parent
065f3d2bd1
commit
eb15f88b6e
@ -0,0 +1,228 @@
|
||||
% 2023-01-15
|
||||
|
||||
% based on the script for task one, this is rcnnscript for direct
|
||||
% classification
|
||||
|
||||
% Requirements:
|
||||
% data has to be provided as PicturesResizedLabelsResizedSignsCutted.zip
|
||||
% in script location
|
||||
% (this script unzips data and renames two files, but there is unlabeled
|
||||
% image-material, that has to be removed by hand after unzipping)
|
||||
|
||||
% additional scriptfiles:
|
||||
% - func_setupData.m
|
||||
% unpack data etc.
|
||||
% - func_groundTruthFromLabelPic.m
|
||||
% (generate groundtruthtablle from image-data)
|
||||
% - augmentData.m
|
||||
% (dataaugmentation für RCNN)
|
||||
% - helperSanitizeBoxes.m
|
||||
% (part of augmentation)
|
||||
% - preprocessData.m
|
||||
% (Resize image and bounding boxes to targetSize.
|
||||
%
|
||||
|
||||
% required add-on(s):
|
||||
% - 'Deep Learning Toolbox Model for ResNet-50 Network'
|
||||
% - 'image processing toolbox'
|
||||
% - 'Computer Vision Toolbox '
|
||||
|
||||
% recommended add-on(s) - if gpu is apt for the job....
|
||||
% - 'Parallel Computing Toolbox'
|
||||
|
||||
% adjustable parameters
|
||||
% - if there is no trained net, it can be trained with this script:
|
||||
% set doTraining to true
|
||||
% - the training can use augmentation or not:
|
||||
% set doAugmentation accordingly
|
||||
|
||||
close all;
|
||||
clear;
|
||||
|
||||
doTraining = true;
|
||||
|
||||
% first we need the data...
|
||||
dataDir = 'Picturedata'; % Destination-Folder for provided (img) Data
|
||||
zippedDataFile = 'PicturesResizedLabelsResizedSignsCutted.zip'; %Data provided by TA
|
||||
grDataFile = 'signDatasetGroundTruth_stepthree.mat';
|
||||
func_setupData_stepthree(dataDir, zippedDataFile, grDataFile);
|
||||
|
||||
%load data
|
||||
data = load(grDataFile);
|
||||
traficSignDataset = data.DataSet;
|
||||
|
||||
% ----- split the dataset into training, validation, and test sets.
|
||||
% Select 60% of the data for training, 10% for validation, and the
|
||||
% rest for testing the trained detector
|
||||
|
||||
rng(0)
|
||||
shuffledIndices = randperm(height(traficSignDataset));
|
||||
idx = floor(0.6 * height(traficSignDataset));
|
||||
|
||||
trainingIdx = 1:idx;
|
||||
trainingDataTbl = traficSignDataset(shuffledIndices(trainingIdx),:);
|
||||
|
||||
validationIdx = idx+1 : idx + 1 + floor(0.1 * length(shuffledIndices) );
|
||||
validationDataTbl = traficSignDataset(shuffledIndices(validationIdx),:);
|
||||
|
||||
testIdx = validationIdx(end)+1 : length(shuffledIndices);
|
||||
testDataTbl = traficSignDataset(shuffledIndices(testIdx),:);
|
||||
|
||||
% ----- use imageDatastore and boxLabelDatastore to create datastores
|
||||
% for loading the image and label data during training and evaluation.
|
||||
|
||||
imdsTrain = imageDatastore(trainingDataTbl{:,'imageFilename'});
|
||||
bldsTrain = boxLabelDatastore(trainingDataTbl(:,2:end))
|
||||
|
||||
imdsValidation = imageDatastore(validationDataTbl{:,'imageFilename'});
|
||||
bldsValidation = boxLabelDatastore(trainingDataTbl(:,2:end))
|
||||
|
||||
imdsTest = imageDatastore(testDataTbl{:,'imageFilename'});
|
||||
bldsTest = boxLabelDatastore(trainingDataTbl(:,2:end))
|
||||
|
||||
% combine image and box label datastores.
|
||||
|
||||
trainingData = combine(imdsTrain,bldsTrain); % erzeugt 'CombinedDatastore
|
||||
validationData = combine(imdsValidation,bldsValidation);
|
||||
testData = combine(imdsTest,bldsTest);
|
||||
|
||||
% display one of the training images and box labels.
