Removal of PQS, fixes to image links in the READMEs

QualityAssessment/README.md

@@ -9,7 +9,6 @@ Functions
 
  - **iq_measures** - Image Quality Measures - various measures of reconstructed image quality
  - **image_show** - Displays image, with additional parameters
- - **pqs** - Script that runs software based on CIPIC PQS version 1 (Picture Quality Scale) image quality measure. Unzip the contents Pqs.zip into subdirectory .\Pqs to make the binary available to the script. The original  CIPIC PQS version 1 software has been modified so there is support for different block size (switch -b).
 
 
 Examples

QualityAssessment/iq_measures.m

@@ -1,6 +1,6 @@
-function [MSE,PSNR,AD,SC,NK,MD,LMSE,NAE,PQS]=iq_measures(A,B,disp)
+function [MSE,PSNR,AD,SC,NK,MD,LMSE,NAE]=iq_measures(A,B,disp)
 %Image Quality Measures - various measures of reconstructed image quality
-%[MSE,PSNR,AD,SC,NK,MD,LMSE,NAE,PQS] = iq_measures(A,B,disp)
+%[MSE,PSNR,AD,SC,NK,MD,LMSE,NAE] = iq_measures(A,B,disp)
 %
 %Input: 
 % A - array containing the original image or its filename
@@ -11,10 +11,11 @@
 %         if (disp ~= 'disp') results are not displayed
 %
 %Note: 
-% Number of specified outputs will defined what is actually computed.
+% Number of specified outputs specifies what is actually computed.
 % if (nargout == 2) only MSE and PSNR are computed
-% if (nargout == 8) the PQS is not computed
-% if (nargout == 9) all measures are computed
+% if (nargout == 3) MSE, PSNR and AD are computed
+% ...
+% if (nargout == 8) all measures are computed
 %
 %Output: 
 % MSE - Mean Squared Error
@@ -25,10 +26,6 @@
 % MD - Maximum Difference
 % LMSE - Laplacian Mean Squared Error
 % NAE - Normalized Absolute Error
-% PQS - Picture Quality Scale 
-%
-%Uses:
-% pqs.m (for computation of PQS)
 %
 %Example:
 % [MSE,PSNR]=iq_measures(A,B);
@@ -40,10 +37,10 @@
 else
     disp = strcmp(disp,'disp');
 end;
-if isstr(A)
+if ischar(A)
     A=imread(A);
 end;
-if isstr(B)
+if ischar(B)
     B=imread(B);
 end;
 if ~isa(A,'double')
@@ -52,13 +49,9 @@
 if ~isa(B,'double')
     B=double(B);
 end;
- if (nargout <= 2)
-    numout = 1;
-elseif (nargout <= 8)
-    numout = 2;
-else
-    numout = 3;
-end;
+
+numout = nargout;
+
 clrcomp = size(A,3); 
 MSE = zeros(1,clrcomp);
 PSNR = zeros(1,clrcomp);
@@ -68,59 +61,75 @@
 MD = zeros(1,clrcomp);
 LMSE = zeros(1,clrcomp);
 NAE = zeros(1,clrcomp);
-PQS = zeros(1,clrcomp);
 for i=1:size(A,3)
  if (disp && (clrcomp > 1)) fprintf('Component %d\n',i);end;
- [MSE(i),PSNR(i),AD(i),SC(i),NK(i),MD(i),LMSE(i),NAE(i),PQS(i)]=measureQ(A(:,:,i),B(:,:,i),disp,numout);
+ [MSE(i),PSNR(i),AD(i),SC(i),NK(i),MD(i),LMSE(i),NAE(i)]=measureQ(A(:,:,i),B(:,:,i),disp,numout);
 end;
 
