總流程類似于幫助文檔幫助文檔圖像回歸
神經網絡結構類似于Unet，由卷積、激活、池化、轉置卷積、深度連接等layers組成
先準備好數據集，我上傳在資源里，可下載https://download.csdn.net/download/qq_51750957/14115738
1.matlab自帶的0-9手寫數字，共10000張，為[28,28]單通道圖像

2.對應的對抗樣本（加了干擾的圖片）10000張（如何生成對抗樣本可參照另一篇）

導入數據

imds = imageDatastore('DigitDataset', 'IncludeSubfolders', true, 'labelsource', 'foldernames');%讀取原始圖片 [imdsTrain,imdsValidation] = splitEachLabel(imds,0.9);%拆分出驗證集 advs = imageDatastore('digit_adv0.15', 'IncludeSubfolders', true, 'labelsource', 'foldernames');%讀取對抗樣本 [advsTrain,advsValidation] = splitEachLabel(advs,0.9);

檢查前十張圖片是否對應

for i=1:10figure(i);subplot(1,2,1);imshow(imread(imds.Files{i}));subplot(1,2,2);imshow(imread(advs.Files{i}));list1{i}=imds.Files{i};list2{i}=advs.Files{i}; end

組合數據，形成輸入輸出

dsTrain = combine(advsTrain,imdsTrain);%合并為兩列數據，第一列為對抗樣本（輸入），第二列為原始圖片（輸出） dsVal = combine(advsValidation,imdsValidation);%同上 dsTrain = transform(dsTrain,@commonPreprocessing);%歸一化到【0,1】,轉為灰度圖，縮放為[32,32]大小，commonPreprocessing定義在最后 dsVal = transform(dsVal,@commonPreprocessing);%同上dsTrain=shuffle(dsTrain);%打亂dataOut = readall(dsTrain);%顯示前8對數據 inputs = dataOut(:,1); responses = dataOut(:,2); minibatch = cat(2,inputs,responses); montage(minibatch','Size',[8 2]) title('Inputs (Left) and Responses (Right)')

設置訓練參數

options = trainingOptions('adam', ...'MaxEpochs',30, ...'MiniBatchSize',32, ...'ValidationData',dsVal, ...'ValidationFrequency',300,...'ValidationPatience',100,...'Shuffle','never', ...'Plots','training-progress', ...'LearnRateSchedule','piecewise',...'LearnRateDropPeriod',5,...'LearnRateDropFactor',0.5,...'Verbose',false);

打開深度學習app，開始搭建網絡，輸入大小[32,32,1]，輸出也是[32,32,1]，輸入和輸出每個像素都在【0,1】范圍內。網絡包含轉置卷積層、深度連接層，比較復雜
搭建好后可導出為lgraph
網絡很長，可以自己修改

創建深度學習網絡架構 該腳本可創建具有以下屬性的深度學習網絡: 層數: 42 連接數: 44 運行腳本以在工作區變量 lgraph 中創建層。 要了解詳細信息，請參閱 從 Deep Network Designer 生成 MATLAB 代碼。 由 MATLAB 于 2021-01-11 20:52:38 自動生成 創建層次圖 創建層次圖變量以包含網絡層。 lgraph = layerGraph();添加層分支 將網絡分支添加到層次圖中。每個分支均為一個線性層組。 tempLayers = [imageInputLayer([32 32 1],"Name","imageinput","Normalization","rescale-zero-one")convolution2dLayer([3 3],32,"Name","conv_1_1","Padding","same")batchNormalizationLayer("Name","batchnorm_1")reluLayer("Name","relu_1_1")convolution2dLayer([3 3],32,"Name","conv_1_2","Padding","same")batchNormalizationLayer("Name","batchnorm_2")reluLayer("Name","relu_1_2")convolution2dLayer([3 3],32,"Name","conv_1_3","Padding","same")batchNormalizationLayer("Name","batchnorm_3")reluLayer("Name","relu_1_3")]; lgraph = addLayers(lgraph,tempLayers);tempLayers = [maxPooling2dLayer([2 2],"Name","maxpool_1","Padding","same","Stride",[2 2])convolution2dLayer([3 3],64,"Name","conv_2","Padding","same")reluLayer("Name","relu_2")]; lgraph = addLayers(lgraph,tempLayers);tempLayers = [dropoutLayer(0.15,"Name","dropout")maxPooling2dLayer([2 2],"Name","maxpool_2","Padding","same","Stride",[2 2])convolution2dLayer([3 3],128,"Name","conv_3","Padding","same")reluLayer("Name","relu_3")]; lgraph = addLayers(lgraph,tempLayers);tempLayers = [maxPooling2dLayer([2 2],"Name","maxpool_3","Padding","same","Stride",[2 2])convolution2dLayer([3 3],256,"Name","conv_5","Padding","same")batchNormalizationLayer("Name","batchnorm_6")reluLayer("Name","relu_7")transposedConv2dLayer([4 4],128,"Name","transposed-conv_1","Cropping",[1 1 1 1],"Stride",[2 2])reluLayer("Name","relu_4")]; lgraph = addLayers(lgraph,tempLayers);tempLayers = [depthConcatenationLayer(2,"Name","depthcat_3")convolution2dLayer([3 3],128,"Name","conv_1_4","Padding","same")batchNormalizationLayer("Name","batchnorm_4")reluLayer("Name","relu_1_4")transposedConv2dLayer([4 4],64,"Name","transposed-conv_2","Cropping",[1 1 1 1],"Stride",[2 2])reluLayer("Name","relu_5")]; lgraph = addLayers(lgraph,tempLayers);tempLayers = [depthConcatenationLayer(2,"Name","depthcat_1")convolution2dLayer([3 3],64,"Name","conv_1_5","Padding","same")batchNormalizationLayer("Name","batchnorm_5")reluLayer("Name","relu_1_5")transposedConv2dLayer([4 4],32,"Name","transposed-conv_3","Cropping",[1 1 1 1],"Stride",[2 2])reluLayer("Name","relu_6")]; lgraph = addLayers(lgraph,tempLayers);tempLayers = [depthConcatenationLayer(2,"Name","depthcat_2")convolution2dLayer([3 3],32,"Name","conv_6","Padding","same")batchNormalizationLayer("Name","batchnorm_7")reluLayer("Name","relu_8")convolution2dLayer([1 1],1,"Name","conv_4","Padding","same")clippedReluLayer(1,"Name","clippedrelu")regressionLayer("Name","regressionoutput")]; lgraph = addLayers(lgraph,tempLayers);% 清理輔助變量 clear tempLayers;連接層分支 連接網絡的所有分支以創建網絡圖。 lgraph = connectLayers(lgraph,"relu_1_3","maxpool_1"); lgraph = connectLayers(lgraph,"relu_1_3","depthcat_2/in2"); lgraph = connectLayers(lgraph,"relu_2","dropout"); lgraph = connectLayers(lgraph,"relu_2","depthcat_1/in1"); lgraph = connectLayers(lgraph,"relu_3","maxpool_3"); lgraph = connectLayers(lgraph,"relu_3","depthcat_3/in1"); lgraph = connectLayers(lgraph,"relu_4","depthcat_3/in2"); lgraph = connectLayers(lgraph,"relu_5","depthcat_1/in2"); lgraph = connectLayers(lgraph,"relu_6","depthcat_2/in1");繪制層 plot(lgraph);

開始訓練，訓練耗時很長，可以適當更改maxEpoch

denoise_net = trainNetwork(dsTrain,lgraph,options);

訓練完成，嘗試預測5張圖片，左邊是輸入，右邊是輸出

for ii=0:5figure(ii+1)I=imread(['digit_adv0.15/',num2str(ii),'/0000.png']);%讀入圖片I=imresize(I,[32,32]);%I=rgb2gray(I);%如果是彩圖，要轉為灰度I=single(I);I=rescale(I);subplot(1,2,1);imshow(I);ypred = predict(denoise_net,I);%預測subplot(1,2,2);imshow(rescale(ypred)); end

成功修復了簡單的對抗樣本

函數

function dataOut = commonPreprocessing(data)dataOut = cell(size(data)); for col = 1:size(data,2)for idx = 1:size(data,1)temp = (data{idx,col});temp = imresize(temp,[32,32]);s=size(size(temp));if s(2)==3temp= rgb2gray(temp);end%temp = rescale(temp);dataOut{idx,col} = temp;end end end

總結

以上是生活随笔為你收集整理的matlab 卷积神经网络 图像去噪 对抗样本修复的全部內容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網站內容還不錯，歡迎將生活随笔推薦給好友。