how to MAKE an image as a svm classifier input ?

i have matlab 2012 sample codes for svm (support vector machine) classifier. In these code "HOW CAN I ADD THE 2D IMAGE AS A INPUT FOR MALE & FEMALE DETECTION

Answer 1

[Uttiya Ghosh]

Hi Anto,

As per my understanding you would like to know the procedure to train a SVM classifier using grayscale images. I have used R2020a version of MATLAB to train trainImageCategoryClassifier function on a set of digits. This function trains a support vector machine (SVM) multiclass classifier using the input bag (bagOfFeatures object). PFB the code required to perform the desired task.

setDir = fullfile(matlabroot,'toolbox','nnet','nndemos', ...
    'nndatasets','DigitDataset');
imds = imageDatastore(setDir,'IncludeSubfolders',true,'LabelSource',...
    'foldernames');
[imdsTrain,imdsTest] = splitEachLabel(imds,0.9,'randomize');
bag = bagOfFeatures(imdsTrain,"VocabularySize",100);
mdl = trainImageCategoryClassifier(imdsTrain,bag);
confMatrix = evaluate(mdl,imdsTest);

For more information, refer to the following links.

https://www.mathworks.com/help/vision/ref/bagoffeatures.html

https://www.mathworks.com/help/vision/ref/trainimagecategoryclassifier.html