COURSES

Undergraduate Course

Digital Image Processing

This course introduces fundamental principles and techniques of Digital Image Processing that includes a fundamental part: image acquisition, display and representation, color space, histogram processing, linear transformation, geometry transformation, morphology process, edge detection; image transformation and feature extraction (KL/PCA), image segmentation, image compression, and a more advanced part: wavelet transformation, etc..

针对计算机系高年级同学开的专业课,在讲解基本图象处理技术的基础上,特别重视编程实践,要求学生不仅在原理和技术上,而且要在程序设计的层次上熟练掌握图象处理技术。

讲解数字图象处理的基本原理和技术,包括基本部分:图象的获取、显示与表示,色彩空间,直方图处理,线性变换,几何变换,形态学处理,边缘检测;图象变换与特征抽取(KL/PCA),图象分割,图象压缩,以及高级部分:小波变换,彩色图象处理,立体视觉,视频编码与压缩等。


Graduate Course

Computer Vision

This course introduces fundamental principles, models, methods and related algorithms of computer vision that include camera geometry model and calibration, low-level visual processing, multiple view geometry, stereopsis, motion estimation and geometric reconstruction, segmentation and clustering, classification and recognition, visual tracking, active shape model, 3D vision, etc..

讲解计算机视觉的基本原理、模型、方法及相关算法,包括摄像机模型与标定,底层视觉处理(图像特征:边缘、角点、轮廓、纹理)、多视角几何,立体视觉(对应点问题),运动分析(光流),运动估计与几何重构(仿射重构,射影重构),分割与聚类,分类与识别,视觉跟踪,活动形状模型,3D视觉等。


Selected Topics in Computer Vision (Image and Vision Computing)

This seminar course aims at introducing some fundamental principles, algorithms and analysis tools in some selected topics of computer vision research that cover related subjects including image processing, pattern recognition and classification and computer vision, etc.. Emphases are paid to an integration approach that combines those related techniques in view of algorithm and computation to practical problems. Methods involved include wavelets, PCA/LDA/ICA, SVMs/Kernel Machine, AdaBoost, ASM&AAM, EM, Kalman filter/EKF, Condensation/Particle filtering/Monte Carlo method. Special research topics include face image processing, biometrics, visual surveillance and stereo vision, panorama, etc..

从若干重要的应用专题研究出发,讲解所涉及的基本原理、算法、工具及最新发展趋势,介绍图象处理、模式识别、计算机视觉的相关内容,包括图象处理、模式识别与分类方法、计算机视觉等内容。重点是从算法和计算的角度综合图象处理、模式识别和计算机视觉所涉及的基本原理和技术,注重培养解决实际问题的能力。
1.色彩模型与色彩分割
2.Gabor滤波与尺度空间理论、小波变换与多分辨率分析
3.Canny边缘检测,SUSAN角点提取
4.图象的几何变换、图象拼图、全景图
5.物体与轮廓跟踪
6.视觉监视与跟踪的关键技术
7.特征抽取PCA(KL)/LDA/ICA
8.统计学习理论与支持向量机方法在模式识别中的应用
9.ASM&AAM, active contour/snake, level set
10.Boosting, Adaboost
11.EM, HMM, Bayesian Network
12.立体视觉与匹配技术
13.生物身份验证技术