p (t) X(t) (5.4) X(t + 1) = X p (t) A. D (5.5) 70

5 A = 2a. r 1 a (5.6) C = 2. r 2 (5.7) In which, X p is the position vector of the prey, t indicates current iteration, X represents the position vector of gray wolf. Where A and C are coefficient vectors and D is the direction vector. In which, r 1, r 2 are random vectors in the range of [0, 1] and a linearly decreases from 2 to Hunting The gray wolves have the intelligence to find the location of its prey. In which, hunt is leaded under the guidance of alpha. The beta and delta might also help in hunting mechanism. The alpha wolfs have the best knowledge about prey position followed by beta and delta in a pack. In GWO, each iterations forwards three best search agents such as, alpha, beta and delta to the next level, all other agents in gray wolf family are treated as gamma. The mathematical expressions are given in Equations (5.8) - (5.10). D = C 1. X X, D β = C 2. X β X, D δ = C 3. X δ X (5.8) X 1 = X A 1. (D ),X 2 = X β A 2. (D β ),X 3 = X δ A 3. (D δ ) (5.9) X(t + 1) = X 1 +X 2 +X 3 3 (5.10) Attacking Prey During the hunting, when the prey stops the movement gray wolves finish the hunt by attacking the prey. During the process of approaching the prey (as per mathematically standard) the gray wolfs need to decrease the value of a, where a is updated from 2 to 0 during the course of iterations. 71

6 5.2.6 Search for Prey The GWO usually perform search based on the positions of the alpha, beta, and delta. The gray wolfs diverge from each other to search for prey and converge to attack the prey. The mathematical modeling of divergence, considered with random values larger than 1 or less than -1 to impel the search agent to diverge away from prey, in which, C vector is treated as the effect of obstacles to approach prey, which appear in the paths of wolves and literally inhibit them, by swiftly and calmly approaching prey and each search agent updates its distance from the prey. The pseudo code for GWO algorithm is given in Figure 5.2. Figure 5.2 Pseudo code of GWO 72

7 5.3 RESULTS AND DISCUSSION The OSMF is tested with 995 sets of medical images and the average results are given. The multi modal medical image fusions used in this section are MR- SPECT fusion, MR-PET fusion, MR-CT fusion and MR: T1- T2 fusion. The various measures used in quantitative analysis are Mutual Information (), Entropy, Standard Deviation (SD),Edge Strength (Q AB F), Structural Similarity Index (SSIM) and Peak Signal to Noise Ratio (PSNR).The mathematical expressions of above measures are given in chapter 3. The OSMF is compared with various states of art fusion techniques such as DCT, IHS, DFT and DWT. In GWO, optimum scale analysis is based on 50 iterations and evaluated the best scale values. The obtained optimum scale values as are given in Table 5.1.In which, the average of best scale values of all the clinical cases of each fusion application is given. The computational analysis is done using Matlab 2010a with Pentium dual core processor with speed 2.30 GHz. The size of the considered input images are 256x256 in JPEG format. Table 5.1 Implementation results of GWO Type of images s 1 best s 2 best MR-SPECT MR-PET MR-CT MR-T1-T

8 5.3.1 MR-SPECT Fusion The MR-SPECT fused modality provides the anatomical structure and metabolic features in a single frame. The OSMF technique is tested for 140 patients MR and SPECT images for performance analysis. The average values of quantitative results are given in Table 5.2. Examples of four MR-SPECT cases among the 140 sets are given in Figure 5.3. The quantitative results shows that the OSMF based fusion has grater performance than that of conventional transform techniques. The PSNR value of conventional FFT is whereas the proposed spectrum masking has The SSIM value of conventional FFT is while the proposed OSMF has Entropy value of FFT based fusion technique is given whereas the proposed technique has Edge quality measure for conventional FFT is while the proposed technique has The obtained results show that the proposed OSMF has greater amount of fusion in terms of structural mapping, edge quality, contrast enhancement and retained information than that of other existing fusion techniques for MR-SPECT multimodal medical image fusion approaches. Table 5.2: Results of MR-SPECT fusion Entropy Q AB F STD SSIM PSNR DCT DWT FFT IHS Proposed

9 MR SPECT MR-SPECT fused MR SPECT MR-SPECT fused MR SPECT MR-SPECT fused SPECT MR-SPECT fused MR Figure 5.3 Examples of MR-SPECT fusion using proposed technique 75

10 Entropy STD SSIM PSNR 10 0 DCT DWT FFT IHS Proposed Figure 5.4 Results of MR-SPECT fusion for Case 1 dataset Entropy STD SSIM PSNR 0 DCT DWT FFT IHS Proposed Figure 5.5 Results of MR-SPECT fusion for Case 2 dataset 76

11 5.3.2 MR-PET Fusion In MR-PET fused modality provides the structural and metabolic information in a single image. The proposed OSMF technique is tested for 375 patients MR and PET images. The average values of MR-PET fusion are compared with various existing techniques and are given in Table 5.3. Examples of four MR-PET fusion cases among the total 375 sets are given in Figure 5.3. The quantitative results shows that the OSMF based fusion has higher fusion than that of existing transform based techniques. The PSNR value of conventional FFT is whereas the OSMF has The SSIM value of conventional FFT is while the OSMF has Entropy value of FFT based fusion technique is given whereas the OSMF technique has Edge quality measure for conventional FFT is while the proposed technique has The obtained MR-PET fusion results show that the OSMF has greater amount of fusion in terms of structural mapping, edge quality, contrast enhancement and retained information than that of other existing fusion techniques. Table 5.3 Results of MR-PET fusion Entropy Q AB F STD SSIM PSNR DCT DWT FFT IHS Proposed

12 MR PET MR-PET fused MR PET MR-PET fused MR PET MR-PET fused MR PET MR-PET fused Figure 5.6 Examples of MR-PET fusion using proposed technique 78

13 Entropy STD SSIM PSNR 0 DCT DWT FFT IHS Proposed Figure 5.7 Results of MR-PET fusion for Case 1 dataset Entropy STD SSIM PSNR 0 DCT DWT FFT IHS Proposed Figure 5.8 Results of MR-PET fusion for Case 2 dataset 79

