respected sir i have code of content based medical image retrieval but this is not working,n when i want to run this code,it give some errors like function gnh= canny( image1,si,sigma,thresh ) | Error: Function definitions are not permitted in th

>> function gnh= canny( image1,si,sigma,thresh ) %% input argument check.

error(NARGINCHK(4,4,nargin)); if(thresh>1) error('Thresh value should be between 0 and 1'); end I=imread(image1.png'); if(size(I,3)==3) I=rgb2gray(I); end figure,imagesc(I),impixelinfo,title('Original Image'),colormap('gray'); lthresh=0.4*thresh; uthresh=thresh; if(uthresh>1.0) uthresh=1; end %% Antialias the image by convolving with gaussian filter

h=fspecial('gaussian',si,sigma); I=im2double(I); I=imfilter(I,h,'conv'); figure,imagesc(I),impixelinfo,title('Original Image after Convolving with gaussian'),colormap('gray'); %% Compute Gaussian derivatives

 x=-si:1:si;
 y=-si:1:si;
gaussx=-(x/(sigma*sigma)).*exp(-(x.*x+y.*y)/(2*sigma*sigma));
gaussy=gaussx';
Ix=imfilter(I,gaussx,'conv');
Iy=imfilter(I,gaussy,'conv');
%%Compute magnitude and orientation of gradient vector.

Mag=sqrt(Ix .^ 2 + Iy .^ 2); Magmax=max(Mag(:)); Mag=Mag/Magmax; figure,imagesc(Mag),impixelinfo,title('Magnitude of gradient Vectors'),colormap('gray'); gnl=zeros(size(I,1),size(I,2)); gnh=zeros(size(I,1),size(I,2)); angle=atand(Iy./Ix); figure,quiver(20*Ix,20*Iy,0),title('Orientation of gradient vectors'),colormap('gray'); %% Apply non maxima suppression and thresholding. The orientation of gradient vecctor is %% resolved into four directions.

for y=2:1:size(I,1)-1 for x=2:1:size(I,2)-1 if(((angle(y,x)>=-22.5)&&(angle(y,x)<22.5))||((angle(y,x)<-157.5))||(angle(y,x)>157.5)) if((Mag(y,x)>Mag(y,x+1)) && (Mag(y,x)>Mag(y,x-1))) if(Mag(y,x)>=lthresh) if(Mag(y,x)>=uthresh) gnh(y,x)=1; else gnl(y,x)=1; end end end end if(((angle(y,x)>=-112.5)&&(angle(y,x)<-67.5))||((angle(y,x)>=67.5)&&(angle(y,x)<112.5))) if((Mag(y,x)>Mag(y+1,x)) && (Mag(y,x)>Mag(y-1,x))) if(Mag(y,x)>=lthresh) if(Mag(y,x)>=uthresh) gnh(y,x)=1; else gnl(y,x)=1; end end end end if(((angle(y,x)>=-67.5)&&(angle(y,x)<-22.5))||((angle(y,x)>=112.5)&&(angle(y,x)<157.5))) if((Mag(y,x)>Mag(y-1,x+1)) && (Mag(y,x)>Mag(y+1,x-1))) if(Mag(y,x)>=lthresh) if(Mag(y,x)>=uthresh) gnh(y,x)=1; else gnl(y,x)=1; end end end end if(((angle(y,x)>=-157.5)&&(angle(y,x)<-112.5))||((angle(y,x)>=22.5)&&(angle(y,x)<67.5))) if((Mag(y,x)>Mag(y+1,x+1)) && (Mag(y,x)>Mag(y-1,x-1))) if(Mag(y,x)>=lthresh) if(Mag(y,x)>=uthresh) gnh(y,x)=1; else gnl(y,x)=1; end end end end end end figure,imagesc(gnh),impixelinfo,title('Image after thresholding'),colormap('gray'); %% Apply Connectivity Analysis.

[row,col]=find(gnh>0); for t=1:1:size(col) if(gnl(row(t)+1,col(t))>0) gnh(row(t)+1,col(t))=1; gnl(row(t)+1,col(t))=0; end if(gnl(row(t),col(t)+1)>0) gnh(row(t),col(t)+1)=1; gnl(row(t),col(t)+1)=0; end if(gnl(row(t)+1,col(t)+1)>0) gnh(row(t)+1,col(t)+1)=1; gnl(row(t)+1,col(t)+1)=0; end if(gnl(row(t)-1,col(t)-1)>0) gnh(row(t)-1,col(t)-1)=1; gnl(row(t)-1,col(t)-1)=0; end if(gnl(row(t)+1,col(t)-1)>0) gnh(row(t)+1,col(t)-1)=1; gnl(row(t)+1,col(t)-1)=0; end if(gnl(row(t)-1,col(t)+1)>0) gnh(row(t)-1,col(t)+1)=1; gnl(row(t)-1,col(t)+1)=0; end if(gnl(row(t)-1,col(t))>0) gnh(row(t)-1,col(t))=1; gnl(row(t)-1,col(t))=0; end if(gnl(row(t),col(t)-1)>0) gnh(row(t),col(t)-1)=1; gnl(row(t),col(t)-1)=0; end end %% Display output Image.

figure,imagesc(gnh),impixelinfo,title('Image after Applying Connectivity Analysis'),colormap('gray'); end function gnh= canny( image1,si,sigma,thresh ) | Error: Function definitions are not permitted in this context.

Answer 1

[Image Analyst]

You can't do this on the command line:

>> function gnh= canny( image1,si,sigma,thresh ) %%input argument check.

If you want to call canny(), and you have assigned values to those inputs already, then you can do this

>> gnh= canny( image1,si,sigma,thresh ) %%input argument check.

without the function keyword. But I think it's best to make a little test script that will assign the values and call the function.