readme.txt

snake例子

 #####                                          #######
#     #  #    #    ##    #    #  ######         #         #   #  ######   ####
#        ##   #   #  #   #   #   #              #          # #   #       #
 #####   # #  #  #    #  ####    #####          #####       #    #####    ####
      #  #  # #  ######  #  #    #              #           #    #            #
#     #  #   ##  #    #  #   #   #              #           #    #       #    #
 #####   #    #  #    #  #    #  ######         #######     #    ######   #### 

			Active Contour Modeling Using Snakes

Program Name : SnakeEyes
Developed by : Brandon Itkowitz ([e-mail])
     Created : November 1998

The enclosed program is an OpenGL / GLUT implementation of a Snake algorithm as described
by Williams and Shah in their paper "A Fast Algorithm for Active Contours and Curvature
Estimation" [1]. The program was originally developed on an SGI O2 and then recompiled under
Windows 98 using Microsoft Visual C++ 6.0. The program under windows differs slightly from
the original SGI version, mostly because of interface issues with GLUT. OpenGL was mainly
used because it preserves portability, although it is not the most optimal graphics library
for just handling pixel buffers.

SnakeEyes was developed for an Image / Video Computing course taught by Stan Sclaroff at
Boston University. A formal writeup for this program can be found on the web at:
http://people.bu.edu/ickey 

This software is freeware. Feel free to copy / distribute it in its
original form along with this Readme file. If you have questions or comments,
please send me email: [e-mail]

--------------------
Quickie User Manual:
--------------------
The purpose of this program is to demonstrate how snakes can be used to seek out regions
of significant contrast in an image. Snakes are best used for latching onto objects in
an image (i.e. objects that stand out from the background of an image). Once a snake has
come to a resting position around an object, the snake can be used as tool for tracking the
motion of the object across multiple frames. This is all much better described in the formal
html writeup for the program.

Basic Usage Tips:

- Use your left mouse button to set snake points. It is preferable to create a rough
  outline of the object, generally larger than the object itself (the snake tends to
  collapse on objects as opposed to expand to fit objects).

- You may adjust a snake point's position at any time by <CTRL>-clicking near the point and
  moving your mouse to the desired location.

- The right mouse button provides a menu for selecting options, such as:
   Wakeup Snake : activate the snake algorithm
    Reset Coeff : reset the snake energy coefficients for all snake points to the default values
Next/Prev Image : if more than one image was passed in on the command-line, you can
		  select it using the next / prev image options. 

- SnakeEyes only supports JPEG images as input
   	
- Adjusting Snake Coefficients
  You can adjust the snake's behavior by controlling its three coefficients
  alpha : continuity
  beta  : curvature
  gamma : image energy bias

  An individual snake point can be adjust by placing your mouse over the point and pressing
  'a' for alpha, 'b' for beta or 'g' for gamma. Without any other modifier keys, the
  nearest snake point to the mouse cursor will increment the respective coefficient. To
  decrement the coefficient, hold down the <CTRL> key while pressing 'a', 'b', or 'g'. 
  To make global changes (i.e. changes to all snake points) hold down the shift key.

-------------
Requirements:
-------------

You need to have Glut and OpenGL to run this demo.

Glut for Windows can be found at

http://trant.sgi.com/opengl/toolkits/glut.html

GLUT is written by Mark J. Kilgard of Silicon Graphics and is a multiplatform
utility toolkit for the OpenGL libraries.

---
References:

[1] Williams, Donna and Shah, Mubarak. "A Fast Algorithm for Active Contours and Curvature
        Estimation", CVGIP: Image Understanding. Vol. 55, No. 1, January 1992. pp. 14-26.