UNABLE TO PERFORM ASSIGMENT BECAUSE THE SIZE OF THE LEFT SIDE IS 1 BY 1260 AND THE SIZE OF THE RIGHT SIDE IS 1 BY 5040

1 次查看（过去 30 天）

Oriol 2023-1-6

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1889477-unable-to-perform-assigment-because-the-size-of-the-left-side-is-1-by-1260-and-the-size-of-the-right

评论： Image Analyst 2023-1-27

function matr_text=ocr(im, bb)
Dir="C:\Users\Usuario\Downloads\vcc2\vcc";
trainingSet = imageDatastore(Dir,'IncludeSubfolders',true,'LabelSource','foldernames');
countEachLabel(trainingSet)
numImages = numel(trainingSet.Files);
img = readimage(trainingSet, 1);
[hog_4x4, vis4x4] = extractHOGFeatures(img,'CellSize',[4 4]);
plot(vis4x4)
cellSize = [4 4];
hogFeatureSize = length(hog_4x4);
trainingFeatures = zeros(numImages,hogFeatureSize,'single');
for i = 1:numImages
    img = readimage(trainingSet,i);
    
    img = rgb2gray(img);
    
    
    % Apply pre-processing steps
    img = imbinarize(img);
    
    trainingFeatures(i,:) = extractHOGFeatures(img, 'CellSize',cellSize);  
end
% Get labels for each image.
trainingLabels = trainingSet.Labels;
classifier = fitcecoc(trainingFeatures, trainingLabels);
[m, n] = size(bb)
for j = 1:m
    img = imcrop(im, bb(j,:));
    figure, imshow(img)
    img = rgb2gray(img);
    
    % Apply pre-processing steps
    img = imbinarize(img);
    img=imresize(img, [45,60])
    figure, imshow(img)
    features(j,:) = extractHOGFeatures(img,'CellSize',cellSize);
end
% Make class predictions using the test features.
predictedLabels = predict(classifier, features);
end