14 5.3.3 MR-CT Fusion In MR-CT fused modality provides the soft and hard tissues in a single image. The proposed OSMF technique is tested for 210 patients MR and CT images. The average values of quantitative fusion metrics for MR-CT in comparison with states of art fusion techniques are given in Table 5.4. Examples of two MR and CT cases among the 210 sets are given in Figure 5.4. The clinical details of both examples are given in chapter 3. The quantitative results shows that the OSMF based fusion has grater performance than that of conventional transform techniques. The PSNR value of conventional FFT is whereas the proposed spectrum masking has The SSIM value of conventional FFT is while the proposed OSMF has Entropy value of FFT based fusion technique is given whereas the proposed technique has Edge quality measure for conventional FFT is while the proposed technique has The obtained results show that the proposed OSMF has greater amount of fusion in terms of structural mapping, edge quality, contrast enhancement and retained information than that of other existing fusion techniques. Table 5.4 Results of MR-CT fusion Entropy Q AB F STD SSIM PSNR DCT DWT FFT IHS Proposed

15 MR CT MR-CT fused MR CT MR-CT fused MR CT MR-CT fused MR CT MR-CT fused Figure 5.9 Examples of MR-CT fusion using proposed technique 81

16 Entropy STD SSIM PSNR 10 0 DCT DWT FFT IHS Proposed Figure 5.10 Results of MR-CT fusion for Case 1 dataset Entropy STD SSIM PSNR 0 DCT DWT FFT IHS Proposed Figure 5.11 Results of MR-CT fusion for Case 2 dataset 82

17 5.3.4 MR-T1-T2 Fusion In MR-T1-T2 fused modality, which provides T1 and T2 weighted slices in a single image. The proposed OSMF technique is tested for 270 patients MR- T1 and MR-T2 images. The average values of fusion quality matrices for MR-T1- MR-T2 in comparison with the states of art fusion techniques are given in Table 5.5. The examples of four MR-T1-T2 image fusions among the 270 sets are given in Figure 5.4. The performance analysis shows that the OSMF based fusion has grater performance than that of conventional transform techniques. The PSNR value of conventional FFT is whereas the proposed spectrum masking has The SSIM value of conventional FFT is while the proposed OSMF has The edge quality of OSMF is while the conventional FFT based fusion has The obtained results show that the proposed OSMF has greater amount of fusion in terms of structural mapping, edge quality, contrast enhancement and retained information than that of other existing fusion techniques. Table 5.5 Results of MR-T1-MR-T2 fusion Entropy Q AB F STD SSIM PSNR DCT DWT FFT IHS Proposed

18 MR-T1 MR-T2 MR-T1-MR-T2 fused MR-T1 MR-T2 MR-T1-MR-T2 fused MR-T1 MR-T2 MR-T1-MR-T2 fused MR-T1 MR-T2 MR-T1-MR-T2 fused Figure 5.12 Examples of MR-T1-MR-T2 fusion using proposed technique 84

19 Entropy STD SSIM PSNR 10 0 DCT DWT FFT IHS Proposed Figure 5.13 Results of MR-T1-MR-T2 fusion for Case 1 dataset Entropy STD SSIM PSNR 0 DCT DWT FFT IHS Proposed Figure 5.14 Results of MR-T1-MR-T2 fusion for Case 2 dataset 85

20 5.4 CONCLUSION In this chapter, presented an Optimum Spectrum Mask Fusion (OSMF) for multi modal medical image fusion.the objective of this technique is to provide contrast improved fusion. The comparison results show that the proposed OSMF provides better quantitative and qualitative values in terms of redundant information, mutual information as well as edge quality with various state of art fusion techniques. The optimization of scale values is selected using GWO. The GWO algorithm has performed optimum selection of scale values irrespective of input modalities. The comparative results are analysed for various image fusion modalities and the average value is compared with existing fusion techniques. The proposed OSMF technique reduced the fusion artefacts in MR- SPECT, MR-PET, MR-CT and MR-T1-T2 fusion. In next chapter, cascaded mask based medical image fusion technique is introduced. The cascaded laplacian wavelet mask can dynamically enhance the soft tissues information as well as hard tissues information in single fusion technique itself. 86

Economic Load Dispatch Using Grey Wolf Optimization

Economic Load Dispatch Using Grey Wolf Optimization RESEARCH ARTICLE OPEN ACCESS Economic Load Dispatch Using Grey Wolf Optimization Dr.Sudhir Sharma,Shivani Mehta, Nitish Chopra 3 Associate Professor, Assistant Professor, Student, Master of Technology

More information

Optimal Unified Power Quality Conditioner Allocation in Distribution Systems for Loss Minimization using Grey Wolf Optimization

Optimal Unified Power Quality Conditioner Allocation in Distribution Systems for Loss Minimization using Grey Wolf Optimization RESEARCH ARTICLE OPEN ACCESS Optimal Unified Power Quality Conditioner Allocation in Distribution Systems for Loss Minimization using Grey Wolf Optimization M. Laxmidevi Ramanaiah*, Dr. M. Damodar Reddy**

More information

BIOMEDICAL IMAGE FUSION FOR BRAIN CANCER DETECTION

BIOMEDICAL IMAGE FUSION FOR BRAIN CANCER DETECTION BIOMEDICAL IMAGE FUSION FOR BRAIN CANCER DETECTION Gaurav Lodhwal, Prof. Jyostana.V.Ogale Abstract Biomedical image processing includes biomedical data gathering, Image forming and processing for diagnosis

More information

System identification in the delta domain: a unified approach using FAGWO algorithm

System identification in the delta domain: a unified approach using FAGWO algorithm System identification in the delta domain: a unified approach using FAGWO algorithm Souvik Ganguli 1*, Gagandeep Kaur 1, Prasanta Sarkar 2 1 Department of Electrical & Instrumentation Engineering, Thapar

More information

S. Das *, M. Chowdhury, and M. K. Kundu Machine Intelligence Unit, Indian Statistical Institute, 203 B.T. Road, Kolkata-108, India