-function [MSE,PSNR,AD,SC,NK,MD,LMSE,NAE,PQS]=measureQ(A,B,disp,numout)
-x=size(A,2);
-y=size(A,1);
+function [MSE,PSNR,AD,SC,NK,MD,LMSE,NAE]=measureQ(A,B,disp,numout)
+PSNR=0;
+AD=0;
+SC=0;
+NK=0;
+MD=0;
+LMSE=0;
+NAE=0;
+
 R=A-B;
 Pk=sum(sum(A.^2));
-MSE=sum(sum(R.^2))/(x*y); % MSE
-if disp~=0 fprintf('MSE (Mean Squared Error) = %f\n',MSE);end;
-% PSNR
-if MSE>0 
-    PSNR=10*log10(255^2/MSE); 
-else 
-    PSNR=Inf;
+
+% MSE
+MSE=sum(sum(R.^2))/(size(A,1)*size(A,2)); % MSE
+if (disp~=0) fprintf('MSE (Mean Squared Error) = %f\n',MSE);end;
+
+if (numout > 1)
+ % PSNR
+ if MSE>0 
+     PSNR=10*log10(255^2/MSE); 
+ else 
+     PSNR=Inf;
+ end;
+ if (disp~=0) fprintf('PSNR (Peak Signal / Noise Ratio) = %f dB\n',PSNR);end;
+end;
+
+if (numout > 2)
+ % AD
+ AD=sum(sum(R))/(size(A,1)*size(A,2)); % AD
+ if disp~=0 fprintf('AD (Average Difference) = %f\n',AD);end;
+end;   
+
+if (numout > 3)
+ % SC
+ Bs = sum(sum(B.^2));
+ if (Bs == 0)
+  SC = Inf;
+ else
+  SC=Pk/sum(sum(B.^2)); 
+ end;
+ if disp~=0 fprintf('SC (Structural Content) = %f\n',SC);end;
+end;    
+
+if (numout > 4)
+ % NK
+ NK=sum(sum(A.*B))/Pk; 
+ if disp~=0 fprintf('NK (Normalised Cross-Correlation) = %f\n',NK);end;
+end;  
+
+if (numout > 5)
+ % MD 
+  MD=max(max(abs(R))); % MD
+  if disp~=0 fprintf('MD (Maximum Difference) = %f\n',MD);end;
+end;
+
+if (numout > 6)
+ % LMSE
+ OP=4*del2(A);
+ LMSE=sum(sum((OP-4*del2(B)).^2))/sum(sum(OP.^2));
+ if disp~=0 fprintf('LMSE (Laplacian Mean Squared Error) = %f\n',LMSE);end;
 end;
-if disp~=0 fprintf('PSNR (Peak Signal / Noise Ratio) = %f dB\n',PSNR);end;
 
-if numout>1
-    AD=sum(sum(R))/(x*y); % AD
-    if disp~=0 fprintf('AD (Average Difference) = %f\n',AD);end;
-    Bs = sum(sum(B.^2));
-    if (Bs == 0)
-        SC = Inf;
-    else
-        SC=Pk/sum(sum(B.^2)); % SC
-    end;
-    if disp~=0 fprintf('SC (Structural Content) = %f\n',SC);end;
-    NK=sum(sum(A.*B))/Pk; % NK
-    if disp~=0 fprintf('NK (Normalised Cross-Correlation) = %f\n',NK);end;
-    MD=max(max(abs(R))); % MD
-    if disp~=0 fprintf('MD (Maximum Difference) = %f\n',MD);end;
-    % LMSE
-    OP=4*del2(A);
-    LMSE=sum(sum((OP-4*del2(B)).^2))/sum(sum(OP.^2));
-    if disp~=0 fprintf('LMSE (Laplacian Mean Squared Error) = %f\n',LMSE);end;
-    NAE=sum(sum(abs(R)))/sum(sum(abs(A))); % NAE
-    if disp~=0 fprintf('NAE (Normalised Absolute Error) = %f\n',NAE);end;
-else 
-    AD=0;
-    SC=0;
-    NK=0;
-    MD=0;
-    LMSE=0;
-    NAE=0;
+if (numout > 7)
+ % NAE
+ NAE=sum(sum(abs(R)))/sum(sum(abs(A))); 
+ if disp~=0 fprintf('NAE (Normalised Absolute Error) = %f\n',NAE);end;
 end;
-if (numout>2)&(x==y) 
-    % PQS
-    PQS=pqs(A,B,x);
-    if disp~=0 fprintf('PQS (Picture Quality Scale) = %f\n',PQS);end;
-else 
-    PQS=0;
-end;

Wavelet/README.md

@@ -114,29 +114,29 @@ Analysis and synthesis filter taps, orthogonality test, Perfect Recostruction (P
 
 	>> scaling_fun('CDF_9x7',5,'d','plot');
 
-![9x7 wavelet scaling function](https://github.com/nsprljan/Matlab/raw/master/Wavelet/CDF_9x7_scaling_analysis.png)
+![9x7 wavelet scaling function](https://github.com/nsprljan/ImageCodingResearchTools/raw/master/Wavelet/CDF_9x7_scaling_analysis.png)
 