|
||||
|
||||
data = read(trainingData);
|
||||
I = data{1};
|
||||
bbox = data{2};
|
||||
annotatedImage = insertShape(I,'Rectangle',bbox);
|
||||
annotatedImage = imresize(annotatedImage,4); % nur fuer Darstellung
|
||||
figure
|
||||
imshow(annotatedImage)
|
||||
|
||||
% ----- Create Faster R-CNN Detection Network
|
||||
|
||||
inputSize = [224 224 3];
|
||||
|
||||
preprocessedTrainingData = transform(trainingData, @(data)preprocessData(data,inputSize));
|
||||
% Achtung: dieser DS wird nur zur Ermittlung der BoundingBoxes verwendet
|
||||
|
||||
% display one of the training images and box labels.
|
||||
while 1==0 %hasdata(preprocessedTrainingData)
|
||||
data = read(preprocessedTrainingData);
|
||||
I = data{1};
|
||||
bbox = data{2};
|
||||
annotatedImage = insertShape(I,'Rectangle',bbox);
|
||||
annotatedImage = imresize(annotatedImage,4); % nur fuer Darstellung
|
||||
figure(1)
|
||||
imshow(annotatedImage)
|
||||
pause(0.100)
|
||||
end
|
||||
|
||||
% Auswahl der anchor boxes
|
||||
% Infos dazu: https://de.mathworks.com/help/vision/ug/estimate-anchor-boxes-from-training-data.html
|
||||
numAnchors = 3;
|
||||
anchorBoxes = estimateAnchorBoxes(preprocessedTrainingData,numAnchors);
|
||||
|
||||
% und das feature CNN
|
||||
featureExtractionNetwork = resnet50;
|
||||
featureLayer = 'activation_40_relu';
|
||||
numClasses = width(traficSignDataset)-1; % also hier: 1, es sollen nur Verkehrsschilder erkannt werden
|
||||
|
||||
lgraph = fasterRCNNLayers(inputSize,numClasses,anchorBoxes,featureExtractionNetwork,featureLayer);
|
||||
|
||||
% Netzwerk ansehen
|
||||
% analyzeNetwork(lgraph)
|
||||
augmentedTrainingData = transform(trainingData,@augmentData_stepthree);
|
||||
trainingData = transform(augmentedTrainingData,@(data)preprocessData(data,inputSize));
|
||||
validationData = transform(validationData,@(data)preprocessData(data,inputSize));
|
||||
|
||||
|
||||
options = trainingOptions('sgdm',...
|
||||
'MaxEpochs',10,...
|
||||
'MiniBatchSize',2,...
|
||||
'InitialLearnRate',1e-3,...
|
||||
'CheckpointPath',tempdir,...
|
||||
'ValidationData',validationData);
|
||||
|
||||
netname = "netDetectorResNet50_stepthree.mat"
|
||||
|
||||
|
||||
|
||||
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]);
|
||||
save netname detector;
|
||||
else
|
||||
% Load pretrained detector for the example.
|
||||
load netname detector;
|
||||
end
|
||||
|
||||
% ----- quick check/test
|
||||
|
||||
I = imresize(I,inputSize(1:2));
|
||||
[bboxes,scores] = detect(detector,I);
|
||||
% Display the results.
|
||||
|
||||
sfigTitle = ""
|
||||
if height(bboxes) > 0
|
||||
I = insertObjectAnnotation(I,'rectangle',bboxes,scores);
|
||||
sfigTitle = "Detected";
|
||||
else
|
||||
sfigTitle = "Not Detected";
|
||||
end
|
||||
|
||||
|
||||
figure;
|
||||
imshow(I);
|
||||
annotation('textbox', [0.5, 0.2, 0.1, 0.1], 'String', sfigTitle)
|
||||
|
||||
% ----- Testing
|
||||
|
||||
testData = transform(testData,@(data)preprocessData(data,inputSize));
|
||||
|
||||
% Run the detector on all the test images.
|
||||
|
||||
detectionResults = detect(detector,testData,'MinibatchSize',4);
|
||||
|
||||
% Evaluate the object detector using the average precision metric.
|
||||
|
||||
[ap, recall, precision] = evaluateDetectionPrecision(detectionResults,testData);
|
||||
% The precision/recall (PR) curve highlights how precise a detector is at varying levels of recall. The ideal precision is 1 at all recall levels. The use of more data can help improve the average precision but might require more training time. Plot the PR curve.
|
||||
|
||||
figure
|
||||
plot(recall,precision)
|
||||
xlabel('Recall')
|
||||
ylabel('Precision')
|
||||
grid on
|
||||
title(sprintf('Average Precision = %.2f', ap))
|
||||
|
||||
|
||||
% ----- Helper functions
|
||||
|
||||
% 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
|
||||
@ -0,0 +1,21 @@
|
||||
function data = augmentData_stepthree(data)
|
||||
% Randomly flip images and bounding boxes horizontally.
