This course focuses on practical issues of using Java 2D and Java 3D APIs for creating 2D and 3D graphics, virtual models and animations. Using Java 2D/3D for data visualization is discussed.
The course helps students to understand usage of Java 2D and Java 3D APIs and to gain practical skills in creating graphics applications in Java.
1. Introduction. Review of Java 2D API.
2. Graphic primitives.
3. Painting and stroking.
4. Transforming. Compositing. Clipping. Rendering Hints.
5. Text and fonts.
6. Images.
7. Image filtering. Printing.
8. Review of Java 3D API.
9. Scene graph.
10. Graphic primitives. Mathematical classes.
11. Geometry classes.
12. Appearance. Attributes. Material.
13. Textures.
14. Lights. Text3D.
15. Interaction with the user. Behavior.
16. Animations. Alpha and Interpolator classes. Morphing.
1. Lecture notes.
2. J.Knudsen, Java 2D Graphics. O'Reilly, 1999, 339 pp.
3. A.E. Walsh and D. Gehringer, Java 3D API Jump-Start. Prentice Hall, 2002, 245 pp.
4. D.Selman, Java 3D Programming, Manning, 2002, 376 pp.
Java Programming I.
Recommended: Computer Graphics; Human Interface and Virtual Reality.
Home task on Java 2D - 40%
Home task on Java 3D - 50%
Attendance - 10%
http://www.u-aizu.ac.jp/~niki/courses/
If we define a robot at a computer interacting with the real world physically, we use so many robots everyday such as elevators, cleaning robots, and so on. For designing, synthesizing, analyzing robots, knowledge on how to represent robot structure and motion in computers are required, as well as one on sensors, actuators, modeling methods, and planning algorithms.
This course offers the introduction to robotics for graduate students in computer science and engineering major.
The course covers fundamentals on robotics such as mechanics, modeling and planning, and robotic intelligence, as well as discussing on current open issues remaining in information processing. After taking this course, the students are expected to be able to answer to the following questions. How are the robot motion and structure represented? What kind of system is needed for robot control and planning? What kind of intelligence are robots required? And so on.
#1 introduction and overview
#2 robot arms: forward kinematics such as robot representation, frames and coordinate systems, homogeneous transformation, Denavit-Hartenberg method
#3 robot arms: inverse kinematics such as exact solution, numerical solution, Jacobian
#4 robot arms: dynamics such as deriving equations of motion, Lagrange method, Newton-Euler method, simulation methods
#5 mobile robots: kinematics, dynamics, non-holonomic constraints.
#6 robot and world representation such as workspaces, configuration spaces, cell decomposition, graph representation, artificial potential methods,
#7 planning of sequences: graph search methods, dynamic programming, reinforcement learning
#8 probabilistic planning: Markov decision process, partially observed Markov decision process, sensor based navigation such as sonar and laser range finder
#9 sensors and actuators
#10 robotics intelligence: behavior based planning
#11 robotic applications
#12 summary
None. Related documents will be distributed in a class
Introduction to robotics in the undergraduate course
Reports on numerical experiments on forward and inverse kinematics, forward dynamics, learning, and so on.
http://iplab.u-aizu.ac.jp/moodle/
This course is intended to introduce you to the mathematical foundations of the modern control theory. The aim of the course is to allow you to develop new skills and analytical tools required to analyze and design methods for the control of both linear and nonlinear dynamical systems.
The course covers fundamentals on the modern control theory using state vectors and system matrices. In the end of the course, the students will be able to use analytical tools to model and control a given physical system. Specifically, they can discuss how to determine if a given dynamical system is controllable and stabilizable. They can design state feedback controllers to change the evolution of a dynamical system. They can optimize the control system design to minimize the control energy spent or achieve control in minimum time.
#1 Introduction and overview
#2 Motion representation method
- Equations of motions and differential equations
- Derivation of Equations of Motion: Lagrange method, Newton-Euler method
#3 Solution of equations of motion
- Mathematics: vectors, matrices, ranks, eigenvlaues and eigenvectors, eigenvalue decomposition, Jordan decomposition, norms
#4 System representations and solutions
- Solutions of continuous time systems, discrete time systems
- Solutions of time-variant systems, time-invariant systems
- Solutions of homogeneous systems and non-homogeneous systems
#5 Stability
- Linear system
- Lyapunov theorem
#6 Controllability and observability
#7 Observers and Kalman filters
#8 Regulators
#9 Optimal control
#11 Intelligent control
#12 Summary
None. Related documets will be distributed in a class
Related courses:
Undergraduate: "Introduction to robotics"
Graduate: "Advanced robotits"
Reports on numerical experiments on control theory
http://iplab.u-aizu.ac.jp/moodle/
This course provides a broad introduction to pattern recognition, machine learning, and related topics.