S. Das *, M. Chowdhury, and M. K. Kundu Machine Intelligence Unit, Indian Statistical Institute, 203 B.T. Road, Kolkata-108, India Progress In Electromagnetics Research B, Vol. 30, 355 370, 2011 MEDICAL IMAGE FUSION BASED ON RIPPLET TRANSFORM TYPE-I S. Das *, M. Chowdhury, and M. K. Kundu Machine Intelligence Unit, Indian Statistical

More information

ELG7173 Topics in signal Processing II Computational Techniques in Medical Imaging

ELG7173 Topics in signal Processing II Computational Techniques in Medical Imaging ELG7173 Topics in signal Processing II Computational Techniques in Medical Imaging Topic #1: Intro to medical imaging Medical Imaging Classifications n Measurement physics Send Energy into body Send stuff

More information

Multi-scale Geometric Summaries for Similarity-based Upstream S

Multi-scale Geometric Summaries for Similarity-based Upstream S Multi-scale Geometric Summaries for Similarity-based Upstream Sensor Fusion Duke University, ECE / Math 3/6/2019 Overall Goals / Design Choices Leverage multiple, heterogeneous modalities in identification

More information

Wavelet Decomposition in Laplacian Pyramid for Image Fusion

Wavelet Decomposition in Laplacian Pyramid for Image Fusion International Journal of Signal Processing Systems Vol. 4, No., February 06 Wavelet Decomposition in Laplacian Pyramid for Image Fusion I. S. Wahyuni Laboratory Lei, University of Burgundy, Dijon, France

More information

Introduction to SPECT & PET TBMI02 - Medical Image Analysis 2017

Introduction to SPECT & PET TBMI02 - Medical Image Analysis 2017 Introduction to SPECT & PET TBMI02 - Medical Image Analysis 2017 Marcus Ressner, PhD, Medical Radiation Physicist, Linköping University Hospital Content What is Nuclear medicine? Basic principles of Functional

More information

A Generative Perspective on MRFs in Low-Level Vision Supplemental Material

A Generative Perspective on MRFs in Low-Level Vision Supplemental Material A Generative Perspective on MRFs in Low-Level Vision Supplemental Material Uwe Schmidt Qi Gao Stefan Roth Department of Computer Science, TU Darmstadt 1. Derivations 1.1. Sampling the Prior We first rewrite

More information

What is Albira and How Does the System Work and Why is it Differentiated Technology? Page 1

What is Albira and How Does the System Work and Why is it Differentiated Technology? Page 1 What is Albira and How Does the System Work and Why is it Differentiated Technology? Page 1 Why Albira? Tri-modality: PET, SPECT, CT Modular and state-of-the-art electronics 6 configurations Novel, Proprietary

More information

Proyecto final de carrera

Proyecto final de carrera UPC-ETSETB Proyecto final de carrera A comparison of scalar and vector quantization of wavelet decomposed images Author : Albane Delos Adviser: Luis Torres 2 P a g e Table of contents Table of figures...

More information

Sparse Solutions of Linear Systems of Equations and Sparse Modeling of Signals and Images: Final Presentation

Sparse Solutions of Linear Systems of Equations and Sparse Modeling of Signals and Images: Final Presentation Sparse Solutions of Linear Systems of Equations and Sparse Modeling of Signals and Images: Final Presentation Alfredo Nava-Tudela John J. Benedetto, advisor 5/10/11 AMSC 663/664 1 Problem Let A be an n

More information

ScienceDirect. Reactive power control of isolated wind-diesel hybrid power system using grey wolf optimization technique

ScienceDirect. Reactive power control of isolated wind-diesel hybrid power system using grey wolf optimization technique Available online at www.sciencedirect.com ScienceDirect Procedia Computer Science 92 (2016 ) 345 354 2nd International Conference on Intelligent Computing, Communication & Convergence (ICCC-2016) Srikanta

More information

Multimedia communications

Multimedia communications Multimedia communications Comunicazione multimediale G. Menegaz gloria.menegaz@univr.it Prologue Context Context Scale Scale Scale Course overview Goal The course is about wavelets and multiresolution

More information

encoding without prediction) (Server) Quantization: Initial Data 0, 1, 2, Quantized Data 0, 1, 2, 3, 4, 8, 16, 32, 64, 128, 256

encoding without prediction) (Server) Quantization: Initial Data 0, 1, 2, Quantized Data 0, 1, 2, 3, 4, 8, 16, 32, 64, 128, 256 General Models for Compression / Decompression -they apply to symbols data, text, and to image but not video 1. Simplest model (Lossless ( encoding without prediction) (server) Signal Encode Transmit (client)

More information

5 Mutual Information and Channel Capacity

5 Mutual Information and Channel Capacity 5 Mutual Information and Channel Capacity In Section 2, we have seen the use of a quantity called entropy to measure the amount of randomness in a random variable. In this section, we introduce several

More information

IMAGE COMPRESSION-II. Week IX. 03/6/2003 Image Compression-II 1

IMAGE COMPRESSION-II. Week IX. 03/6/2003 Image Compression-II 1 IMAGE COMPRESSION-II Week IX 3/6/23 Image Compression-II 1 IMAGE COMPRESSION Data redundancy Self-information and Entropy Error-free and lossy compression Huffman coding Predictive coding Transform coding

More information

CHAPTER 4 PRINCIPAL COMPONENT ANALYSIS-BASED FUSION

CHAPTER 4 PRINCIPAL COMPONENT ANALYSIS-BASED FUSION 59 CHAPTER 4 PRINCIPAL COMPONENT ANALYSIS-BASED FUSION 4. INTRODUCTION Weighted average-based fusion algorithms are one of the widely used fusion methods for multi-sensor data integration. These methods

More information

Harris s Hawk Multi-Objective Optimizer for Reference Point Problems

Harris s Hawk Multi-Objective Optimizer for Reference Point Problems Int'l Conf. Artificial Intelligence ICAI'16 287 Harris s Hawk Multi-Objective Optimizer for Reference Point Problems A. Sandra DeBruyne 1, B. Devinder Kaur 2 1,2 Electrical Engineering and Computer Science

More information

Digital Image Processing

Digital Image Processing Digital Image Processing, 2nd ed. Digital Image Processing Chapter 7 Wavelets and Multiresolution Processing Dr. Kai Shuang Department of Electronic Engineering China University of Petroleum shuangkai@cup.edu.cn

More information

A Brief Introduction to Medical Imaging. Outline

A Brief Introduction to Medical Imaging. Outline A Brief Introduction to Medical Imaging Outline General Goals Linear Imaging Systems An Example, The Pin Hole Camera Radiations and Their Interactions with Matter Coherent vs. Incoherent Imaging Length

More information

Why is Deep Learning so effective?

