Difference between revisions of "Kifu: Go game record (kifu) generator"

Revision as of 20:38, 10 August 2008

Image change detection:

Using vertical and horizontal RGB components sums of a thresholded difference between a reference image and the current one,

it is easy to compute the screen coordinates of a played stone

Geometric projection:

It is necessary to map each point from a reference rectangle to a point in the displayed shape, or vice-versa;

without considering scene, camera or observer parameters.

The only data available are the corner coordinates on the projection plane and the reference rectangle proportions.

Grid intersection detection:

As the camera has not been calibrated, and the corner coordinates are detected from image change,

the projected coordinates will be somewhat approximative.

So it is necessary to detect the nearest grid intersection for each computed grid point.

Using the accumulative algorithm described above on a square portion of the rectifed image

around the computed intersection should be sufficient, but the result is maybe be more exact

doing line detection on the square edges to compute the intersection using line geometry laws.

Using the resulting grid(s) and the real size or proportions of the object(s) in the scene,

maybe it becomes possible to calibrate the camera with Tsai algorithms or others.

Methods for mapping the coordinates:

"2.2 Perspective transformation with two vanishing points" (pages 2 and 3, equations 7 and 10)

Inverse homography and plane image rectification (page 14)

"Inferring Projective Mappings" (page 3)

Links:

irc://irc.freenode.net/#kifu

Photointerpretation and Small Scale Stereoplotting with Digitally Rectified Photographs with Geometrical constraints: http://cipa.icomos.org/fileadmin/papers/potsdam/2001-21-gf01a.pdf

@@ Line 4: / Line 4: @@
 it is easy to compute the screen coordinates of a played stone
 '''Geometric projection:'''
@@ Line 13: / Line 14: @@
 The only data available are the corner coordinates on the projection plane
 and the reference rectangle proportions.
 '''Grid intersection detection:'''
-As the camera has not been calibrated, the projected coordinates are somewhat approximative.
+As the camera has not been calibrated, and the corner coordinates are detected from image change,
+the projected coordinates will be somewhat approximative.
 So it is necessary to detect the nearest grid intersection for each computed grid point.
+Using the accumulative algorithm described above on a square portion of the rectifed image
+around the computed intersection should be sufficient, but the result is maybe be more exact
+doing line detection on the square edges to compute the intersection using line geometry laws.
+Using the resulting grid(s) and the real size or proportions of the object(s) in the scene,
+maybe it becomes possible to calibrate the camera with Tsai algorithms or others.
@@ Line 42: / Line 59: @@
 http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.9.7803
-with related java source code here: http://www.developpez.net/forums/showthread.php?t=591698
+with related java and C source code here: http://www.developpez.net/forums/showthread.php?t=591698