Topics include: Linear Models for Regression and Classification; Neural Network; Kernel Methods; Bayesian Decision Theory; Clustering and Classification; Feature extraction from signals and images; Sequence and Signal design and their applications;
Signal Processing and Image Processing for Instrumentation and Communications; Complex Event Processing and its Applications
Complex Event Processing. The course will also discuss some important applications of pattern recognition and machine
learning.
The objectives of this course is to provide fundamental concepts, theories, and algorithms for pattern recognition and machine learning, which are widely used in various kinds of fields.
1. Overview of topics of pattern classification and its applications
Overview of topics of pattern classification and its applications will be presented. The overview of this course will be illustrated.
2. Linear Models for Regression
This lecture introduces the models that fit a linear equation
to observed data between two variables. Least-squares estimation
and related techniques will be discussed.
3. Linear Models for Classification
A linear classifier attempts to make a classification decision
on an object by the value of a liner combination of the object's
characteristics. Discriminative models, such as perceptron,
will be discussed.
4. Neural Network
An artificial neural network is a mathematical model that consists
of an interconnected group of artificial neurons. Supervised
learning, such as back-propagation algorithm, will be
studied.
5. Kernel Methods
Kernel methods refer to a class of pattern analysis algorithms
that map the data into a high dimensional feature space. Algorithms
including support vector machine and principal components analysis
will be introduced.
6. Bayesian Decision Theory
Bayesian decision making with both discrete probabilities
and continuous probabilities will be studied.
7. Clustering and Classification
Classification is to assign objects to classes on the
basis of measurements made on the objects, while
clustering is to group a set of objects in such a way
that objects in the same group are more similar to
each other than to those in other groups. Both
centroid-based clustering and distribution-based
clustering will be introduced.
8. Feature extraction from signals and images
Various kinds of feature extraction from signals and images will be discussed. Several advantages and disadvantages of digital signal processing and digital image processing will be discussed related with feature extraction and pattern recognition.
9-10. Sequence and Signal design and their applications
Sequences and signals which have special properties are used for pattern recognition and its related applications. Various kinds of sequences and signals will be introduced. Furthermore, several sequence constructions and applications of the sequences to communications, instrumentations will be illustrated.
11-12. Signal Processing and Image Processing for Instrumentation
Signal processing, image processing, and various kinds of pattern recognition techniques are user for instrumentation. In these lecture, various kinds of signal processing and image processing for instrumentation are discussed.
13. Signal Processing for Communications
Signal processing is widely used for communications. Various kinds of signal processing schemes, which are used for communications, will be discussed.
14. Complex Event Processing and its Applications
Complex Event Processing is a powerful technique for various kinds of applications. As an application of pattern recognition, complex event processing and its applications will be discussed.
15. Final Review
Final Review will be provided.
Textbooks will be informed before the course will start.
Assignments
Bioinformatics is to implement information technology to the research of molecular biology for the analysis of DNA, RNA, protein, and metabolism. Recent applications have been extended to system biology, drug design, and personalized medicine of cancer therapy. Due to the huge exponentially increasing number of DNA sequence data, it is urgent to train experts and engineers, who are family with the basic knowledge, analysis methods, and software tools of bioinformatics. In this course, students will learn the mathematical and biological basis of bioinformatics, genetic analysis and database search, gene discovery, and applications of informatics.
The goal is to train students to master the mathematical and biological basis of bioinformatics, the basic algorithms for nucleotide and protein sequence analysis, genetic database search and analysis, and the commonly used software and internet tools of bioinformatics.
