This preview shows pages 1–3. Sign up to view the full content.
A HIghlight Review Topic List for CSE803
Fall, 2007
Ch 1 INTRODUCTION AND APPLICATIONS
1. What are the major issues in computer vision?
Can you give some lowlevel issues and highlevel
issues? Why do you rank them as low or high level?
2. What are the possible application areas of computer vision techniques?
Ch 2 SENSING AND IMAGE FORMATION
2.1 Imaging Geometry
1. Perspective projection model
2. Orthographic projection model
u
=
mx,
v
=
my
when
m
6
= 1, sometimes it is called weak perspective model.
3. Rotation and translation
4. Homogeneous coordinates
5. Single camera calibration:
External parameters.
Internal parameters.
Projection matrix.
Camera
distortion
6. Stereo camera system: Con±guration. Calibration. Stereo triangulation and uncertainties.
2.2 Image Brightness
1. Radiance
2. Image Irradiance
3. Lens
2.3 Color
1. Tristimulus theory, primary colors
2. CIE Color space
3. Color models: RGB, IHS, YIQ. The meanings of IHS components.
2.4 Fourier Transform and Filters
1. Fourier transform: De±nition. Properties.
2. Convolution theorem:
f
(
x
)
*
g
(
x
) corresponds to
F
(
u
)
G
(
u
).
3. Discrete Fourier transform. Discrete convolution.
4. Filtering using convolution. Frequency domain ±ltering. Highpass, lowpass and bandpass.
5. 2D Fourier transform and 2D ±ltering.
6. Sampling.
1
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document2.5 Imaging Devices
1. Still cameras and video cameras (CCD, Vidicon, etc)
2. Digital images: spatial and gray level resolutions
3. Sensing range directly: structured light ranging, spot ranging, ultrasound ranging
Ch 3 EARLY PROCESSING
3.1 Intrinsic images
1. Defnition oF intrinsic images
2. Examples oF intrinsic images. What are intrinsic images and what are not?
3.2 Some basic image operations
1. Template matching (sumoFsquares, correlation, normalized correlation).
2. Interest operator.
3. Histogram and histogram equalization.
3.3 Feature detection
1. Gradient edge detectors.
2. Edge detection with Facet model.
3. LaplacianoFGaussian operator and zerocrossing: Uncertainty relation. Shape oF LaplacianoFGaussian
operator in space domain and Frequency domain. Bandpass property oF LaplacianoFGaussian oper
ator.
The Four Factors in image Formation.
Primal sketch and 21/2 D sketch.
This is the end of the preview. Sign up
to
access the rest of the document.
 Fall '07
 WENG

Click to edit the document details