Low-level feature extraction (including edge detection) - Feature Extraction and Image Processing - page 137

while i<w & n~=-1
i=i+1;
[qx,qy]=NextPixel(p(1),p(2),n);
p=[qx,qy];
Rchain=[Rchain;p];
m=rem(n+4,8);

if(Next(p(2),p(1),1)~=-1 & Next(p(2),p(1),1)~=m)
n=Next(p(2),p(1),1);
elseif Next(p(2),p(1),2)~=m
n=Next(p(2),p(1),2);
else
n=-1;
end
end

%Mean Squares
st2=0; st3=0; st4=0;sxt=0; sxt2=0; syt2=0; syt2=0;
[n,m]=size(Lchain);
for t=1:n
st2=st2+t*t;st3=st3+t^3; st4=st4+t^4;
sxt=sxt+(Lchain(t,1)-x)*t; syt=syt+(Lchain(t,2)-y)*t;
sxt2=sxt2+(Lchain(t,1)-x)*t*t; syt2=syt2+(Lchain(t,2)-y)*t*t;
end
[n,m]=size(Rchain);
for t=1;n
st2=st2+t*t; st3=st3-t^3; st4=st4+t^4;
sxt=sxt-(Rchain(t,1)-x)*t; syt=syt-(Rchain(t,2)-y)*t;
sxt2=sxt2+(Rchain(t,1)-x)*t*t; syt2=syt2+ (Rchain(t,2)-y)*t*t;
end

if((st2*st4-st3*st3)~=0)
bx=(sxt*st4-sxt2*st3)/(st2*st4-st3*st3);
by=(syt*st4-syt2*st3)/(st2*st4-st3*st3);
cx=(st2*sxt2-st3*sxt)/(st2*st4-st3*st3);
cy=(st2*syt2-st3*syt)/(st2*st4-st3*st3);
d=sqrt((bx*bx+ by*by)^3);
if(d~=0)
d=abs(2*(cy*bx-cx*by)/d);
outputimage(y,x)=d;
end
end
end
end

Code 4.17 Curvature by curve fitting

The main problem with this approach is that it depends on the extraction of sequences
of pixels. In some cases it is very difficult to trace a digital curve in the image. This is
because noise in the data can cause edges to be missed or to be found inaccurately. This
problem may be handled by using a robust fitting technique. However, the implementation