返回信息流% =================================================================
% Sparse Feature Fidelity (SFF)
% on LIVE database
% Copyright(c) 2013 Hua-wen Chang
% All Rights Reserved.
% ----------------------------------------------------------------------
% Please refer to the following paper
%
% Hua-wen Chang, Hua Yang, Yong Gan, and Ming-hui Wang, "Sparse Feature Fidelity
% for Perceptual Image Quality Assessment", IEEE Transactions on Image Processing,
% vol. 22, no. 10, pp. 4007-4018, October 2013
% ----------------------------------------------------------------------
% =================================================================
function [OB,metrics] = LIVE()
load('LIVE.mat'); % load database information and DMOS data
load('W.mat'); % load the feature detector, W
Score = zeros(779,1);
h = waitbar(0,'Please wait...');
i = 0;
for iPoint = 1:982
if (live_imSTD(iPoint) == 0)
continue;
else
%READ A REFERENCE IMAGE
Ir = imread(['D:\IMDB\databaserelease2\refimgs\' live_imName{iPoint}]);
%READ A DISTORTED IMAGE
if (iPoint > 0)&&(iPoint <= 227) % JPEG2000
Id = imread(['D:\IMDB\databaserelease2\jp2k\img' num2str(iPoint) '.bmp']);
elseif (iPoint > 227)&&(iPoint <= 460) % JPEG
Id = imread(['D:\IMDB\databaserelease2\jpeg\img' num2str(iPoint-227) '.bmp']);
elseif (iPoint > 460)&&(iPoint <= 634) % White Noise
Id = imread(['D:\IMDB\databaserelease2\wn\img' num2str(iPoint-460) '.bmp']);
elseif (iPoint > 634)&&(iPoint <= 808) % Gaussian Blur
Id = imread(['D:\IMDB\databaserelease2\gblur\img' num2str(iPoint-634) '.bmp']);
elseif (iPoint > 808)&&(iPoint <= 982) % Fast Fading
Id = imread(['D:\IMDB\databaserelease2\fastfading\img' num2str(iPoint-808) '.bmp']);
end
i = i+1;
Score(i) = SFF(Ir,Id,W);
end
waitbar(iPoint/982);
end
close(h);
SB = live_imDMOS; % Subjective Score
OB = Score; % Objective Score
metric_1 = corr(SB, OB, 'type', 'pearson'); % Pearson linear correlation coefficient (without mapping)
metric_2 = corr(SB, OB, 'type', 'spearman'); % Spearman rank-order correlation coefficient
metric_3 = corr(SB, OB, 'type', 'kendall'); % Kendall rank-order correlation coefficient
metrics = [metric_1;metric_2;metric_3];
figure,scatter(OB,SB,'*');
这是一条镜像帖。来源：北邮人论坛 / matlab / #10512同步于 2013/11/28
Matlab机器人发帖
发一个图像质量评价主客观相关度测试的程序源码
ahujack52
2013/11/28镜像同步0 回复
订阅后，新回复会通过你的通知中心匿名送达。
0 条回复
暂无回复 · 你可以订阅本帖等待新回复。