Why is Deep Learning so effective? Ma191b Winter 2017 Geometry of Neuroscience The unreasonable effectiveness of deep learning This lecture is based entirely on the paper: Reference: Henry W. Lin and Max Tegmark, Why does deep and cheap

More information

Image Quality and Adaptive Imaging

Image Quality and Adaptive Imaging Image Quality and Adaptive Imaging Matthew A. Kupinski Associate Professor College of Optical Sciences University of Arizona Tucson, Arizona November 7, 2012 Introduction Imaging equation The need for

More information

Application of a GA/Bayesian Filter-Wrapper Feature Selection Method to Classification of Clinical Depression from Speech Data

Application of a GA/Bayesian Filter-Wrapper Feature Selection Method to Classification of Clinical Depression from Speech Data Application of a GA/Bayesian Filter-Wrapper Feature Selection Method to Classification of Clinical Depression from Speech Data Juan Torres 1, Ashraf Saad 2, Elliot Moore 1 1 School of Electrical and Computer

More information

International Electrical Engineering Journal (IEEJ) Vol. 7 (2017) No.9, pp ISSN

International Electrical Engineering Journal (IEEJ) Vol. 7 (2017) No.9, pp ISSN Optimal Number Size and Location of Distributed Generation Units in Radial Distribution Systems Using Grey Wolf Optimizer Abdel-Rahman Sobieh 1 *, M. Mandour*, Ebtisam M. Saied*, and M.M. Salama* *Electrical

More information

Sparse Solutions of Systems of Equations and Sparse Modelling of Signals and Images

Sparse Solutions of Systems of Equations and Sparse Modelling of Signals and Images Sparse Solutions of Systems of Equations and Sparse Modelling of Signals and Images Alfredo Nava-Tudela ant@umd.edu John J. Benedetto Department of Mathematics jjb@umd.edu Abstract In this project we are

More information

Introduction to Information Theory. Uncertainty. Entropy. Surprisal. Joint entropy. Conditional entropy. Mutual information.

Introduction to Information Theory. Uncertainty. Entropy. Surprisal. Joint entropy. Conditional entropy. Mutual information. L65 Dept. of Linguistics, Indiana University Fall 205 Information theory answers two fundamental questions in communication theory: What is the ultimate data compression? What is the transmission rate

More information

Dept. of Linguistics, Indiana University Fall 2015

Dept. of Linguistics, Indiana University Fall 2015 L645 Dept. of Linguistics, Indiana University Fall 2015 1 / 28 Information theory answers two fundamental questions in communication theory: What is the ultimate data compression? What is the transmission

More information

Introduction to Medical Imaging. Medical Imaging

Introduction to Medical Imaging. Medical Imaging Introduction to Medical Imaging BME/EECS 516 Douglas C. Noll Medical Imaging Non-invasive visualization of internal organs, tissue, etc. I typically don t include endoscopy as an imaging modality Image

More information

Department of Electrical Engineering, Polytechnic University, Brooklyn Fall 05 EL DIGITAL IMAGE PROCESSING (I) Final Exam 1/5/06, 1PM-4PM

Department of Electrical Engineering, Polytechnic University, Brooklyn Fall 05 EL DIGITAL IMAGE PROCESSING (I) Final Exam 1/5/06, 1PM-4PM Department of Electrical Engineering, Polytechnic University, Brooklyn Fall 05 EL512 --- DIGITAL IMAGE PROCESSING (I) Y. Wang Final Exam 1/5/06, 1PM-4PM Your Name: ID Number: Closed book. One sheet of

More information

Nuclear Medicine RADIOPHARMACEUTICAL CHEMISTRY

Nuclear Medicine RADIOPHARMACEUTICAL CHEMISTRY Nuclear Medicine RADIOPHARMACEUTICAL CHEMISTRY An alpha particle consists of two protons and two neutrons Common alpha-particle emitters Radon-222 gas in the environment Uranium-234 and -238) in the environment

More information

SHIFT INVARIENT AND EIGEN FEATURE BASED IMAGE FUSION

SHIFT INVARIENT AND EIGEN FEATURE BASED IMAGE FUSION SHIFT INVARIENT AND EIGEN FEATURE BASED IMAGE FUSION S.B.G.Tilak Babu 1, V.Satyanarayana 1 and Ch.Srinivasarao 2 1 Dept. of ECE,AEC, Surampalem,Kakinada, India 2 Dept. of ECE,JNTUK, UCEV,Vizianagaram,

More information

A Riemannian Framework for Denoising Diffusion Tensor Images

A Riemannian Framework for Denoising Diffusion Tensor Images A Riemannian Framework for Denoising Diffusion Tensor Images Manasi Datar No Institute Given Abstract. Diffusion Tensor Imaging (DTI) is a relatively new imaging modality that has been extensively used

More information

Transform Coding. Transform Coding Principle

Transform Coding. Transform Coding Principle Transform Coding Principle of block-wise transform coding Properties of orthonormal transforms Discrete cosine transform (DCT) Bit allocation for transform coefficients Entropy coding of transform coefficients

More information

Multimedia Networking ECE 599

Multimedia Networking ECE 599 Multimedia Networking ECE 599 Prof. Thinh Nguyen School of Electrical Engineering and Computer Science Based on lectures from B. Lee, B. Girod, and A. Mukherjee 1 Outline Digital Signal Representation

