Course Objectives and Targets:
This course is a fundamental course in digital image processing theory. During the course we explore the basic tools for processing and restoring of gray-level images. We discuss the basic aspects of lossless/lossy image compression and the use of transformation in image processing.
1. Introduction to vision and image processing.
2. 2D signals and linear systems.
3. Sampling and reconstruction: uniform sampling, aliasing, general sampling grids.
4. Quantization: scalar, visual perception, color quantization.
5. Image enhancement: Intensity transformations, histogram processing, spatial filtering, smoothing, sharpening.
6. Image restoration and reconstruction: maximum-likelihood, maximum a posteriori.
7. 2D discrete transforms.
8. Wavelets and multiresolution processing.
9. Image compression: fundamental concepts of information theory, image redundancies, error-free compression, lossy compression, watermarking.
10. Introduction to computer vision.
1. Sufficient background in Digital Signal Processing. In particular, we assume knowledge of the following topics:Linear systems. Fourier transforms of continuous and discrete signals. Frequency responses of continuous and discrete systems. Sampling theorem and perfect reconstruction. Discrete Fourier Transform (DFT) and its properties. Cyclic convolution and its relation to DFT.
2. Sufficient background in Probability. In particular, we assume knowledge of the following topics: Conditional probability. Random variables. Transformation of random variables. Expectations and moments. Discrete and continuous distributions.
Vector of random variables and their joint distributions. Conditional distributions and expectations.
Lecturer: Professor Israel Cohen
Homework – 16%:
Throughout the course 6 homework assignments will be given. The homework is to be submitted individually. The homework can be downloaded from the course website.
Attendance is mandatory for at least 10 of the lectures and 10 of the recitations.
Final Exam – 84%:
A grade of 55% at least is required on the exam in order to pass the course.
Contact Hours per Week (4 weeks, total 28 lectures, 16 recitations):
Lecture: 7 Hours
Recitation: 4 Hour
Rafael C. Gonzalez and Richard E. Woods, Digital Image Processing (3rd Edition), Prentice-Hall, New Jersey, 2008.
Rafael C. Gonzalez and Richard E. Woods, Digital Image Processing Using MATLAB (2nd Edition), Prentice-Hall, New Jersey, 2009.
Anil. K. Jain, Fundamentals of Digital Image Processing, Prentice Hall, New Jersey, 1989.
STAT364/664: Information theory
Yihong Wu, Yale University, Fall 2016
Welcome to STAT364/664! This is a graduate-level introduction to the mathematics of Information Theory, with emphasis on the modern aspects of non-asymptotics and information spectrum methods. We will also discuss various connections and applications to statistics and computer science.
The first lecture is Jan 17.
The lecture on Feb 9 has been cancelled due to ❄❄❄.
Midterm: Feb 28 Tuesday in class. You can bring one letter-size (8.5x11’’) sheet of notes. Calculators or other electronic devices are NOT allowed. Here is an excellent idea of creating cheat sheet.
Final: May 8 Monday 2-5pm (according to the assignment scheme), location: HLH17 03. You can bring two letter-size (8.5x11’’) sheets of notes. Calculators or other electronic devices are NOT allowed.
Lectures: Tuesday and Thursday at 1135-1250 in Room 107, 24 Hillhouse.
Office hours: Tuesday 2-4pm, Room 203, 24 Hillhouse.
Teaching fellow : Jason Klusowski
Lecture notes (updated Jun 2016) co-developed with Prof. Yury Polyanskiy at MIT