Image is an essential form of information representation and communication in modern society. Nowadays, billions of images are generated daily in various applications ranging from photography, entertainment, education, and defense to medicine. Digital image processing provides fundamental modeling and computing solutions for image capture, filtering, segmentation, quantization, resampling, super-resolution, and compression solutions.
In this introductory course on digital image processing, we cover the basics in both theory and practice of image sampling, quantization, filtering, and deep convolution and their applications in segmentation, super-resolution, classification, and compression.
Upon completion of the course, students should be able to understand the current state of the art in the image processing area, have practical programming and system skills to address these problems, and have the basis for future research exploration and industry careers in this area.
Probability and Random Process, Signal and Systems, Digital Signal Processing
Dr. Tong Chen, firstname.lastname@example.org, Room 311, 电子楼
Jiaxin Li, email@example.com, Room 311, 电子楼
Tues. 9-11, 仙 II 303
Every Tues Before and After Class (or By appointment)
R. C. Gonzalez, R. E. Woods, Digital Image Processing, 4th Ed.
Midterm: 40%, Final: 40%, Assignments: 20%
Syllabus and Course Material:
Week #1 数字图像处理简介 Introduction to the digital image processing
Week #2 图像格式 - 色彩空间 Image Formation - Color Space [pdf]
点操作和量化 Point Operations & Quantization [pdf]
Digital Image Processing Tutorial [pdf]
1) deriving the mean, var, and histogram of a given image (e.g., Lena)
2) histogram quantization of a given image with uniform quantization and distribution-aware quantization (4 bins and 8 bins with MSE report)
Week #3 线性滤波 Linear Filtering [pdf]
傅立叶变换 Fourier Transform [pdf]
频域滤波 Frequency-Domain Filtering [pdf]
Week #6 Image Compression [pdf]
Week #7 Neural Image Compression [pdf]