Is there any python library function to perform blob detection?

Question

I was trying to implement blob detection in Python. I have gone through an implemented Matlab code. But, there is a problem I couldn't understand the code. Is there any function related to this one in OpnenCV, sciktimage, etc. package?

The name of the file of the implemented code is "ExtractNLargestBlobs.m" and the code is-

function binaryImage = ExtractNLargestBlobs(binaryImage, numberToExtract)
try
    % Get all the blob properties.  Can only pass in originalImage in version R2008a and later.
    [labeledImage, numberOfBlobs] = bwlabel(binaryImage);
    blobMeasurements = regionprops(labeledImage, 'area');
    % Get all the areas
    allAreas = [blobMeasurements.Area];
    if numberToExtract > 0
        % For positive numbers, sort in order of largest to smallest.
        % Sort them.
        [sortedAreas, sortIndexes] = sort(allAreas, 'descend');
    elseif numberToExtract < 0
        % For negative numbers, sort in order of smallest to largest.
        % Sort them.
        [sortedAreas, sortIndexes] = sort(allAreas, 'ascend');
        % Need to negate numberToExtract so we can use it in sortIndexes later.
        numberToExtract = -numberToExtract;
    else
        % numberToExtract = 0.  Shouldn't happen.  Return no blobs.
        binaryImage = false(size(binaryImage));
        return;
    end
    % Extract the "numberToExtract" largest blob(a)s using ismember().
    biggestBlob = ismember(labeledImage, sortIndexes(1:numberToExtract));
    % Convert from integer labeled image into binary (logical) image.
    binaryImage = biggestBlob > 0;
catch ME
    errorMessage = sprintf('Error in function ExtractNLargestBlobs().\n\nError Message:\n%s', ME.message);
    fprintf(1, '%s\n', errorMessage);
    uiwait(warndlg(errorMessage));
end

and this file is accessing from the file namely main.m and the code line that is used to access are-

%--------------------blob detection-------------------------------
            mask = ExtractNLargestBlobs(mask2, 1);

The parameter mask2 consists of an Image previously gone through morphological analysis. And the variable mask assigns an image that image is expected to be a clean image that doesn't have the unwanted regions in the image.

Actually, I was implementing the code in Python and I'm not familiar with Matlab. Unfortunately, I couldn't find the above code implemented in Python. Is there any Python library method that could be used to replace the above method? and to overcome the situations like this is it okay to follow the given solution from stackoverflow?

I have found a python implemented code by fmw42 but my code won't suit with the solution despite modifying the code lines - here is the solution

Answer 1

If you are free to mark down some blobs or ROI on an image and train a simple detector try out dlib.

An example code of how to train the object detector in Python is provided in the project itself (here).

The object detector that you get here is HOG-SVM. In short, it uses the HOG descriptor of regions of interest (positive, you mark these using the imglab tool) and non-interest (negative) samples to train a SVM classifier.

DLIB has been built for tackling real-world machine learning problems hence the inference or testing is very fast. The classifier is pretty robust as well if trained with proper parameters and data.

NOTE: To install dlib you need CMake. For labelling you will need the imglab tool which is included in the project.