|
||||
tform = randomAffine2d('XReflection',true);
|
||||
sz = size(data{1});
|
||||
rout = affineOutputView(sz,tform);
|
||||
|
||||
% jitter
|
||||
data{1} = jitterColorHSV(data{1},...
|
||||
Contrast=0.2,...
|
||||
Hue=0,...
|
||||
Saturation=0.1,...
|
||||
Brightness=0.2);
|
||||
|
||||
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
|
||||
@ -0,0 +1,135 @@
|
||||
function [ ] = func_groundTruthFromLabelPic( dataStorePicturePath, dataStoreLabelPath, outFile )
|
||||
%
|
||||
% erstellt us einem Picture-Datastore und einem Label-Datastore
|
||||
% eine Groundtruth-Tabelle, wie sie z.B. in FasterRCNN.m
|
||||
% benoetigt wird
|
||||
%
|
||||
% file von tas
|
||||
% adaptiert als func 2022/12/28 vh
|
||||
%
|
||||
|
||||
labelDS = imageDatastore(dataStoreLabelPath, 'IncludeSubfolders', true);
|
||||
pictureDS = imageDatastore(dataStorePicturePath, 'IncludeSubfolders', true);
|
||||
|
||||
labelCount = numel(labelDS.Files)
|
||||
pictureCount = numel(pictureDS.Files)
|
||||
|
||||
if labelCount ~= pictureCount
|
||||
fprintf("!!!! Error: Die Anzahl der Bilder und Anzahl der LabelPicture sind ungleich -> Abbruch");
|
||||
return
|
||||
end
|
||||
|
||||
fprintf("-----------------------------------------------------------\n");
|
||||
fprintf("Picture-Verzeichnis: %s\n", dataStorePicturePath)
|
||||
fprintf("Anzahl Images: %d\n", pictureCount)
|
||||
fprintf("Label-Verzeichnis: %s\n", dataStoreLabelPath)
|
||||
fprintf("Anzahl Images: %d\n", labelCount)
|
||||
fprintf("BE PATIENT.... \n")
|
||||
fprintf("-----------------------------------------------------------\n");
|
||||
|
||||
% table anlegen
|
||||
sz = [pictureCount 6];
|
||||
varTypes = ["cellstr","cell","cell","cell","cell","logical"];
|
||||
varNames = ["imageFilename","30GBS","50GBS","60GBS","AndereGBS","valid"];
|
||||
DataSet = table('Size',sz,'VariableTypes',varTypes,'VariableNames',varNames);
|
||||
|
||||
rng(0)
|
||||
shuffledIndices = randperm(pictureCount);
|
||||
|
||||
% los gehts
|
||||
for i = 1:pictureCount
|
||||
shuffeldIndex = shuffledIndices(i);
|
||||
[imPic imPic_INFO]= readimage(pictureDS, shuffeldIndex);
|
||||
[im_path imPic_name im_ext]=fileparts(imPic_INFO.Filename);
|
||||
|
||||
[imLabel imLabel_INFO]= readimage(labelDS, shuffeldIndex);
|
||||
[lbl_path imLabel_name im_ext]=fileparts(imLabel_INFO.Filename);
|
||||
|
||||
%fprintf("picture: %s label: %s\n", imPic_name, imLabel_name);
|
||||
%fprintf("picturepath: %s labelpath: %s\n", im_path, lbl_path);
|
||||
box = [0,0,0,0]; %default if theres no labelimg
|
||||
|
||||
v = true;
|
||||
|
||||
if ~strcmp(imPic_name, imLabel_name)
|
||||
fprintf("!!!! Error: zum Picture gibt es kein entsprechendes LabelPicture -> Abbruch");
|
||||
imPic_name
|
||||
imLabel_name
|
||||
|
||||
else
|
||||
|
||||
% LabelRegion aus Image ausschneiden
|
||||
bw = imLabel;
|
||||
s = regionprops(bw, 'BoundingBox');
|
||||
box = cat(1, s.BoundingBox); % structure to matrix
|
||||
box = round(box);
|
||||
|
||||
% falls mehrere Marker vorhanden sind, ignorieren wir den Datensatz
|
||||
% das dürfte für das Training Region detection besser sein ??