1. Biological basis: Cell structure and function, DNA, RNA, and protein
2. Basis of probability and statistics: Probability basis, Bayes’ theorem, probability distribution, histogram, regression, correlation coefficient, t test, and etc
3. Basis of Pattern recognition: Linear classification, Bayes classification, principal component analysis, Hidden Markov models and support vector machine
4. Basis of Data mining: Data preprocessing, mining frequent patterns, associations, and correlations, classification and prediction, and cluster analysis
5. Molecular biology database: DNA/Protein database, Genome database, motif-domain database, data retrieval, and data search
6. Sequence and genetic analysis: Pairwise alignment, multiple alignment, and BLAST/PSI-BLAST, FASTA
7. Gene discovery and data analysis: Microarray, cluster analysis
8. Genome analysis and genome medicine: Molecular phylogenetic tree: algorithm and application
9. Protein structure and prediction: 1st~4th Protein structure, PDB data, homologous protein
10. Computational chemistry: Molecular dynamics, force field, computer software, and etc
11. Special lecture by outside specialist
12. System biology and medicine: Application of genome research in genetic diseases: diagnosis and therapy
はじめてのバイオインフォマティクス 編者： 藤博幸 講談社
Handout will be distributed in class.
Probability and statistics
Physics and chemistry
Database and network
Attendance 20%
Homework 40%
Project 40%
東京大学 バイオインフォマティクス集中講義 監修： 高木 利久
バイオインフォマティクス事典 日本バイオインフォマティクス学会編集
日本バイオインフォマティクス学会 （http://www.jsbi.org/）
バイオインフォマティクス技術者認定試験（http://www.jsbi.org/nintei/）
Biosignals cover a wide spectrum in time and frequency domains. Biosignal detection is a procedure by which we can determines the quantity that characterizes the property or state of human biological condition. Detection modalities include applying various engineering principles in instrumentation. This course will provide introductory knowledge on human body and basic concepts of medical instrumentation, and especially enhance some aspects in detecting a variety of biosignals that differ from industrial measurement.
1. To understand the fundamental skill in applying engineering principles to detect various vital signs.
2. To learn basic specialties in biosignal detection which differ from industrial measurement in some aspects.
1. Introduction
2. Direct Pressure
3. Indirect Pressure
4. Direct Blood Flow
5. Indirect Blood Flow
6. Respiration
7. Motion and Force
8. Body Temperature
9. Bioelectric Measurements
10. Biomagnetic Measurements
11. Electrochemical Measurements of Biochemical Quantities
12. Physical Measurements of Biochemical Quantities
13. Chemical Analyzers
14. Bioimaging
15. Unconstrained Monitoring
Biomedical Sensors and Instruments, 2nd edition, Tatsuo Togawa et al., CRC Press, ISBN: 9781420090789, Publication Date: March 22, 2011
1. Physics and chemistry
2. Electricity and electronics
Research report and presentation
http://i-health.u-aizu.ac.jp/IBSD
エンジニア コンピューターサイエンティストとして科学分野、実践としての 訓練のあるコースです。このコースは技術的な要素だけでなく、オリジナル的な素の表現を適切さより も主体的に学べる事を重視しています。他のコースと大きく違うところは、”正解”を誰とでも共有するという事で、 芸術分野、コンピューターミュージックは、オリジナリティーを奨励していま す。
美的な要素だけでなく技術的な要素も関係しています。オーディオエンジニアや音楽家と同様研究者にとっても有用なコースです。このコースは、リーディング課題、ミュージック理論のドリル、ラボでの実践 など創作力豊かな技術的プログラムを重視しています。私達は，作曲家や演奏家がどのようにコンピューター（声の合成、ビジュアル ミュージック）を使っているか、基本的なミュージック学（チューニング、音 程、和音、音階、調和、旋律、調子、記譜法）やミディインターフェース （ミディ,修正、デイジー チェーン, 制御装置、連続楽譜記憶装置、シンセサイザー、 サウンドモジュール、音調生成プログラムとサンプル）を使ったDTM（ディス クトップ ミュージック）の調査しています。ミュージック理論の授業では、ラボセッションでの学生達の作曲，編曲、各自 の曲、試聴をコンピュータミュージックスタジオ（ワークステーションによる キーボード，シンセサイザーや多種なMIDIコントローラと音声ソフトウェア ー）で行う事も折り込まれています。
このコースを選択すると、学生達は音楽理論の理解とDTMの演習を習得し、基本的な音の編集、ミキシング、エフェクトの使い方を学ぶことによって、自分自身で作曲をする事ができるようになります。
* 基本
o コンピューターを使って、作曲者と演奏者による概観
o コンピューターミュージック言語と譜表
o 音楽学
+ 音程, 間隔
+ 和音
+ 音階
+ 調和
+ 旋律
+ 調子
+ 記譜法
* 音楽と音： オーディトリー ディスプレーとコントロール
o シンセサイザー音楽の演奏
o ミディ
+ ゼネラル ミディ
+ デイジー チェーン
+ 制御装置
+ 連続楽譜記憶装置
+ シンセサイザー、サウンドモジュール、音色、ジェネレーター、サンプル
* コンピューターミュージックソフト使用
o 曲の編集とミキシング
o デスクトップミュージック
o 作曲、配列、演奏
o エフェクト処理
o 音と音楽を素材のインターネットによる調査
o 音楽ソフトウェアーの使用
o 作曲と表音法、連鎖
o パッチス
o 総合
* コンピューターミュージック ハードの使用
* Edly's Music Theory for Practical People, Second Edition, Musical EdVentures, 2005. (ISBN 0-9661616-0-2)
* Curtis Roads, コンピュータ音楽?歴史・テクノロジー・アート, 東京電機大学出版局, 2001. (ISBN-10 4501532106, ISBN-13 978-4501532109)
* Various materials prepared by the instructor
数学、物理
(No prior knowledge of music is assumed.)