More information

Copula based Divergence Measures and their use in Image Registration

Copula based Divergence Measures and their use in Image Registration 7th European Signal Processing Conference (EUSIPCO 009) Glasgow, Scotland, August 4-8, 009 Copula based Divergence Measures and their use in Image Registration T S Durrani and Xuexing Zeng Centre of Excellence

More information

A. I, II, and III B. I C. I and II D. II and III E. I and III

A. I, II, and III B. I C. I and II D. II and III E. I and III BioE 1330 - Review Chapters 7, 8, and 9 (Nuclear Medicine) 9/27/2018 Instructions: On the Answer Sheet, enter your 2-digit ID number (with a leading 0 if needed) in the boxes of the ID section. Fill in

More information

Compression. What. Why. Reduce the amount of information (bits) needed to represent image Video: 720 x 480 res, 30 fps, color

Compression. What. Why. Reduce the amount of information (bits) needed to represent image Video: 720 x 480 res, 30 fps, color Compression What Reduce the amount of information (bits) needed to represent image Video: 720 x 480 res, 30 fps, color Why 720x480x20x3 = 31,104,000 bytes/sec 30x60x120 = 216 Gigabytes for a 2 hour movie

More information

The Use of Locally Weighted Regression for the Data Fusion with Dempster-Shafer Theory

The Use of Locally Weighted Regression for the Data Fusion with Dempster-Shafer Theory The Use of Locally Weighted Regression for the Data Fusion with Dempster-Shafer Theory by Z. Liu, D. S. Forsyth, S. M. Safizadeh, M.Genest, C. Mandache, and A. Fahr Structures, Materials Performance Laboratory,

More information

Introduction to Computer Vision. 2D Linear Systems

Introduction to Computer Vision. 2D Linear Systems Introduction to Computer Vision D Linear Systems Review: Linear Systems We define a system as a unit that converts an input function into an output function Independent variable System operator or Transfer

More information

MEDICAL EQUIPMENT: NUCLEAR MEDICINE. Prof. Yasser Mostafa Kadah

MEDICAL EQUIPMENT: NUCLEAR MEDICINE. Prof. Yasser Mostafa Kadah MEDICAL EQUIPMENT: NUCLEAR MEDICINE Prof. Yasser Mostafa Kadah www.k-space.org Recommended Textbook Introduction to Medical Imaging: Physics, Engineering and Clinical Applications, by Nadine Barrie Smith

More information

Basic Principles of Video Coding

Basic Principles of Video Coding Basic Principles of Video Coding Introduction Categories of Video Coding Schemes Information Theory Overview of Video Coding Techniques Predictive coding Transform coding Quantization Entropy coding Motion

More information

A Novel System for Wind Speed Forecasting Based on Multi-Objective Optimization and Echo State Network

A Novel System for Wind Speed Forecasting Based on Multi-Objective Optimization and Echo State Network Article A Novel System for Wind Speed Forecasting Based on Multi-Objective Optimization and Echo State Network Jianzhou Wang, Chunying Wu * and Tong Niu School of Statistics, Dongbei University of Finance

More information

SIGNAL COMPRESSION. 8. Lossy image compression: Principle of embedding

SIGNAL COMPRESSION. 8. Lossy image compression: Principle of embedding SIGNAL COMPRESSION 8. Lossy image compression: Principle of embedding 8.1 Lossy compression 8.2 Embedded Zerotree Coder 161 8.1 Lossy compression - many degrees of freedom and many viewpoints The fundamental

More information

MRI in Practice. Catherine Westbrook MSc, DCRR, CTC Senior Lecturer Anglia Polytechnic University Cambridge UK. John Talbot MSc, DCRR

MRI in Practice. Catherine Westbrook MSc, DCRR, CTC Senior Lecturer Anglia Polytechnic University Cambridge UK. John Talbot MSc, DCRR MRI in Practice Third edition Catherine Westbrook MSc, DCRR, CTC Senior Lecturer Anglia Polytechnic University Cambridge UK and Carolyn Kaut RothRT(R) (MR) (CT) (M) (CV) Fellow SMRT (Section for Magnetic

More information

Transform coding - topics. Principle of block-wise transform coding

Transform coding - topics. Principle of block-wise transform coding Transform coding - topics Principle of block-wise transform coding Properties of orthonormal transforms Discrete cosine transform (DCT) Bit allocation for transform Threshold coding Typical coding artifacts

More information

Digital Image Processing Lectures 25 & 26

Digital Image Processing Lectures 25 & 26 Lectures 25 & 26, Professor Department of Electrical and Computer Engineering Colorado State University Spring 2015 Area 4: Image Encoding and Compression Goal: To exploit the redundancies in the image

More information

arxiv: v1 [cs.ne] 12 Jun 2018

arxiv: v1 [cs.ne] 12 Jun 2018 Using Chaos in Grey Wolf Optimizer and Application to Prime Factorization Harshit Mehrotra and Dr. Saibal K. Pal arxiv:1806.04419v1 [cs.ne] 12 Jun 2018 Abstract The Grey Wolf Optimizer (GWO) is a swarm

More information

Inverse Problems in Image Processing

Inverse Problems in Image Processing H D Inverse Problems in Image Processing Ramesh Neelamani (Neelsh) Committee: Profs. R. Baraniuk, R. Nowak, M. Orchard, S. Cox June 2003 Inverse Problems Data estimation from inadequate/noisy observations

More information

Time Evolution of ECoG Network Connectivity in Patients with Refractory Epilepsy

Time Evolution of ECoG Network Connectivity in Patients with Refractory Epilepsy Grand Valley State University ScholarWorks@GVSU Masters Theses Graduate Research and Creative Practice 8-2018 Time Evolution of ECoG Network Connectivity in Patients with Refractory Epilepsy Michael J.

More information

Contents. Acknowledgments

Contents. Acknowledgments Table of Preface Acknowledgments Notation page xii xx xxi 1 Signals and systems 1 1.1 Continuous and discrete signals 1 1.2 Unit step and nascent delta functions 4 1.3 Relationship between complex exponentials

More information

Technical University of Denmark

Technical University of Denmark Technical University of Denmark Page 1 of 11 pages Written test, 9 December 2010 Course name: Introduction to medical imaging Course no. 31540 Aids allowed: none. "Weighting": All problems weight equally.

More information

EL-GY 6813/BE-GY 6203 Medical Imaging, Fall 2016 Final Exam

EL-GY 6813/BE-GY 6203 Medical Imaging, Fall 2016 Final Exam EL-GY 6813/BE-GY 6203 Medical Imaging, Fall 2016 Final Exam (closed book, 1 sheets of notes double sided allowed, no calculator or other electronic devices allowed) 1. Ultrasound Physics (15 pt) A) (9

More information

Automated Segmentation of Low Light Level Imagery using Poisson MAP- MRF Labelling

Automated Segmentation of Low Light Level Imagery using Poisson MAP- MRF Labelling Automated Segmentation of Low Light Level Imagery using Poisson MAP- MRF Labelling Abstract An automated unsupervised technique, based upon a Bayesian framework, for the segmentation of low light level

More information

Velocity Images. Phase Contrast Technique. G. Reiter 1,2, U. Reiter 1, R. Rienmüller 1

Velocity Images. Phase Contrast Technique. G. Reiter 1,2, U. Reiter 1, R. Rienmüller 1 Velocity Images - the MR Phase Contrast Technique G. Reiter 1,2, U. Reiter 1, R. Rienmüller 1 SSIP 2004 12 th Summer School in Image Processing, Graz, Austria 1 Interdisciplinary Cardiac Imaging Center,

More information

Unit 6 Nuclear Radiation Parent Guide. What is radioactivity and why are things radioactive?

Unit 6 Nuclear Radiation Parent Guide. What is radioactivity and why are things radioactive? Unit 6 Nuclear Radiation Parent Guide What is radioactivity and why are things radioactive? The nucleus of an atom is comprised of subatomic particles called protons and neutrons. Protons have a positive

More information

Independent Component Analysis on the Basis of Helmholtz Machine

Independent Component Analysis on the Basis of Helmholtz Machine Independent Component Analysis on the Basis of Helmholtz Machine Masashi OHATA *1 ohatama@bmc.riken.go.jp Toshiharu MUKAI *1 tosh@bmc.riken.go.jp Kiyotoshi MATSUOKA *2 matsuoka@brain.kyutech.ac.jp *1 Biologically

More information

Over-enhancement Reduction in Local Histogram Equalization using its Degrees of Freedom. Alireza Avanaki

Over-enhancement Reduction in Local Histogram Equalization using its Degrees of Freedom. Alireza Avanaki Over-enhancement Reduction in Local Histogram Equalization using its Degrees of Freedom Alireza Avanaki ABSTRACT A well-known issue of local (adaptive) histogram equalization (LHE) is over-enhancement

More information

Image fusion based on bilateral sharpness criterion in DT-CWT domain

Image fusion based on bilateral sharpness criterion in DT-CWT domain Int. J. Computational Vision and Robotics, Vol. 4, Nos. 1/2, 2014 161 Image fusion based on bilateral sharpness criterion in DT-CWT domain Priya Ranjan Muduli* Department of Electrical Engineering, Indian

More information

Prof. Mohd Zaid Abdullah Room No:

Prof. Mohd Zaid Abdullah Room No: EEE 52/4 Advnced Digital Signal and Image Processing Tuesday, 00-300 hrs, Data Com. Lab. Friday, 0800-000 hrs, Data Com. Lab Prof. Mohd Zaid Abdullah Room No: 5 Email: mza@usm.my www.eng.usm.my Electromagnetic

More information

Compressed Sensing: Extending CLEAN and NNLS

Compressed Sensing: Extending CLEAN and NNLS Compressed Sensing: Extending CLEAN and NNLS Ludwig Schwardt SKA South Africa (KAT Project) Calibration & Imaging Workshop Socorro, NM, USA 31 March 2009 Outline 1 Compressed Sensing (CS) Introduction

More information

Radioisotopes and PET

Radioisotopes and PET Radioisotopes and PET 1 Radioisotopes Elements are defined by their number of protons, but there is some variation in the number of neutrons. Atoms resulting from this variation are called isotopes. Consider

More information

CS 231A Section 1: Linear Algebra & Probability Review. Kevin Tang

CS 231A Section 1: Linear Algebra & Probability Review. Kevin Tang CS 231A Section 1: Linear Algebra & Probability Review Kevin Tang Kevin Tang Section 1-1 9/30/2011 Topics Support Vector Machines Boosting Viola Jones face detector Linear Algebra Review Notation Operations

More information

Principles of MRI EE225E / BIO265. Instructor: Miki Lustig UC Berkeley, EECS

Principles of MRI EE225E / BIO265. Instructor: Miki Lustig UC Berkeley, EECS Principles of MRI EE225E / BIO265 Instructor: Miki Lustig UC Berkeley, EECS Today... Administration http://inst.eecs.berkeley.edu/~ee225e/sp16/ Intro to Medical Imaging and MRI Medical Imaging (Before

More information

CS 231A Section 1: Linear Algebra & Probability Review

CS 231A Section 1: Linear Algebra & Probability Review CS 231A Section 1: Linear Algebra & Probability Review 1 Topics Support Vector Machines Boosting Viola-Jones face detector Linear Algebra Review Notation Operations & Properties Matrix Calculus Probability

More information

Radionuclide Imaging MII Positron Emission Tomography (PET)

Radionuclide Imaging MII Positron Emission Tomography (PET) Radionuclide Imaging MII 3073 Positron Emission Tomography (PET) Positron (β + ) emission Positron is an electron with positive charge. Positron-emitting radionuclides are most commonly produced in cyclotron

More information

AS to BS in Medical Dosimetry. Core Curriculum Course Descriptions

AS to BS in Medical Dosimetry. Core Curriculum Course Descriptions AS to BS in Medical Dosimetry Core Curriculum Course Descriptions MD301 RADIATION DOSIMETRY This course focuses on introducing radiation terminology used in radiation dosimetry. Fundamental dose calculation

More information

Operational modal analysis using forced excitation and input-output autoregressive coefficients

