PSNR计算

purpose_857

im1=imread('lena2.bmp');
  H% D6 V* x- ]8 h%im1=im2double(im1);
- N, n  D7 @' Y) u+ w8 sim2=imread('lena_rbhv.bmp');
%im2=im2double(im2);
2 [0 Z. S+ ?! e5 A  C- l/ b3 \% u   
if (size(im1))~=(size(im2))
    error('错误：两个输入图象的大小不一致');
end
9 |' J% R2 }6 y7 |6 D9 ^
4 I  b" m( u9 z! L; R: o0 p7 R7 a    [m,n] = size(im1);
    A = double(im1);
, h1 A3 e" k8 O- B% x; N    B = double(im2);
    D = sum( sum( (A-B).^2 ) );%||A-B||^2
- U. ~5 n4 z6 y3 f8 d; G( T    MSE = D / (m * n);
if  D == 0
    error('两幅图像完全一样');
/ n0 D5 u, o, j2 _    PSNR = 200;
else
    PSNR = 10*log10( (255^2) / MSE );                                                        
end


3 E8 k5 p$ I0 c# j从网上COPY了下 稍微改了下  我不知道这一行为什么出错  请指教
2 i4 L2 m3 q* `2 v! C. F; g2 O3 Q1 I??? Subscript indices must either be real positive integers or logicals.
. I8 ~! M# b! X! j
Error in ==> PSNR_ERRO at 14
    D = sum( sum( (A-B).^2 ) );%||A-B||^2

I_believe

D = sum( sum( (A-B).^2 ) );指的是什么，为什么有两从sum？

peerless2021

function main()
clc; close all;
: S5 i1 @0 h- zI=imread('cameraman.tif'); %用的是原图
. D9 b2 j4 p. j  K; pI1=imnoise(I,'gaussian',0,0.01);
figure;
subplot(1, 2, 1); imshow(I); title('原图像');
subplot(1, 2, 2); imshow(I1); title('加噪声图像');
[PSNR, MSE] = psnr(I, I1)
% 图像峰值信噪比PSNR、均方根误差MSE
+ N8 ]3 ]7 T% L2 q6 d1 {, afunction [PSNR, MSE] = psnr(X, Y)
, Q$ z$ U7 ]. Y; d% 计算峰值信噪比PSNR、均方根误差MSE
% 如果输入Y为空，则视为X与其本身来计算PSNR、MSE
. _% ]* g0 N! f% q0 Q9 C
% ~% H' |& O6 T% ]0 [) N  H8 aif nargin<2
* l- W% m( ]( K4 S' n    D = X;
else
    if any(size(X)~=size(Y))
4 V8 `  P" D9 G% B6 s& l! R        error('The input size is not equal to each other!');
9 I2 {5 O# G& J0 z( _    end
6 t/ q3 c  T5 ?3 j) W    D = X-Y;
end
MSE = sum(D(: ).*D(: ))/prod(size(X));
PSNR = 10*log10(255^2/MSE);
function main()
clc; close all;
! ?! T% d. @$ ]9 e0 qI=imread('cameraman.tif'); %用的是原图
I1=imnoise(I,'gaussian',0,0.01);
figure;
! s4 `3 L( L, \$ Asubplot(1, 2, 1); imshow(I); title('原图像');
subplot(1, 2, 2); imshow(I1); title('加噪声图像');
[PSNR, MSE] = psnr(I, I1)
( m" N, R( n/ U% 图像峰值信噪比PSNR、均方根误差MSE
function [PSNR, MSE] = psnr(X, Y)
% 计算峰值信噪比PSNR、均方根误差MSE
% 如果输入Y为空，则视为X与其本身来计算PSNR、MSE
+ K, E4 \5 J$ Y4 V2 @5 S3 e1 ~
( z( V9 \& J+ k* Z6 r; F6 y$ lif nargin<2
$ f" [8 J7 }" H) k9 V& |6 r    D = X;
else
    if any(size(X)~=size(Y))
        error('The input size is not equal to each other!');
" u& ^3 r, b, J    end
    D = X-Y;
7 c: g1 e. O6 `7 D' @end
MSE = sum(D(: ).*D(: ))/prod(size(X));
; F6 N( c, u0 \" VPSNR = 10*log10(255^2/MSE);
用这个算法算下吧，I为原图 I1为加噪图像
3 G+ X0 B- ?4 Y' b, d' y4 H, n. H