homework, lab exercises (progressive song composition, culminating in informal end-of-term recital)
中間試験、期末試験があります。
http://www.edly.com/mtfpp.html
http://www.apple.com/ilife/GarageBand
http://www.pgmusic.com/bbmac.htm
http://audacity.sourceforge.net/?lang=en
http://www.f2dd.info/Developer/Applications/Utilities/Speech/Repeat%20After%20Me%20User%27s%20Guide.pdf
This course deals with advanced architectures for synthetic worlds from a unified view of software and hardware. This course first treats with the fundamentals of computer graphics and computer architectures, and then, presents application-specific computer architectures and parallel algorithms for 3D real-time image synthesis. In particular, parallel architectures/algorithms for polygon rendering and ray tracing are discussed in detail.
1. Introduction
2. Fundamentals of 3D Computer Graphics
3. Advanced Rendering Techniques
4. Hardware Elements for Graphics Systems
5. Parallel Polygon Rendering
6. Parallel Ray Tracing
7. Parallel Volume Rendering
* J. D. Foley, A. van Dam, Computer Graphics, 2nd edition, 1995.
* T. Sagishima, T. Nishizawa, and S. Asahara, Parallel Processing for Computer Graphics (in Japanese), Corona Publishing, 1991.
* Handouts
* Selected journal/conference papers
Computer Graphics, Computer Architecture
Attendance, Presentation, Reports
http://web-int.u-aizu.ac.jp/~nisim/vr_arch/
Biomedical imaging has been an essential diagnostic and therapeutic tool in clinical and basic medicine since the invention of X-ray photographer. Current imaging technology include X-ray photographer, X-ray CT, MRI, ultrasonic imaging, nuclear medicine imaging, endoscopic and laparoscopic imaging technology, and etc. Nowadays, the advancement of medicine requires the scientists and engineers to invent novel imaging modalities, improve the imaging quality and speed of current technology, and the software for accurate and quick analysis of medical images. We expect to train our students to obtain the physical and mathematical knowledge of biomedical imaging, understand the characteristics of different imaging technologies, and have the ability to do further research in biomedical image processing and analysis.
We will train our students to master the theoretical basis of biomedical imaging, understand the characteristics and utilities of different imaging technologies, and have some basic abilities to conduct biomedical image processing and analysis.
1. X-ray CT: Basis of physics and mathematics, system and reconstruction algorithms
2. MRI: Physics and chemistry, system and reconstruction algorithms
3. Ultrasonic imaging: Physics, transducer, and A/B/C/D/F/M modes
4. Nuclear medicine and other imaging modalities: PET, SPECT, OCT, EIT, molecular imaging, and etc.
5. Endoscope and laparoscope: Basis of optics, CCD, CMOS, applications in diagnosis and therapies, and recent development
6. Image processing: Artifacts removal, enhancement, transformation, and etc.
7. Image segmentation: Laplacian filter, snake deformation, and region growing
8. Characteristic extraction from medical images: Preprocessing, region of interest, texture analysis, and characteristic extraction
9. Image information retrieval and registration: Retrieval and analysis of shape and texture, and image registration
10. Computer-aided diagnosis: Reviews on statistics, Bayes’ theorem, classification algorithms, cluster analysis, mammography and angiography
11. Special lecture by outside specialist
12. 3D visualization: Automatic and semi-automatic 3D image reconstruction from 2D slices
13. Surgical navigation system: Imaging and image processing technology for surgical navigation system
Mathematics and Physics of Emerging Biomedical Imaging by National Research council (free download from http://www.e-booksdirectory.com/details.php?ebook=3692),
Handout will be distributed in class.
