使用 YOLO v3 深度学习进行目标检测

unzip vehicleDatasetImages.zip
data = load("vehicleDatasetGroundTruth.mat");
vehicleDataset = data.vehicleDataset;

vehicleDataset(1:4,:)

vehicleDataset.imageFilename = fullfile(pwd,vehicleDataset.imageFilename);

rng(0);
shuffledIndices = randperm(height(vehicleDataset));
idx = floor(0.6 * length(shuffledIndices) );

trainingIdx = 1:idx;
trainingDataTbl = vehicleDataset(shuffledIndices(trainingIdx),:);

validationIdx = idx+1 : idx + 1 + floor(0.1 * length(shuffledIndices) );
validationDataTbl = vehicleDataset(shuffledIndices(validationIdx),:);

testIdx = validationIdx(end)+1 : length(shuffledIndices);
testDataTbl = vehicleDataset(shuffledIndices(testIdx),:);

imdsTrain = imageDatastore(trainingDataTbl{:,"imageFilename"});
bldsTrain = boxLabelDatastore(trainingDataTbl(:,"vehicle"));

imdsValidation = imageDatastore(validationDataTbl{:,"imageFilename"});
bldsValidation = boxLabelDatastore(validationDataTbl(:,"vehicle"));

imdsTest = imageDatastore(testDataTbl{:,"imageFilename"});
bldsTest = boxLabelDatastore(testDataTbl(:,"vehicle"));

trainingData = combine(imdsTrain,bldsTrain);
validationData = combine(imdsValidation,bldsValidation);
testData = combine(imdsTest,bldsTest);

validateInputData(trainingData);
validateInputData(validationData);
validateInputData(testData);

augmentedTrainingData = transform(trainingData,@augmentData);

augmentedData = cell(4,1);
for k = 1:4
    data = read(augmentedTrainingData);
    augmentedData{k} = insertShape(data{1,1},"Rectangle",data{1,2});
    reset(augmentedTrainingData);
end
figure
montage(augmentedData,BorderSize=10)

reset(augmentedTrainingData);

networkInputSize = [227 227 3];

rng(0)
trainingDataForEstimation = transform(trainingData,@(data)preprocessData(data,networkInputSize));
numAnchors = 6;
[anchors,meanIoU] = estimateAnchorBoxes(trainingDataForEstimation,numAnchors)

area = anchors(:,1).*anchors(:,2);
[~,idx] = sort(area,"descend");
anchors = anchors(idx,:);
anchorBoxes = {anchors(1:3,:)
    anchors(4:6,:)
    };

baseNetwork = imagePretrainedNetwork("squeezenet");
classNames = trainingDataTbl.Properties.VariableNames(2:end);

yolov3Detector = yolov3ObjectDetector(baseNetwork,classNames,anchorBoxes, ...
    DetectionNetworkSource=["fire9-concat","fire5-concat"],InputSize=networkInputSize);

options = trainingOptions("adam", ...
    GradientDecayFactor=0.9, ...
    SquaredGradientDecayFactor=0.999, ...
    InitialLearnRate=0.001, ...
    LearnRateSchedule="none", ...
    MiniBatchSize=8, ...
    L2Regularization=0.0005, ...
    MaxEpochs=80, ...
    PreprocessingEnvironment="parallel", ...
    ResetInputNormalization=true, ...
    Shuffle="every-epoch", ...
    VerboseFrequency=20, ...
    ValidationFrequency=1000, ...
    CheckpointPath=tempdir, ...
    ValidationData=validationData, ...
    Metrics = mAPObjectDetectionMetric(Name="map50"));

doTraining = false;
if doTraining       
    % Train the YOLO v3 detector.
    [yolov3Detector,info] = trainYOLOv3ObjectDetector(augmentedTrainingData,yolov3Detector,options);
else
    % Load pretrained detector for the example.
    yolov3Detector = downloadPretrainedYOLOv3Detector();
end

results = detect(yolov3Detector,testData,MiniBatchSize=8,Threshold=0.01);

metrics = evaluateObjectDetection(results,testData);

[precision,recall] = precisionRecall(metrics);
AP = averagePrecision(metrics);

figure
plot(recall{:},precision{:})
xlabel("Recall")
ylabel("Precision")
grid on
title("Average Precision = "+AP)

data = read(testData);
I = data{1};

[bboxes,scores,labels] = detect(yolov3Detector,I);

I = insertObjectAnnotation(I,"rectangle",bboxes,scores);
figure
imshow(I)

function data = augmentData(A)
% Apply random horizontal flipping, and random X/Y scaling. Boxes that get
% scaled outside the bounds are clipped if the overlap is above 0.25. Also,
% jitter image color.

data = cell(size(A));
for ii = 1:size(A,1)
    I = A{ii,1};
    bboxes = A{ii,2};
    labels = A{ii,3};
    sz = size(I);

    if numel(sz) == 3 && sz(3) == 3
        I = jitterColorHSV(I, ...
            Contrast=0, ...
            Hue=0.1, ...
            Saturation=0.2, ...
            Brightness=0.2);
    end
    
    % Randomly flip image.
    tform = randomAffine2d(XReflection=true,Scale=[1 1.1]);
    rout = affineOutputView(sz,tform,BoundsStyle="centerOutput");
    I = imwarp(I,tform,OutputView=rout);
    
    % Apply same transform to boxes.
    [bboxes,indices] = bboxwarp(bboxes,tform,rout,OverlapThreshold=0.25);
    bboxes = round(bboxes);
    labels = labels(indices);
    
    % Return original data only when all boxes are removed by warping.
    if isempty(indices)
        data(ii,:) = A(ii,:);
    else
        data(ii,:) = {I,bboxes,labels};
    end
end
end

function data = preprocessData(data,targetSize)
% Resize the images and scale the pixels to between 0 and 1. Also scale the
% corresponding bounding boxes.

for ii = 1:size(data,1)
    I = data{ii,1};
    imgSize = size(I);
    
    % Convert an input image with single channel to 3 channels.
    if numel(imgSize) < 3 
        I = repmat(I,1,1,3);
    end
    bboxes = data{ii,2};

    I = im2single(imresize(I,targetSize(1:2)));
    scale = targetSize(1:2)./imgSize(1:2);
    bboxes = bboxresize(bboxes,scale);
    
    data(ii,1:2) = {I,bboxes};
end
end

function detector = downloadPretrainedYOLOv3Detector()
% Download a pretrained yolov3 detector.
if ~exist("yolov3SqueezeNetVehicleExample_21aSPKG.mat","file")
    if ~exist("yolov3SqueezeNetVehicleExample_21aSPKG.zip","file")
        disp("Downloading pretrained detector...");
        pretrainedURL = "https://ssd.mathworks.com/supportfiles/vision/data/yolov3SqueezeNetVehicleExample_21aSPKG.zip";
        websave("yolov3SqueezeNetVehicleExample_21aSPKG.zip",pretrainedURL);
    end
    unzip("yolov3SqueezeNetVehicleExample_21aSPKG.zip");
end
pretrained = load("yolov3SqueezeNetVehicleExample_21aSPKG.mat");
detector = pretrained.detector;
end