Appeal No. 1997-0888
Application 08/367,683
independent claims 1 and 12, and states that Dvorzsak and
Ghazizadeh teach one half of the double template not
corresponding with a center of the ideal acquisition target.
Although we find that Dvorzsak does not teach using a double
template to locate a single acquisition target, we look to see
whether Ghazizadeh makes up for this deficiency.
Ghazizadeh discloses a technique for character
recognition in which seven sub-regions of the character are
analyzed to form sequences of alternating dark and light bands
that represent one small area of the character. The character
image fits within a rectangular array R of pixels that is
fixed for that character (col. 3, lines 57-59). The technique
is illustrated in figures 1A-1D for the lower case letter "b."
Two horizontal "slices" (a thin sub-array of pixels chosen
from the rectangular array R that represents the image of the
character), as shown in figure 1A, provide dark-and-light
sequences 11S and 12S corresponding to the dark and light
areas of the character at the slice. A vertical slice 13
produces a sequence 13S. "Masks" are applied to the left,
right, top, and bottom of the character to produce
dark-and-light sequences 14S, 15S, 16S, and 17S. The
- 6 -
Page: Previous 1 2 3 4 5 6 7 8 9 10 Next
Last modified: November 3, 2007