Physics and chemistry
Electricity and electronics
Probability and statistics
Attendance 20%
Homework 40%
Project 40%
はじめての核医学画像処理
http://www.ne.jp/asahi/ma-ku/104216/
C言語で学ぶ医用画像処理 著者：広島国際大学保健医療学部 石田 隆行 編 オーム
社
This course is a practical introduction to the finite element method. It focuses on algorithms of the finite element method for solid mechanics modeling. Mesh generation and visualization issues are considered.
The course helps students to understand main algorithms of the finite element method and to gain practical skills in finite element programming.
1. Introduction.
2. Formulation of finite element equations.
3. Finite element equations for heat transfer.
4. Finite element method for solid mechanics problems.
5. Finite element equations.
6. Assembly of the global equation system.
7. Finite elements. Two-dimensional triangular element.
8. Two-dimensional isoparametric elements.
9. Three-dimensional isoparametric elements.
10. Discretization.
11. Mesh generation.
12. Assembly algorithms. Displacement boundary conditions.
13. Solution of finite element equations.
14. Nonlinear Problems.
15. Visualization of finite element models and results.
16. Contouring on higher-order surfaces.
1. Lecture Notes
2. Gennadiy Nikishkov, Programming Finite Elements in Java. Springer, 2010, 402 pp.
Numerical Analysis, Programming in any language. Computer Graphics course is recommended.
Exercises - 40%
Home task - 40%
Attendance - 20%
http://www.u-aizu.ac.jp/~niki/courses/
The course deals with practical issues of using Java for creating games with 2D and 3D graphics and animation. Usefulness of Java 2D/3D and Jogl for game programming is discussed.
Upon completion of this course, the student should: Know basics of game development; Know commonly used Java-based techniques for the development of games with graphics and animation; Be able to use Java for game programming.
1. Introduction. Java for game programming.
2. Example of 2D game.
3. Movements and collision detection.
4. Java 2D drawing and transformation.
5. Images and sounds.
6. Car parking game.
7. Collision detection. Game panel. Deployment.
8. Game design concepts.
9. Scene graphs and Java 3D.
10. Java 3D: Behaviors, interaction, and animation.
11. JOGL API. Geometry.
12. JOGL: color, lights, materials.
13. Sprites.
14. Character animation.
15. Sprite animation.
16. Game Engines.
1. A.Davison, Killer Game Programming. O’Reilly, 2005, 969 pp.
2. A.Davison, Pro Java 6 3D game Development. O'Reilly, 2007, 498 pp.
Students are assumed to have taken the course that covered basics of programming in Java. Computer Graphics course is recommended.
Home tasks - 90%
Attendance - 10%
http://www.u-aizu.ac.jp/~niki/courses/
In order to determine your research theme related computer vision and image processing, you need to know the latest status of these fields. Actually, image processing needs many technical and conceptual background from computational algorithms such as Monte-Carlo, forests, dynamic programming, belief propagation, statistical analysis and so on.
In the lecture of image processing in the undergraduate course, we learned the concept of digital images and some basic techniques for analyzing image patterns, and this course provides fundamental algorithms how to understand images or patterns and the status which is necessary technically and conceptually to conduct your master/doctor thesis.
We aim to present the fundamental knowledge for reading and writing academic papers related computer vision and image processing.
1. Segmentation - Mean-shift, Sneaks, Watershed
2. Clustering - k-Means, Fuzzy c-Means, Sequential Clustering, Hieralchical Clustering, Learning Vector Quantization
3. Statistical Analysis - Principal Component Analysis, Eigenface, Latent Semantic Indexing
4. Statistical Analysis - Independent Component Analysis, Sparse Component Analysis, Bag of Visual Words
5. Pattern Matching - DTW, Dikstra Algorithm, Longest Common Sub-sequence
6. Pattern Matching - Continuous DP, Time-Space CDP
7. Pattern Matching - 2DCDP for spotting recognition of Images
8. Pattern Matching - Random Forests, Graph Cut
9. Understanding - Bayesian Net
10. Understanding - Hidden Markov Model
11. Understanding - Self Organization Map, Quantification Method IV
12. Understanding - Support Vector Machine
13. Etc. - Perceptron, Back-Propagation
No special textbook. All materials are provided at each class.
