拉普拉斯金字塔加权融合代码matlab

以下是拉普拉斯金字塔加权融合的MATLAB代码：

% 生成高斯金字塔
function pyramid = generate_gaussian_pyramid(image, levels)
    pyramid = {image};
    for i = 2:levels
        image = imresize(image, 0.5);
        pyramid{i} = image;
    end
end

% 生成拉普拉斯金字塔
function pyramid = generate_laplacian_pyramid(image, levels)
    gaussian_pyramid = generate_gaussian_pyramid(image, levels);
    pyramid = cell(1, levels);
    for i = 1:levels-1
        expanded = imresize(gaussian_pyramid{i+1}, size(gaussian_pyramid{i}));
        pyramid{i} = gaussian_pyramid{i} - expanded;
    end
    pyramid{levels} = gaussian_pyramid{levels};
end

% 加权融合
function fused_pyramid = weighted_fusion(laplacian_pyramid_1, laplacian_pyramid_2, mask_pyramid)
    fused_pyramid = cell(1, length(laplacian_pyramid_1));
    for i = 1:length(laplacian_pyramid_1)
        weight1 = double(mask_pyramid{i}) / 255.0;
        weight2 = 1.0 - weight1;
        fused_pyramid{i} = uint8(weight1 .* double(laplacian_pyramid_1{i}) + weight2 .* double(laplacian_pyramid_2{i}));
    end
end

% 重建图像
function image = reconstruct_image(pyramid)
    image = pyramid{end};
    for i = length(pyramid)-1:-1:1
        image = imresize(image, size(pyramid{i}));
        image = image + pyramid{i};
    end
end

% 加载两张图片和掩膜
image1 = imread('image1.jpg');
image2 = imread('image2.jpg');
mask = imread('mask.jpg');

% 将掩膜转换为三通道
mask = cat(3, mask, mask, mask);

% 生成拉普拉斯金字塔和掩膜金字塔
levels = 6;
laplacian_pyramid_1 = generate_laplacian_pyramid(image1, levels);
laplacian_pyramid_2 = generate_laplacian_pyramid(image2, levels);
mask_pyramid = generate_gaussian_pyramid(mask, levels);

% 加权融合
fused_pyramid = weighted_fusion(laplacian_pyramid_1, laplacian_pyramid_2, mask_pyramid);

% 重建图像
fused_image = reconstruct_image(fused_pyramid);

% 显示结果
imshow(fused_image);

在MATLAB代码中，`generate_gaussian_pyramid` 用于生成高斯金字塔，`generate_laplacian_pyramid` 用于生成拉普拉斯金字塔，`weighted_fusion` 用于加权融合，`reconstruct_image` 用于重建图像。在代码中，我们首先加载了两张图片和一个掩膜，然后将掩膜转换为三通道，接着生成拉普拉斯金字塔和掩膜金字塔，并使用 `weighted_fusion` 函数进行加权融合，最后使用 `reconstruct_image` 函数重建图像，并显示结果。