Operational modal analysis using forced excitation and input-output autoregressive coefficients Operational modal analysis using forced excitation and input-output autoregressive coefficients *Kyeong-Taek Park 1) and Marco Torbol 2) 1), 2) School of Urban and Environment Engineering, UNIST, Ulsan,

More information

Deterministic sampling masks and compressed sensing: Compensating for partial image loss at the pixel level

Deterministic sampling masks and compressed sensing: Compensating for partial image loss at the pixel level Deterministic sampling masks and compressed sensing: Compensating for partial image loss at the pixel level Alfredo Nava-Tudela Institute for Physical Science and Technology and Norbert Wiener Center,

More information

Fourier Transforms 1D

Fourier Transforms 1D Fourier Transforms 1D 3D Image Processing Alireza Ghane 1 Overview Recap Intuitions Function representations shift-invariant spaces linear, time-invariant (LTI) systems complex numbers Fourier Transforms

More information

EE67I Multimedia Communication Systems

EE67I Multimedia Communication Systems EE67I Multimedia Communication Systems Lecture 5: LOSSY COMPRESSION In these schemes, we tradeoff error for bitrate leading to distortion. Lossy compression represents a close approximation of an original

More information

Medical Biophysics II. Final exam theoretical questions 2013.

Medical Biophysics II. Final exam theoretical questions 2013. Medical Biophysics II. Final exam theoretical questions 2013. 1. Early atomic models. Rutherford-experiment. Franck-Hertz experiment. Bohr model of atom. 2. Quantum mechanical atomic model. Quantum numbers.

More information

NONRIGID IMAGE REGISTRATION WITH TWO-SIDED SPACE-FRACTIONAL PARTIAL DIFFERENTIAL EQUATIONS

NONRIGID IMAGE REGISTRATION WITH TWO-SIDED SPACE-FRACTIONAL PARTIAL DIFFERENTIAL EQUATIONS NONRIGID IMAGE REGISTRATION WITH TWO-SIDED SPACE-FRACTIONAL PARTIAL DIFFERENTIAL EQUATIONS Clarissa C. Garvey, Nathan D. Cahill, Andrew Melbourne, Christine Tanner, Sebastien Ourselin, and David J. Hawkes

More information

Phase-Correlation Motion Estimation Yi Liang

Phase-Correlation Motion Estimation Yi Liang EE 392J Final Project Abstract Phase-Correlation Motion Estimation Yi Liang yiliang@stanford.edu Phase-correlation motion estimation is studied and implemented in this work, with its performance, efficiency

More information

Neural Networks. Prof. Dr. Rudolf Kruse. Computational Intelligence Group Faculty for Computer Science

Neural Networks. Prof. Dr. Rudolf Kruse. Computational Intelligence Group Faculty for Computer Science Neural Networks Prof. Dr. Rudolf Kruse Computational Intelligence Group Faculty for Computer Science kruse@iws.cs.uni-magdeburg.de Rudolf Kruse Neural Networks 1 Supervised Learning / Support Vector Machines

More information

MSc Project Feature Selection using Information Theoretic Techniques. Adam Pocock

MSc Project Feature Selection using Information Theoretic Techniques. Adam Pocock MSc Project Feature Selection using Information Theoretic Techniques Adam Pocock pococka4@cs.man.ac.uk 15/08/2008 Abstract This document presents a investigation into 3 different areas of feature selection,

More information

Image Degradation Model (Linear/Additive)

Image Degradation Model (Linear/Additive) Image Degradation Model (Linear/Additive),,,,,,,, g x y h x y f x y x y G uv H uv F uv N uv 1 Source of noise Image acquisition (digitization) Image transmission Spatial properties of noise Statistical

More information

Robust Monte Carlo Methods for Sequential Planning and Decision Making

Robust Monte Carlo Methods for Sequential Planning and Decision Making Robust Monte Carlo Methods for Sequential Planning and Decision Making Sue Zheng, Jason Pacheco, & John Fisher Sensing, Learning, & Inference Group Computer Science & Artificial Intelligence Laboratory

More information

IMAGE COMPRESSION IMAGE COMPRESSION-II. Coding Redundancy (contd.) Data Redundancy. Predictive coding. General Model

IMAGE COMPRESSION IMAGE COMPRESSION-II. Coding Redundancy (contd.) Data Redundancy. Predictive coding. General Model IMAGE COMRESSIO IMAGE COMRESSIO-II Data redundancy Self-information and Entropy Error-free and lossy compression Huffman coding redictive coding Transform coding Week IX 3/6/23 Image Compression-II 3/6/23

More information

Communication Theory and Engineering

Communication Theory and Engineering Communication Theory and Engineering Master's Degree in Electronic Engineering Sapienza University of Rome A.A. 018-019 Information theory Practice work 3 Review For any probability distribution, we define

More information

L. Yaroslavsky. Fundamentals of Digital Image Processing. Course

L. Yaroslavsky. Fundamentals of Digital Image Processing. Course L. Yaroslavsky. Fundamentals of Digital Image Processing. Course 0555.330 Lec. 6. Principles of image coding The term image coding or image compression refers to processing image digital data aimed at

More information

RLE = [ ; ], with compression ratio (CR) = 4/8. RLE actually increases the size of the compressed image.

RLE = [ ; ], with compression ratio (CR) = 4/8. RLE actually increases the size of the compressed image. MP/BME 574 Application Solutions. (2 pts) a) From first principles in class, we expect the entropy of the checkerboard image to be since this is the bit depth of the image and the frequency of each value

More information

Medical Imaging. Norbert Schuff, Ph.D. Center for Imaging of Neurodegenerative Diseases

Medical Imaging. Norbert Schuff, Ph.D. Center for Imaging of Neurodegenerative Diseases Uses of Information Theory in Medical Imaging Norbert Schuff, Ph.D. Center for Imaging of Neurodegenerative Diseases Norbert.schuff@ucsf.edu With contributions from Dr. Wang Zhang Medical Imaging Informatics,

More information

Performance evaluation of DWT, SWT and NSCT for fusion of PET and CT Images using different fusion rules.

Performance evaluation of DWT, SWT and NSCT for fusion of PET and CT Images using different fusion rules. Biomedical Research 2016; 27 (1): 12-11 ISSN 0970-98X www.biomedres.info Performance evaluation of DWT, SWT and NSCT for fusion of PET and CT using different fusion rules. KP Indira 1 *, R Rani Hemamalini

More information

ECE472/572 - Lecture 11. Roadmap. Roadmap. Image Compression Fundamentals and Lossless Compression Techniques 11/03/11.

ECE472/572 - Lecture 11. Roadmap. Roadmap. Image Compression Fundamentals and Lossless Compression Techniques 11/03/11. ECE47/57 - Lecture Image Compression Fundamentals and Lossless Compression Techniques /03/ Roadmap Preprocessing low level Image Enhancement Image Restoration Image Segmentation Image Acquisition Image

More information

Midterm Review. Yao Wang Polytechnic University, Brooklyn, NY 11201

Midterm Review. Yao Wang Polytechnic University, Brooklyn, NY 11201 Midterm Review Yao Wang Polytechnic University, Brooklyn, NY 11201 Based on J. L. Prince and J. M. Links, Medical maging Signals and Systems, and lecture notes by Prince. Figures are from the textbook.

More information

Empirical Mean and Variance!

Empirical Mean and Variance! Global Image Properties! Global image properties refer to an image as a whole rather than components. Computation of global image properties is often required for image enhancement, preceding image analysis.!

More information

The family of hierarchical classes models: A state-of-the-art overview. Iven Van Mechelen, Eva Ceulemans, Iwin Leenen, and Jan Schepers

The family of hierarchical classes models: A state-of-the-art overview. Iven Van Mechelen, Eva Ceulemans, Iwin Leenen, and Jan Schepers The family of hierarchical classes models: state-of-the-art overview Iven Van Mechelen, Eva Ceulemans, Iwin Leenen, and Jan Schepers University of Leuven Iven.VanMechelen@psy.kuleuven.be Overview of the

More information

2.3. Clustering or vector quantization 57

2.3. Clustering or vector quantization 57 Multivariate Statistics non-negative matrix factorisation and sparse dictionary learning The PCA decomposition is by construction optimal solution to argmin A R n q,h R q p X AH 2 2 under constraint :

More information

Matrix Decomposition Algorithms for Feature Extraction

Matrix Decomposition Algorithms for Feature Extraction Matrix Decomposition Algorithms for Feature Extraction Derrick Pisani Department of Computer Science and AI, University of Malta Abstract. Clinical decision support software is a delicate system which,

More information

Introduction p. 1 Compression Techniques p. 3 Lossless Compression p. 4 Lossy Compression p. 5 Measures of Performance p. 5 Modeling and Coding p.

Introduction p. 1 Compression Techniques p. 3 Lossless Compression p. 4 Lossy Compression p. 5 Measures of Performance p. 5 Modeling and Coding p. Preface p. xvii Introduction p. 1 Compression Techniques p. 3 Lossless Compression p. 4 Lossy Compression p. 5 Measures of Performance p. 5 Modeling and Coding p. 6 Summary p. 10 Projects and Problems

More information

SINGLE CHANNEL SPEECH MUSIC SEPARATION USING NONNEGATIVE MATRIX FACTORIZATION AND SPECTRAL MASKS. Emad M. Grais and Hakan Erdogan

SINGLE CHANNEL SPEECH MUSIC SEPARATION USING NONNEGATIVE MATRIX FACTORIZATION AND SPECTRAL MASKS. Emad M. Grais and Hakan Erdogan SINGLE CHANNEL SPEECH MUSIC SEPARATION USING NONNEGATIVE MATRIX FACTORIZATION AND SPECTRAL MASKS Emad M. Grais and Hakan Erdogan Faculty of Engineering and Natural Sciences, Sabanci University, Orhanli

More information

Today. Probability and Statistics. Linear Algebra. Calculus. Naïve Bayes Classification. Matrix Multiplication Matrix Inversion

Today. Probability and Statistics. Linear Algebra. Calculus. Naïve Bayes Classification. Matrix Multiplication Matrix Inversion Today Probability and Statistics Naïve Bayes Classification Linear Algebra Matrix Multiplication Matrix Inversion Calculus Vector Calculus Optimization Lagrange Multipliers 1 Classical Artificial Intelligence

More information

Procesamiento de Imágenes y Bioseñales

Procesamiento de Imágenes y Bioseñales Procesamiento de Imágenes y Bioseñales Dr. Víctor Castañeda Agenda Physical basis of X-ray- CT, NMR, Ultrasound, Nuclear Medicine Sensors (cameras, gamma probes, microphone) Computational Tomography (CT)

More information

Research Article A Novel Differential Evolution Invasive Weed Optimization Algorithm for Solving Nonlinear Equations Systems

Research Article A Novel Differential Evolution Invasive Weed Optimization Algorithm for Solving Nonlinear Equations Systems Journal of Applied Mathematics Volume 2013, Article ID 757391, 18 pages http://dx.doi.org/10.1155/2013/757391 Research Article A Novel Differential Evolution Invasive Weed Optimization for Solving Nonlinear

More information

Introduction to the Course and the Techniques. Jeffry R. Alger, PhD Ahmanson-Lovelace Brain Mapping Center Department of Neurology

Introduction to the Course and the Techniques. Jeffry R. Alger, PhD Ahmanson-Lovelace Brain Mapping Center Department of Neurology Introduction to the Course and the Techniques Jeffry R. Alger, PhD Ahmanson-Lovelace Brain Mapping Center Department of Neurology (jralger@ucla.edu) CTSI Neuroimaging April 2013 Rationale for the Course

More information

AQA Physics /7408

AQA Physics /7408 AQA Physics - 7407/7408 Module 10: Medical physics You should be able to demonstrate and show your understanding of: 10.1 Physics of the eye 10.1.1 Physics of vision The eye as an optical refracting system,

More information

Context Adaptive Space Quantization for Image Coding

Context Adaptive Space Quantization for Image Coding Context Adaptive Space Quantization for Image Coding by Jeffrey Erbrecht A thesis presented to the University of Waterloo in fulfillment of the thesis requirement for the degree of Master of Applied Science

More information