Question

我应该设计并实现一个自适应滤波器，以消除医学图像中的脉冲噪声！我是图像处理的新手。并且不知道如何设计过滤器！我检查了预定义的过滤器......它们不是我想要的！请帮助这是我的B.S.项目！

Answer 1

通常使用median滤镜处理脉冲噪音为了构建自适应滤波器，我将使用统计数据来确定窗口内是否有某些东西要平滑。

我会使用windows处理图像在每个窗口中，我都会检查中位数和平均值如果它们彼此相距很远，我会应用median过滤器，否则，应用本地LPF过滤器什么都不做。

很简单......

Answer 2

我找到了一篇关于它的论文，这是我实施的代码

function adaptive()
    I = imread('1.png');
    x = rgb2gray(I);
    %%-------adding Noise-----------------------
    disp('Noise density lies between 0 and 1');
    disp(' ');
    ND = input('Enter Noise Density [0.5] : ');
    if isempty(ND)
        ND = 0.5;
    end
    y=x;
    Narr = rand(size(y));
    N = Narr;
    N(N>=ND)=0;
    N1 = N;
    N1 = N1(N1>0);
    Imn=min(N1(:));
    Imx=max(N1(:));
    N=(((N-Imn).*(255-0))./(Imx-Imn));
    y(Narr<ND) = N(Narr<ND);
    y=double(y);
    %%------noise detection------------------------
    adj = imadjust(y);
    L = max(adj);
    [M,N] = size(y);
    for i = 1:M
        for j = 1:N
            if y(i,j) == L(j)-1 || y(i,j) == 0 
                alpha(i,j) = 1;     %%the value 1 presents the “noise pixel” 
            else
                alpha(i,j) = 0;     %%the value 0 presents the “noise-free pixel”.
            end
         end
    end
    K=0;
    for i = 1:M
        for j = 1:N
            if alpha(i,j) == 1; 
                K=K+1;      %%total number of the “noise pixel”, K.
            end
        end
    end
    n = K / (M*N);         %%the ratio of the “noise pixels” to 
                            %%the total number of pixels contained in the image

    %%------noise elimination------------------------
    %%R=1;
    R = floor(0.5 * sqrt( 7 / (1-n) ));
     m = ones(size(y(:)));   
    for i = 1:M
        for j = 1:N
            if alpha(i,j) == 1 %% this is a noise pixel
                nnum=0;     %%noise number
                while nnum <= 8
                    R = R+1;  %% This will make the filter window 2pixels larger
                    clear tmp;
                    if i-R <= 0 || i+R >= M
                        while i-R <= 0 
                            R=R-1;
                        end
                        while i+R >= M
                            R=R-1;
                        end
                    end
                    if j-R <= 0 || j+R >= N
                        while i-R <= 0 
                            R=R-1;
                        end
                        while i+R >= M
                            R=R-1;
                        end
                    end
                    tmp = y( i-R:i+R , j-R:j+R ); %% filter window
                    w = 2*R +1 ;    %% Size of filter window
                    for s = 1:w
                        for p = 1:w
                            if tmp(s,p) == 0 || tmp(s,p) == L(j)-1 
                                nnum=nnum+1;    %%number of free noise pixels in the window
                            end
                        end
                    end
                end

                clear temp;
                temp = y( i-R:i+R , j-R:j+R ); %% filter window
                S = sort(temp(:));
                if S(1) < S(R+1) && S(R+1) < S(w)  && 0 < S(R+1) && S(R+1) < L(j)
                    m(i,j) = S(R+1);
                end
                if S(1) >= S(R+1) || S(R+1) >= S(w) || S(R+1) == L(j) && S(R+1) == 0
                    t = y(i,j-1);
                    m(i,j) = t;
                end
            end
        end
    end

    for i = 1:M
        for j = 1:N
            if alpha(i,j) == 0      %% the value of z(i,j) is copied directly as the value of y(i,j).
                z = y(i,j);
            else
                z = m(i,j);   %%the output value z(i,j) is equal to m(i,j).
            end
        end
    end 
    imshow(x);
    figure,imshow(y);
    figure,imshow(z);
end

我有错误，请帮忙！

错误：内存不足。键入HELP MEMORY以获取选项。自适应错误（第96行） m（i，j）= t;

和

下标索引必须是实数正整数或逻辑值。自适应错误（第73行） tmp = y（i-R：i + R，j-R：j + R）; %%过滤器窗口

如何设计和实现一个用于脉冲噪声消除的自适应滤波器？

2 个答案: