【图像评价】基于Gabor特征的屏幕内容图像质量评估模型matlab代码
1 简介
本文公开了一种基于Gabor特征互信息的全参考图像质量评价方法,属于图像处理领域.本发明方法具体实施包括如下步骤:(一)输入参考图像和失真图像;(二)对参考图像和失真图像,分别利用二维Gabor滤波器进行滤波以提取Gabor特征;(三)分别建立概率分布函数,求出滤波后图像每一个像素点的概率分布及联合概率分布;(四)求出参考图像和失真图像Gabor特征的互信息;(五)建立图像质量评价函数MIGF评价图像质量.本文利用二维Gabor滤波器提取视觉特征,并通过参考图像和失真图像的互信息来建立图像质量评价函数,评价结果符合人类视觉主观认识.
2 部分代码
%% Demo
%----------------------------------------------------------------------
%
% This is an implementation of the algorithm for calculating the
% Gabor features-basedmodel (GFM) index between two images.
%
% Please refer to the following paper
%
% Z. Ni, H. Zeng, L. Ma, J. Hou, J. Chen, and K.-K. Ma, “A Gabor
% Feature-based Quality Assessment Model for the Screen Content Images,”
% IEEE Transactions on Image Processing, 2018.
%----------------------------------------------------------------------
clc
clear
close
origImg = imread('.\SCI07.bmp');
distImg = imread('.\SCI07_2_4.bmp');
fprintf('Image load finished. \n')
[gfm, gfm_map] = GFM(origImg, distImg);
fprintf('The score of the distorted image is %.4f \n', gfm)
function [score, SimMatrix] = GFM(Im1,Im2)
% ========================================================================
% GFM Index for SCI, Version 1.0
% Copyright(c) 2017 Zhangkai Ni, Huanqiang Zeng, Lin Ma, Junhui Hou,
% Jing Chen, and Kai-Kuang Ma
% All Rights Reserved.
%
% ----------------------------------------------------------------------
% Permission to use, copy, or modify this software and its documentation
% for educational and research purposes only and without fee is here
% granted, provided that this copyright notice and the original authors'
% names appear on all copies and supporting documentation. This program
% shall not be used, rewritten, or adapted as the basis of a commercial
% software or hardware product without first obtaining permission of the
% authors. The authors make no representations about the suitability of
% this software for any purpose. It is provided "as is" without express
% or implied warranty.
%----------------------------------------------------------------------
%
% This is an implementation of the algorithm for calculating the
% Gabor features-basedmodel (GFM) index between two images.
%
% Please refer to the following paper
%
% Z. Ni, H. Zeng, L. Ma, J. Hou, J. Chen, and K.-K. Ma, “A Gabor
% Feature-based Quality Assessment Model for the Screen Content Images,”
% IEEE Transactions on Image Processing, 2018.
%
%----------------------------------------------------------------------
%
%Input : (1) Im1: the first image being compared, which is a RGB image
% (2) Im2: the second image being compared, which is a RGB image
%
%Output: (1) score: is the similarty score calculated using GFM algorithm.
% GFM considers the luminance component and chrominance component of images.
%
% (2) SimMatrix: is the local quality map of the distorted image
%
%-----------------------------------------------------------------------
%
%Usage:
%Given 2 test images img1 and img2. For gray-scale images, their dynamic range should be 0-255.
%For colorful images, the dynamic range of each color channel should be 0-255.
%
%[score, SimMatrix] = GFM(img1, img2);
%-----------------------------------------------------------------------
%%%%% Transform into an opponent color space
L1 = 0.06 * double(Im1(:,:,1)) + 0.63 * double(Im1(:,:,2)) + 0.27 * double(Im1(:,:,3));
L2 = 0.06 * double(Im2(:,:,1)) + 0.63 * double(Im2(:,:,2)) + 0.27 * double(Im2(:,:,3));
M1 = 0.30 * double(Im1(:,:,1)) + 0.04 * double(Im1(:,:,2)) - 0.35 * double(Im1(:,:,3));
M2 = 0.30 * double(Im2(:,:,1)) + 0.04 * double(Im2(:,:,2)) - 0.35 * double(Im2(:,:,3));
N1 = 0.34 * double(Im1(:,:,1)) - 0.60 * double(Im1(:,:,2)) + 0.17 * double(Im1(:,:,3));
N2 = 0.34 * double(Im2(:,:,1)) - 0.60 * double(Im2(:,:,2)) + 0.17 * double(Im2(:,:,3));
%%%%%%%%%%%%%%%%
To = 330;
Tc = 100;
lambda = 0.04;
3 仿真结果
4 参考文献
[1]丁勇等. "基于Gabor特征互信息的全参考图像质量评价方法." CN, CN102497576 B.