Course website: https://sites.google.com/site/uaizuipu2013/
Prerequisites and other related courses which include important concepts relevant to the course:
Image processing and signal processing in the undergraduate school.
(1) Two Reports for the given themes (TSCDP, Bayesian Net)
(2) Brief presentation of the latest status of your master thesis.
https://sites.google.com/site/uaizuipu2013/
This course concerns the method for Document Analysis and Recognition. We will discuss on the advanced techniques of Document Analysis and Recognition and create the new idea based on this research theme. Especially, we focus on the current technologies related on the on-line/off-line recognition, analysis, and its application.
At the completion of this course, students will be able to:
Have the overview of the Document Analysis and Recognition.
Be able to know how can be implemented for the programming of this area.
Introduction to Document Analysis and Recognition(DAR).
Fundamentals on on-line recognition and off-line recognition.
Pattern recognition on DAR.
Current problems and solving methods of this area such as text recognition, handwriting recognition and other applications:
- pen-based interactive system,
- oriental-pen writing/drawing simulation,
- handwritten font generation,
- signature verification, writer identification,
- handwritten gesture recognition using Wii remote controller,
- segmentation free recognition, latin and non-latin character recognition,
- cursive handwritten character and symbols, layout analysis,
- smartphone system, other application system,
- the application and the relation to image recognition and computer vision,
The presentation of some application program.
Students' investigation work:
-Investigation, presentation and discussion about current techniques and producing the new idea.
-Programming related on Document Analysis and Recognition.
Materials collected from books/papers of journal and proceeding which are selected and provided by the instructor.
Permission of the instructors.
Interest in the area of Document Analysis and Recognition.
Investigation and presentation (40%)
Attendance and positive class participation (20%)
Programming project(40%)
References:
A. C. Downton, S. Impedovo, Progress in Handwriting Recognition, World Scientific; ISBN 981-02-3084-2 (Sep. 1996)
S.-W. Lee, Advances in Handwriting Recognition, World Scientific; ISBN 981-02-3715-4 (1999)
T. Pavlidis, Structural Pattern Recognition, Springer-Verlag; ISBN 3-540-08463-0 (1980)
Useful Links:
Course Web Site: http://web-int.u-aizu.ac.jp/~jpshin/GS/DAR.html
The purpose of this course is to study the fundamentals of spatial hearing
and its application to virtual environments. By using two ears, human among
other species, are able to determine the direction from where a sound is being
emitted in a real environment. For virtual environments (e.g., movies, games,
recorded or live concerts) is desirable to provide the spatial cues found in
nature to increase the realism of the scene. Besides reviewing the underlying
theories of spatial hearing, this course focuses in practical implementations
of binaural hearing techniques, so the course is intense in hands-on exercises,
assignments, and projects mainly based on Pure-data programming language
(http://puredata.info).
Students who approve this course are expected to understand the basic
underlying mechanisms of spatial hearing, as well as the literature and
terminology on this topic.
Given some application constraints (real-time, computing power, etc.)
students at the end of the term should be able to decide which of the
presented techniques is best for creating the 3D aural illusion.
Upon completion of this course, students should be able to successfully
implement virtual 3D sound environments based on head-related transfer
functions (HRTF) and multi-speaker systems.
Session 1. Introductions and introduction to PD
Session 2. Quanti cation of sound
Session 3. Spatial hearing
Session 4. Psychoacoustics
Session 5. Temporal cues
Session 6. Intensity cues
Session 7. Head-related impulse response and transfer function
Session 8. Motion perception
Session 9. Presentation of mid-term projects
Session 10. Distance perception
Session 11. Headphone techniques
Session 12. Headphone techniques II
Session 13. Loudspeaker techniques
Session 14. Loudspeaker techniques II
Session 15. Applications
Session 16. Presentation of fnal projects
Durand R. Begault, 3-D Sound for Virtual Reality and Multimedia, Academic Press, 2000.
Various materials prepared by the instructors
Durand R. Begault, 3-D Sound for Virtual Reality and Multimedia, Aca-
demic Press, 2000.
Various materials prepared by the instructors
Exercises and quizzes 10%
Assignments 30%
Mid-term project 30%
Final project 30%
Course website: http://sonic.u-aizu.ac.jp/spatialHearing/
J. Blauert, Spatial Hearing: The Psychophysics of Human Sound Local-
ization. MIT Press, 1997.
