Teacher: Prof. Liu Yuncai

1
???????COMPUTER VISION

• Teacher Prof. Liu Yuncai 
• Reference books
• 1.Machine Vision, by Ramerh Jian, et al (Major
  Reference)
• 2.Computer Vision, by Dama H. Ballard, et al.
• 3.Vision, By David Marr.
• 4.Introductory Techniques for 3-D Computer
  Vision, by Emanuele Trucco.
• 5. Computer Vision and Image Processing, by Scatt
  E.Umbaugh.

2
6. Algorithm for Image Processing and Computer
Vision, by J.R.Parker.7. ????,??? ??8.
???????????, ????9 .???????????????, ????
Rule of gradingHomework 10Examination 6
0Project 30Important DeadlinesHand in
project proposal May 11, 2005Hand in project
report July 6, 2005 All written in
Chinese (????) Final Examination July 6, 2005
3
Course 1 Introduction

• 1.What is Computer Vision
• Computer vision aims to recover useful
  information about a (3D) scene from its 2D
  projections (images).
• 3D depth / 3D structure
• 3D motion analysis
• 3D surfaces and orientation
• status of 3D objects, so on
• meaning of the actions of 3D objects

4
Every task is tough since it is to be understood
from flat images!Computer vision is also called
as Machine visionRobot visionImage
understandingScene analysis (in early years)
5
(No Transcript)
6
(No Transcript)
7
????1)  ?? (Introduction)2)  ???? (Image
Filtering)3)  ??????? (Binary Image
Processing)4)  ?? (Region)5)  ???? (Edge
Detection)6)  ???? (Stereo)7)  ????????
(Motion)8)  ?? (Contours)9)  ?? (Texture)
 10)  ????? (Shading)11)  ??? (Optic
Flows)12)  ???? (Calibration)13)  ?????
(Curves and Surfaces)14)  ???? (Dynamic
Vision)15) ???? (Object Recognition)
8
2. Related fields image processing
computer vision pattern recognition
artificial Intelligence neural network
psychophysics
9
3. Perspective Projection
pin-hole camera model

• projection center (i.e., camera) is at the
  origin of scene coordinate system with optic axis
  of camera pointing to z-axis direction of
  coordinates
• origin of image coordinate system is located at
  (0, 0, f) of scene coordinate system (f is focus
  length of the camera), the X- and Y-axes of image
  coordinates are parallel with x- and y- axes of
  scene coordinates, respectively
• camera does not have any lens distortions.

10
z
11

• (1) Image point

12

• Image Line
• 3D line l is projected onto image plane through
  projecting plane to get image line L

Let we use a vector to
represent the image line L of l. Since We
have,
or It is the equation of projecting plane, the
normal is . Thus, we will use the
normal of the projecting plane of 3D line to
denote image plane by.
13
4. Digital Image

• Sampling digitalize in image plane (
  directions)
• say, , , and so on.
• quantization digitalize in image intensity, say
  8 bits (256 levels).
• for an images having only 2 gray levels, we call
  it as binary image.
•  
•  

14
Image of 64 gray level
Binary image