matlab unet图像分割代码

以下是一个简单的 MATLAB UNet 图像分割代码示例：

%% 数据准备
imds = imageDatastore('path_to_images');
pxds = pixelLabelDatastore('path_to_labels', classes);

%% 定义 UNet 模型
inputSize = [256 256 3];
numClasses = numel(classes);
layers = [
    imageInputLayer(inputSize)
    convolution2dLayer(3, 64, 'Padding', 'same')
    batchNormalizationLayer
    reluLayer
    maxPooling2dLayer(2, 'Stride', 2)

    convolution2dLayer(3, 128, 'Padding', 'same')
    batchNormalizationLayer
    reluLayer
    maxPooling2dLayer(2, 'Stride', 2)

    convolution2dLayer(3, 256, 'Padding', 'same')
    batchNormalizationLayer
    reluLayer
    maxPooling2dLayer(2, 'Stride', 2)

    convolution2dLayer(3, 512, 'Padding', 'same')
    batchNormalizationLayer
    reluLayer
    transposedConv2dLayer(2, 512, 'Stride', 2, 'Cropping', 'same')

    concatenationLayer(3)
    convolution2dLayer(3, 256, 'Padding', 'same')
    batchNormalizationLayer
    reluLayer
    transposedConv2dLayer(2, 256, 'Stride', 2, 'Cropping', 'same')

    concatenationLayer(3)
    convolution2dLayer(3, 128, 'Padding', 'same')
    batchNormalizationLayer
    reluLayer
    transposedConv2dLayer(2, 128, 'Stride', 2, 'Cropping', 'same')

    concatenationLayer(3)
    convolution2dLayer(3, 64, 'Padding', 'same')
    batchNormalizationLayer
    reluLayer
    convolution2dLayer(3, numClasses, 'Padding', 'same')
    softmaxLayer
    pixelClassificationLayer
];

%% 训练模型
opts = trainingOptions('adam', ...
    'MaxEpochs', 20, ...
    'MiniBatchSize', 8, ...
    'InitialLearnRate', 1e-3, ...
    'Shuffle', 'every-epoch', ...
    'ValidationData', {imds_val, pxds_val}, ...
    'ValidationFrequency', 10, ...
    'Verbose', true, ...
    'Plots', 'training-progress');
trainedNet = trainNetwork(imds, pxds, layers, opts);

%% 对测试集进行预测
im_test = imread('path_to_test_image');
im_test_resized = imresize(im_test, [256 256]);
im_test_resized = double(im_test_resized)/255;
pred = semanticseg(im_test_resized, trainedNet);
imshow(im_test_resized)
imshow(label2rgb(pred))

请注意，此示例代码仅用于说明如何使用 MATLAB 实现 UNet 模型进行图像分割。实际上，您需要根据您的数据集和任务进行修改和调整。