Bregman, Albert S., Auditory Scene Analysis: The Perceptual Orga-
nization of sound. Cambridge, Massachusetts: The MIT Press, 1990
(hardcover)/1994 (paperback).
www-crca.ucsd.edu/~msp/Pd_documentation/
http://hyperphysics.phy-astr.gsu.edu/hbase/sound/soucon.html
This course is focused on the use of technology for language teaching. Each student will create a final project that demonstrates the lessons and work involved in this class. Final projects should exemplify authentic language
courses.
Students will:
- Create a customized CMS course that clearly displays its structure.
- Enhance the course with html and/or javascript code.
- Use customized modules, themes, activities, blocks and/or other augmentations to Moodle.
- Indicate in a syllabus the CALL materials that are created by the student
and borrowed from the Internet. Describe why each item is included.
- Organize the materials in an appropriate systematic manner.
- Describe the decision making process that informed the creation of the project.
Weekly project work based on the course objectives.
Reading material will be provided by the instructor.
- Create a customized CMS course that clearly displays its structure. 20%
- Enhance the course with html and/or javascript code. 15%
- Use customized modules, themes, activities, blocks and/or other augmentations to Moodle. 15%
- Indicate in a syllabus the CALL materials that are created by the student
and borrowed from the Internet. Describe why each item is included. 15%
- Organize the materials in an appropriate systematic manner. 15%
- Describe the decision making process that informed the creation of the project. 20%
http://moodle.u-aizu.ac.jp/moodle
This course introduces students to the field of automatic speech recognition (ASR). It gives basic knowledge about speech science, i.e. speech production and perception by humans, digital speech processing techniques and speech models. Some fundamental aspects of the pattern recognition theory are given in order to make clear the specifics of classifier design and main parameter estimation methods. The Hidden Markov model is explained in detail since it is the main tool of current speech recognition technology. The design and training of the main parts of an ASR system, i.e. acoustic and language models is given, and their usage for decoding is explained. The whole process of ASR system design is presented and some more advanced methods such as noise robustness, discriminative training, and Bayesian networks are briefly described
The objective of this course is to make students familiar with the fundamentals of pattern recognition theory and its application to speech recognition in particular, as well as to teach them how to build classifiers starting with feature extraction methods and ending with system evaluation techniques.
1.Course overview. Sounds and human speech systems. Phonetics.
Sound levels, Speech production, Speech perception.
Phonemes, Co-articulation, Syllables and words.
2.Speech signal processing.
Short-time Fourier Analysis, Linear Predictive Coding.
Cepstral processing, Pitch extraction.
3.Pattern classification.
Bayes' decision theory, Classifiers design.
Maximum likelihood estimation, MAP estimation.
Vector quantization, EM algorithm.
4.Hidden Markov model.
Dynamic programming and DTW, Forward algorithm, Viterbi algorithm, Baum-Welch algorithm.
Types of HMM, HMM limitations.
5.Acoustic modeling.
Context independent model units, Context dependent model units.
Training.
6.Language modeling.
Context free grammars, N-grams.
Perplexity.
7.Search algorithms.
Decoder basics.
Viterbi search, Stack search.
8.Environmental robustness.
Channel distortions, Additive noises.
Spectral subtraction, Viener filter.
9.ASR system design.
Data collection and labeling.
Model training, Evaluation.
10.Advanced algorithms.
Discriminative training, HMM/NN.
Bayesian networks, HMM/BN.
Handouts
Basic knowledge about probability theory, distributions and random variables is required. Familiarity with some signal processing techniques, like filters, digital Fourier transform is a plus.
Attendance: 30 points
Laboratory exercises: 30 points
Project: 40 points
Books:
L. Rabiner, B. Juang, Fundamentals Of Speech Recognition, Prentice-Hall, 1993.
X. Huang, A. Acero, H. Hon, Spoken Language Processing: A guide to theory, Algorithm, and System Development, Prentice-Hall, 2001.
S. Theodoridis, K. Koutroumbas, Pattern Recognition, 4th edition, Elsevier, 2009.
本講義では、調音のメカニズムとその計測方法について学習する。また、調音と音響の関連性についても焦点を当てていく。調音については超音波やビデオなどを用いて調べ、音響音声についてはPraat（オープンソースの音響分析ソフトウェア）を使用して調べる。
本講義を通して、学生は以下のことが可能になる。
(1) 人の音声がどのように生成されるか、また、調音の変化が音響音声にどのような影響を与えるかを説明できる。
(2) 音声データ採集のために超音波機器を扱える。
(3) ソフトウェアを使用して超音波画像を処理できる。
(4) 音響音声を分析でき、また、音響データを自動分析する簡単なスクリプトを書ける。
(5) 波形やフォルマント、ＦＦＴ、サイン波による音声合成といった音響の概念を理解できる
第１週：音声の生成過程と調音の計測方法
第２週：音声分類における音響特性
第３週：超音波による音声データの採集と分析
第４週：調音と音響の関係I
第５週：調音と音響の関係II
第６週：視聴覚による言語知覚
第７週：音声の変動性I ― 話者間における違い
第８週：音声の変動性II ― 言語間における違い
ハンドアウトやその他の資料については、講義のウェブサイト上にアップロードする予定。
先修科目は特に無し。
課題やプロジェクトについては後日アナウンスする。
講義のウェブサイトは以下のURLから利用可能。
http://aizuben.u-aizu.ac.jp/moodle/
Biomedical modeling and visualization is an important technology to extract useful information and discover the biomedical mechanisms buried in the huge amount of data produced in the basic biomedical researches and clinical medical practice. This course will introduce how to implement computer information technology in biomedical modeling and visualization. Main lecture contents include computer modeling and simulation of biological cells, organs, and systems, mathematical basis for biomedical modeling and simulation, physiological modeling and simulation, and biomedical visualization. Homework and projects will be assigned based on measured data in Biomedical Information Technology lab and medical database available in the Internet.
This course will help students to obtain the skills and experiences in implementing computer information technology to biomedicine. Through this course, it will strengthen students' R&D ability in future biomedical research and work.
1. Biomedical modeling and visualization: its application in clinical and basic medicine
2. Mathematical basis for biomedical modeling and simulation
3. Cellular level modeling and simulation: Hodgkin-Huxley model
4. Tissue level modeling and simulation: rule-based model and reaction-diffusion model
5. Construction and visualization of biological models with realistic shapes
6. Organic modeling and simulation: whole-heart model
7. Computer simulation of arrhythmias: atrial fibrillation, supraventericular tachycardias, and ventricular fibrillation
8. Physiological modeling and simulation: heart rate variability, and its linear and nonlinear dynamics
9. Topics on other biomedical modeling and simulation: cerebral networks, bioheat transfer, biomechanics, biofluid mechanics, and etc.
10. General-purpose GPU in biomedical modeling and visualization
Handout will be distributed in class.
Digital signal processing
Computer graphics
Biomedical information technology
Image processing
Attendance 20%
Homework 40%
Project an presentation 40%
http://www.physiome.jp/
http://www.physiome.org.nz/
http://www.nlm.nih.gov/
http://ecg.mit.edu/
http://www.u-aizu.ac.jp/~zhuxin/course
Biosignal enhancement, feature extraction and physiological interpretation are important aspects in biomedical engineering field. Various biosignals can be manipulated through proper representation, transformation, visualization and optimization. This course will introduce fundamental concepts and approaches, such as filtering, detection, estimation, and classification, in various biosignal processing and data mining.
1. To provide a clear picture of biosignal from detection to application by following the course "Introduction to Biosignal Detection?".
2. To provide advanced knowledge for students who are considering pursuing further study in biomedical engineering field.
1. Introduction
2. Biosignal measurement
3. Signal separation
4. Event detection
5. Data preprocessing
6. Time domain analysis
7. Frequency domain analysis
8. Chaotic analysis - 1
9. Chaotic analysis - 2
10. Envelope detection
11. Model estimation and predication
12. Trend and cycle
13. Detection of change
14. Classification
15. Clustering
Practical Biomedical Signal Analysis Using MATLAB (Series in Medical Physics and Biomedical Engineering)
Katarzyn J. Blinowska and Jaroslaw Zygierewicz, CRC Press; 1 edition (September 12, 2011), ISBN-10: 1439812020, ISBN-13: 978-1439812020
Introduction to Biosignal Detection
Probability and Statistics
Discrete Mathematics and Linear Algebra
Digital Signal Processing
Research report and presentation
http://i-health.u-aizu.ac.jp/BPDM/index.html