【图像增强】基于Step和Polynomial 滤波实现图像增强附matlab代码

1 内容介绍

本文提出的Step过滤和多项式过滤方法的实现  提供对部分曲线结构的增强。

该方法应用于地球物理图像，以识别线性模式存在于考古遗址 [1] 和故障检测 [2] 中的地下建筑结构。该方法有效地结合了旋转​和尺度不变滤波器和像素标记方法，提供对大多数线结构的强大增强和检测分别为二维灰度图像。

2 仿真代码

% I :  given image

% grammes: size, 

% DTheta: angle vector

% DPaxos:  vector of width of main wave

% typeOfFilter : type of step filter in {0,1} : 1 step function, 0 Polynomial filter 

% Y: filter output 

function [Y]= step_pol_Filtering(I,grammes, DTheta,DPaxos,typeOfFilter)  

gabors = length(DTheta);

gabors2 = length(DPaxos);

for j=1:gabors2,

    bw = DPaxos(j);

   

    for i=1:gabors,

        theta = DTheta(i);

        if typeOfFilter == 1,

            gb=mask_fn(grammes,bw,theta,0);

        else

            gb=mask_fn_smooth(grammes,bw,theta,0);

        end

        gb = gb / sum(sum(gb.^2)); %normalazation in energy

        G = conv2(I,gb,'same'); %idio size me I

        if i == 1 && j == 1,

            Y = abs(G);

        else

            Y = max(abs(Y),abs(G));

        end

    end

end

del = 8;

Y(1:del,:) = 0;

Y(:,1:del) = 0;

Y(size(I,1)-[0:del-1],:) = 0;

Y(:,size(I,2)-[0:del-1]) = 0;

i = 1;

if typeOfFilter == 1,

    gb=mask_fn(grammes,bw,pi/4,1);

else

    gb=mask_fn_smooth(grammes,bw,pi/4,1);

end

function gb=mask_fn(grammes,bw,theta,toPlot)

% bw    = bandwidth, (1)

% gamma = aspect ratio, (0.5)

%

% lambda= wave length, (>=2)

% theta = angle in rad, [0 pi)

sz=grammes;

if mod(sz,2)==0, sz=sz+1;end

[x y]=meshgrid(-fix(sz/2):1:fix(sz/2),fix(sz/2):-1:fix(-sz/2));

% Rotation 

x_theta=x*cos(theta)+y*sin(theta);

y_theta=-x*sin(theta)+y*cos(theta);

for i=1:length(x),

    for j=1:length(y),

        d = abs(x_theta(i,j));

        u = x_theta(i,j);

        v = y_theta(i,j);

        gb(i,j) = 0;

        if d < bw,

            gb(i,j) = 1;

        elseif d <= 3*bw

            gb(i,j) = -1;

        end

    end

end

[n1 n2] = find(gb == -1);

[p1 p2] = find(gb == 1);

nv_n1 = -length(p1)/length(n1);

for i=1:length(n1),

    gb(n1(i),n2(i))= nv_n1;

end

gb = gb/sum(sum(gb.^2));

%gb=exp(-.5*(x_theta.^2/sigma_x^2+y_theta.^2/sigma_y^2)).*cos(2*pi/lambda*x_theta+psi);

if toPlot == 1,

 figure;

 imagesc(gb);

 colormap(gray);

 title('Filter');

mean(mean(gb))

end

function gb=mask_fn_smooth(grammes,bw,theta,toPlot)

% bw    = bandwidth, (1)

% grammes = filter size

%

% toPlot= 1 for plotting  

% theta = angle in rad, [0 pi)

sz=grammes;

if mod(sz,2)==0, sz=sz+1;end

[x y]=meshgrid(-fix(sz/2):1:fix(sz/2),fix(sz/2):-1:fix(-sz/2));

% Rotation 

x_theta=x*cos(theta)+y*sin(theta);

y_theta=-x*sin(theta)+y*cos(theta);

    

for i=1:length(x),

    for j=1:length(y),

        d = abs(x_theta(i,j));

        u = x_theta(i,j);

        v = y_theta(i,j);

        gb(i,j) = 0;

        if d < bw,

            gb(i,j) = 1-((d/bw)^2);

        elseif d <= 3*bw

            gb(i,j) = 0.5*(-1+(((d/bw)-2)^2));

        end

    end

end

[u1 v1] = find(gb > 0);

[u2 v2] = find(gb < 0);

m1 = 0;

N1 = length(u1);

for i=1:length(u1),

    m1 = m1+gb(u1(i),v1(i));

end

m1 = m1/N1;

m2 = 0;

N2 = length(u2);

for i=1:length(u2),

    m2 = m2+gb(u2(i),v2(i));

end

m2 = m2/N2;

c = -m1*N1/(m2*N2);

for i=1:length(u2),

    gb(u2(i),v2(i)) = c*gb(u2(i),v2(i));

end

gb = gb/sum(sum(gb.^2));

%gb=exp(-.5*(x_theta.^2/sigma_x^2+y_theta.^2/sigma_y^2)).*cos(2*pi/lambda*x_theta+psi);

if toPlot == 1,

 figure;

 imagesc(gb);

 colormap(gray);

 title('Filter');

figure;

 mesh(gb);

  title('Filter in 3D view');

  

mean(mean(gb))

end

3 运行结果

4 参考文献

[1] C. Panagiotakis, E. Kokinou and A. Sarris, Curvilinear Structure Enhancement and Detection in Geophysical 

Images, IEEE Trans. on Geoscience and Remote Sensing, vol. 49, no. 6, pp. 2040-2048, 2011.

[2] C. Panagiotakis and E. Kokinou, Linear Pattern Detection of Sea Faults via a Topology and Shape Optimization Method, 

IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 8, no. 1, pp. 3-11, 2015. 

博主简介：擅长智能优化算法、神经网络预测、信号处理、元胞自动机、图像处理、路径规划、无人机等多种领域的Matlab仿真，相关matlab代码问题可私信交流。

部分理论引用网络文献，若有侵权联系博主删除。