|
||||
if (height(box) > 1)
|
||||
v = false;
|
||||
end
|
||||
box = box(1,:);
|
||||
|
||||
if (numel(box) ~= 4)
|
||||
fprintf("Boxkoordinaten nicht ok: %s %s\n", imPic_name, imLabel_name)
|
||||
box
|
||||
v = false
|
||||
end
|
||||
|
||||
end
|
||||
|
||||
a = [];
|
||||
b = [];
|
||||
c = [];
|
||||
d = [];
|
||||
|
||||
|
||||
if contains(im_path,'30GBS') == 1
|
||||
a = num2cell(box, 2);
|
||||
elseif contains(im_path,'50GBS') == 1
|
||||
b = num2cell(box, 2);
|
||||
elseif contains(im_path,'60GBS') == 1
|
||||
c = num2cell(box, 2);
|
||||
else
|
||||
d = num2cell(box, 2);
|
||||
end
|
||||
|
||||
|
||||
%check for boxes which are somewhat wrong
|
||||
if v
|
||||
if (box(3) < 8.) ...
|
||||
|| (box(4) < 8.) ...
|
||||
|| (abs(box(3) - box(4)) > 2) ...
|
||||
|| (box(1) + box(3) > 1024) ...
|
||||
|| (box(2) + box(4) > 768)
|
||||
fprintf("boxkoordinaten nicht i.o. %s (%d %d %d %d) \n", imLabel_name, box(1),box(2),box(3),box(4));
|
||||
v = false;
|
||||
end
|
||||
end
|
||||
|
||||
DataSet(shuffeldIndex,:) = {imPic_INFO.Filename,a,b,c,d,v};
|
||||
|
||||
% display one of the training images and box labels.
|
||||
if (shuffeldIndex == 4)
|
||||
annotatedImage = insertShape(imPic,'Rectangle',box);
|
||||
figure
|
||||
imshow(annotatedImage)
|
||||
end
|
||||
|
||||
end
|
||||
|
||||
%Die Daten sind teilweise nicht in Ordnung, am einfachsten ist natürlich
|
||||
%die Einträge zu löschen, die nicht gut sind.
|
||||
fprintf("Groundtruth hat %d eintraege vor der Bereinigung \n ", height(DataSet) )
|
||||
|
||||
toDelete = DataSet.valid == false;
|
||||
DataSet(toDelete,:) = [];
|
||||
|
||||
DataSet.valid=[];
|
||||
|
||||
fprintf("Groundtruth hat %d eintraege nach der Bereinigung \n", height(DataSet) )
|
||||
|
||||
save(outFile, 'DataSet' );
|
||||
|
||||
|
||||
@ -0,0 +1,43 @@
|
||||
function [ ] = func_setupData( dataDir, zippedDataFile, grDataFile )
|
||||
|
||||
% script func_setupData:
|
||||
% - entpackt die trainingsdaten,
|
||||
% - räumt ein bisschen auf, bzw. liefert Hinweise zum aufräumen
|
||||
% - erstellt grounddata
|
||||
% - (alles nur, falls es das nicht schon gibt)
|
||||
|
||||
if not(exist(dataDir , 'dir'))
|
||||
% unzip data
|
||||
if (not(exist(zippedDataFile , 'file')))
|
||||
fprintf("Data file is missing please copy %s to script folder !", zippedDataFile);
|
||||
return;
|
||||
end
|
||||
|
||||
fprintf("unzipping Data");
|
||||
unzip (zippedDataFile, dataDir)
|
||||
|
||||
%rename files correctly (there are two wrong):
|
||||
fprintf("fix faulty falenames\n");
|
||||
movefile (append(dataDir, '/Labels_1024_768/60GBS/60GBS_Gruppe05_SS21_Nr49.png'), append(dataDir, '/Labels_1024_768/60GBS/60GBS_Gruppe05_SS21_49.png'));
|
||||
movefile (append(dataDir, '/Labels_1024_768/60GBS/60GBS_Gruppe05_SS21_Nr50.png'), append(dataDir, '/Labels_1024_768/60GBS/60GBS_Gruppe05_SS21_50.png'));
|
||||
|
||||
fprintf("Labeldata is incomplete\n");
|
||||
fprintf("please compare data in keinGBS and delete what's to much manually\n");
|
||||
fprintf("otherwise grounddatageneration will fail\n");
|
||||
frpintf("then restart script \n");
|
||||
return;
|
||||
|
||||
end
|
||||
|
||||
% generate Grounddata from Pictures
|
||||
|
||||
if (not(exist(grDataFile , 'file')))
|
||||
dataStorePicturePath = append(pwd,'/', dataDir,'/Pictures_1024_768/');
|
||||
dataStoreLabelPath = append(pwd,'/', dataDir, '/Labels_1024_768/');
|
||||
|
||||
|
||||
|
||||
% Die Tabelle wird in einer Funktion erstellt und gespeichert
|
||||
% dabei werden Datensätze, wo die Label nicht passen entfernt.
|
||||
func_groundTruthFromLabelPic_stepthree(dataStorePicturePath, dataStoreLabelPath, grDataFile);
|
||||
end
|
||||
Loading…
Reference in new issue