matlab-(含教程)基于PSO优化的多光谱图像融合算法matlab仿真

clc; clear; close all; warning off; addpath(genpath(pwd)); load PAN.mat; %% loading the MS image load MS.mat; %% Loading the PAN image MSWV_db = double(MS); PANWV_db = double(PAN); MS_ORG = double(MS); MSWV_US = imresize(MSWV_db, 1/4, 'bicubic'); MSWV_US = imresize(MSWV_US, 4, 'bicubic'); MSWV_DG = MSWV_US; PANWV_DS = imresize(PANWV_db, 1/4, 'bicubic'); PANWV_US = imresize(PANWV_DS, 4, 'bicubic'); R = MSWV_US(:,:,1); G = MSWV_US(:,:,2); B = MSWV_US(:,:,3); NIR = MSWV_US(:,:,4); for i=1:size(MSWV_US,3) bandCoeffs(i) = max(max(MSWV_US(:,:,i))); MSWV_US(:,:,i) = MSWV_US(:,:,i)/bandCoeffs(i); end P = PANWV_DS; panCoeff = max(max(P)); P = P/panCoeff; CostFunction=@(x) ERGAS_Index(x); % Cost Function nVar = 4; % Number of Decision Variables VarSize = [1 nVar]; % Size of Decision Variables Matrix VarMin = 0; % Lower Bound of Variables VarMax = 1; % Upper Bound of Variables MaxIt = 25; % Maximum Number of Iterations nPop = 10; % Population Size (Swarm Size) phi1 = 2.05; phi2 = 2.05; phi = phi1+phi2; chi = 2/(phi-2+sqrt(phi^2-4*phi)); w = chi; % Inertia Weight wdamp = 1; % Inertia Weight Damping Ratio c1 = chi*phi1; % Personal Learning Coefficient c2 = chi*phi2; % Global Learning Coefficient % Velocity Limits VelMax = 0.1*(VarMax-VarMin); VelMin = -VelMax; empty_particle.Position = []; empty_particle.Cost = []; empty_particle.Velocity = []; empty_particle.Best.Position = []; empty_particle.Best.Cost = []; particle = repmat(empty_particle,nPop,1); GlobalBest.Cost = 0; for i = 1:nPop % Initialize Position particle(i).Position = unifrnd(VarMin,VarMax,VarSize); % Initialize Velocity particle(i).Velocity = zeros(VarSize); % Evaluation particle(i).Cost = CostFunction(particle(i).Position); % Update Personal Best particle(i).Best.Position = particle(i).Position; particle(i).Best.Cost = particle(i).Cost; % Update Global Best if particle(i).Best.Cost > GlobalBest.Cost GlobalBest = particle(i).Best; end end BestCost = zeros(MaxIt,1); for it = 1:MaxIt % figure for i=1:nPop % Update Velocity particle(i).Velocity = w*particle(i).Velocity ... +c1*rand(VarSize).*(particle(i).Best.Position-particle(i).Position) ... +c2*rand(VarSize).*(GlobalBest.Position-particle(i).Position); % Apply Velocity Limits particle(i).Velocity = max(particle(i).Velocity,VelMin); particle(i).Velocity = min(particle(i).Velocity,VelMax); % Update Position particle(i).Position = particle(i).Position + particle(i).Velocity; % Velocity Mirror Effect IsOutside=(particle(i).Position<VarMin | particle(i).Position>VarMax); particle(i).Velocity(IsOutside)=-particle(i).Velocity(IsOutside); % Apply Position Limits particle(i).Position = max(particle(i).Position,VarMin); particle(i).Position = min(particle(i).Position,VarMax); particle(i).Cost = CostFunction(particle(i).Position); % Update Personal Best if particle(i).Cost<particle(i).Best.Cost particle(i).Best.Position = particle(i).Position; particle(i).Best.Cost = particle(i).Cost; % Update Global Best if particle(i).Best.Cost < GlobalBest.Cost GlobalBest = particle(i).Best; end end end BestCost(it) = GlobalBest.Cost; disp(['Iteration ' num2str(it) ': Best Cost = ' num2str(BestCost(it))]); w = w*wdamp; end BestSol = GlobalBest; lamda = 10^-9; eps = 10^-10; % PAN weight in edge detector F_P = expEdge(P, lamda,eps); % MS weight in edge detector Red_W = max(max(R)); % Red component R = R/Red_W; F_R = expEdge(R, lamda, eps); Green_W = max(max(G)); % Green component G = G/Green_W; F_G = expEdge(G, lamda,eps); Blue_W = max(max(B)); % Blue component B = B/Blue_W; F_B = expEdge(B, lamda,eps); NIR_W=max(max(NIR)); % NIR component NIR=NIR/NIR_W; F_NIR = expEdge(NIR, lamda,eps); W = impGradDes(MSWV_US,P); % Optimal weights for spectral bands I = W(1)*MSWV_US(:,:,1) + W(2)*MSWV_US(:,:,2) + W(3)*MSWV_US(:,:,3) + W(4)*MSWV_US(:,:,4); P = (P-mean(P(:)))*std(I(:))/std(P(:)) + mean(I(:)); % Histogram matching W_R = 4*R./(R + B + G + NIR); W_B = 4*B./(R + B + G + NIR); W_G = 4*G./(R + B + G + NIR); W_NIR = 4*NIR./(R + B + G + NIR); F_PSO(:,:,1) = MSWV_US(:,:,1) + W_R.*(BestSol.Position(1)*F_P + (1-BestSol.Position(1))*F_R).*(P-I); F_PSO(:,:,2) = MSWV_US(:,:,2) + W_G.*(BestSol.Position(2)*F_P + (1-BestSol.Position(2))*F_G).*(P-I); F_PSO(:,:,3) = MSWV_US(:,:,3) + W_B.*(BestSol.Position(3)*F_P + (1-BestSol.Position(3))*F_B).*(P-I); F_PSO(:,:,4) = MSWV_US(:,:,4) + W_NIR.*(BestSol.Position(4)*F_P + (1-BestSol.Position(4))*F_NIR).*(P-I); for i = 1:size(MS_ORG, 3) F_PSO(:,:,i) = F_PSO(:,:,i)*bandCoeffs(i); end figure, imshow(uint8(MSWV_DG(:,:,1:3)),'border','tight'); figure, imshow(uint8(PANWV_DS),'border','tight'); figure, imshow(uint8(F_PSO(:,:,1:3)),'border','tight'); addpath Objective_Evaluation Methods = {'PSO'}; ERGAS = ERGAS(MS_ORG,F_PSO, 4); SAM = SAM(MS_ORG,F_PSO); RASE = RASE(MS_ORG,F_PSO); RMSE = RMSE(MS_ORG,F_PSO); UIQI = uqi(MS_ORG,F_PSO); CC = CC(MS_ORG,F_PSO); T = table(ERGAS, SAM, RASE, RMSE, UIQI, CC, 'RowNames', Methods)