 ### Wavelet filter taps ###
 Low pass filter taps, (scaling function with 0 iterations) with **scaling\_fun**:
 
 	>> scaling_fun('LeGall_5x3',0,'d');
 
-![5x3 wavelet h0 taps](https://github.com/nsprljan/Matlab/raw/master/Wavelet/LeGall_5x3_h0.png)	
+![5x3 wavelet h0 taps](https://github.com/nsprljan/ImageCodingResearchTools/raw/master/Wavelet/LeGall_5x3_h0.png)	
 
 ### Frequency and phase characteristic ###
 With **wavelet\_char**:
 
 	>> wavelet_char('CDF_9x7','CDF_9x7','d','plot');  
 
-![9x7 wavelet frequency characteristic](https://github.com/nsprljan/Matlab/raw/master/Wavelet/wavelet_9x7_char_freq.png)	
-![9x7 wavelet phase characteristic](https://github.com/nsprljan/Matlab/raw/master/Wavelet/wavelet_9x7_char_phase.png)
+![9x7 wavelet frequency characteristic](https://github.com/nsprljan/ImageCodingResearchTools/raw/master/Wavelet/wavelet_9x7_char_freq.png)	
+![9x7 wavelet phase characteristic](https://github.com/nsprljan/ImageCodingResearchTools/raw/master/Wavelet/wavelet_9x7_char_phase.png)
 
 ### 2D wavelet ###
 As a tensor product of 1D wavelets, with **wavelet2D**:
 
 	>> wavelet2D('LeGall_5x3',5,'d','l','l');
 
-![2D wavelet](https://github.com/nsprljan/Matlab/raw/master/Wavelet/2Dwavelet.png)		
+![2D wavelet](https://github.com/nsprljan/ImageCodingResearchTools/raw/master/Wavelet/2Dwavelet.png)		
 
 ### Wavelet packet decomposition ###
 
@@ -147,10 +147,10 @@ As a tensor product of 1D wavelets, with **wavelet2D**:
 
 Wavelet packet subband tree:
 
-![Wavelet packet subband tree](https://github.com/nsprljan/Matlab/raw/master/Wavelet/wavelet_packet_subband_tree.png)	
+![Wavelet packet subband tree](https://github.com/nsprljan/ImageCodingResearchTools/raw/master/Wavelet/wavelet_packet_subband_tree.png)	
 
 Wavelet packet transform coefficients:
 
-![Wavelet packet transform coefficients](https://github.com/nsprljan/Matlab/raw/master/Wavelet/Lena_transform_coefficients.jpg)
+![Wavelet packet transform coefficients](https://github.com/nsprljan/ImageCodingResearchTools/raw/master/Wavelet/Lena_transform_coefficients.jpg)
 
 

ZerotreeCoding/README.md

@@ -1,7 +1,7 @@
 Zerotree Coding Toolbox
 =======================
 
-Image compression based on wavelets, using zerotrees of wavelet coefficients. Uses [Wavelet](https://github.com/nsprljan/Matlab/tree/master/Wavelet) and [Quality Assessment](https://github.com/nsprljan/Matlab/tree/master/QualityAssessment) toolboxes. The SPIHT binaries required for function spspiht.m can be downloaded from the SPIHT image compression homepage, see below. 
+Image compression based on wavelets, using zerotrees of wavelet coefficients. Uses [Wavelet](https://github.com/nsprljan/ImageCodingResearchTools/tree/master/Wavelet) and [Quality Assessment](https://github.com/nsprljan/ImageCodingResearchTools/tree/master/QualityAssessment) toolboxes. The SPIHT binaries required for function spspiht.m can be downloaded from the SPIHT image compression homepage, see below. 
 
 
 Functions
@@ -21,7 +21,7 @@ Examples
 --------
 SPIHT algorithm flow-chart:
 
-  ![SPIHT algorithm flow-chart](https://github.com/nsprljan/Matlab/raw/master/ZerotreeCoding/SPIHT_flowchart.png)	
+  ![SPIHT algorithm flow-chart](https://github.com/nsprljan/ImageCodingResearchTools/raw/master/ZerotreeCoding/SPIHT_flowchart.png)	
 
 **spiht_wpackets** with dyadic DWT, as in the original SPIHT: