odexample.m

data = load('vehicleDatasetGroundTruth.mat');
vehicleDataset = data.vehicleDataset;
rng(0)
shuffledIndices = randperm(height(vehicleDataset));
idx = floor(0.6 * height(vehicleDataset));

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);
data = read(trainingData);
I = data{1};
bbox = data{2};
annotatedImage = insertShape(I,'Rectangle',bbox);
annotatedImage = imresize(annotatedImage,2);
figure
imshow(annotatedImage)

inputSize = [224 224 3];
% preprocessedTrainingData = transform(trainingData, @(data)preprocessData(data,inputSize));
trainingData = transform(trainingData,@scaleGT);


numAnchors = 3;
% return;

anchorBoxes = estimateAnchorBoxes(trainingData,numAnchors)
featureExtractionNetwork = resnet50;
featureLayer = 'activation_40_relu';
numClasses = width(vehicleDataset)-1;
lgraph = fasterRCNNLayers(inputSize,numClasses,anchorBoxes,featureExtractionNetwork,featureLayer); 
data = read(trainingData);
I = data{1};
bbox = data{2};
annotatedImage = insertShape(I,'Rectangle',bbox);
annotatedImage = imresize(annotatedImage,2);
figure
imshow(annotatedImage)
% augmentedTrainingData = transform(trainingData,@augmentData);
% augmentedData = cell(4,1);
% for k = 1:4
%     data = read(augmentedTrainingData);
%     augmentedData{k} = insertShape(data{1},'Rectangle',data{2});
%     reset(augmentedTrainingData);
% end
% figure
% montage(augmentedData,'BorderSize',10)


options = trainingOptions('sgdm',...
    'MaxEpochs',10,...
    'MiniBatchSize',2,...
    'InitialLearnRate',1e-3,...
    'CheckpointPath',tempdir,...
    'ValidationData',validationData);
doTraining = false;
if doTraining
    % Train the Faster R-CNN detector.
    % * Adjust NegativeOverlapRange and PositiveOverlapRange to ensure
    %   that training samples tightly overlap with ground truth.
    [detector, info] = trainFasterRCNNObjectDetector(trainingData,lgraph,options, ...
        'NegativeOverlapRange',[0 0.3], ...
        'PositiveOverlapRange',[0.6 1]);
else
    % Load pretrained detector for the example.
    pretrained = load('fasterRCNNResNet50EndToEndVehicleExample.mat');
    detector = pretrained.detector;
end
I = imread(testDataTbl.imageFilename{2});
I = imresize(I,inputSize(1:2));
[bboxes,scores] = detect(detector,I);
I = insertObjectAnnotation(I,'rectangle',bboxes,scores);
figure
imshow(I)


function data = augmentData(data)
% Randomly flip images and bounding boxes horizontally.
tform = randomAffine2d('XReflection',true);
sz = size(data{1});
rout = affineOutputView(sz,tform);
data{1} = imwarp(data{1},tform,'OutputView',rout);

% Sanitize box data, if needed.
data{2} = helperSanitizeBoxes(data{2}, sz);

% Warp boxes.
data{2} = bboxwarp(data{2},tform,rout);
end

function data = preprocessData(data,targetSize)
% Resize image and bounding boxes to targetSize.
sz = size(data{1},[1 2]);
scale = targetSize(1:2)./sz;
data{1} = imresize(data{1},targetSize(1:2));

% Sanitize box data, if needed.
data{2} = helperSanitizeBoxes(data{2}, sz);

% Resize boxes.
data{2} = bboxresize(data{2},scale);
end

function data = scaleGT(data)  
    targetSize = [224 224];
    % data{1} is the image
    scale = targetSize./size(data{1},[1 2]);
    data{1} = imresize(data{1},targetSize);
    % data{2} is the bounding box
    data{2} = bboxresize(data{2},scale);
end