How to find seed point for region growing algorithm?

Actually my project is brain tumor segmentation in MRI images. I want to segment the brain MRI images using region growing technique. How can I find a better seed point that detects the brain tumor efficiently.Sample images are attached.

[Sai praneeth Boddapati]

hi asia malik i am doing the same project and i want the code for tumor segemntation using region growing can you please help me with the code

[Image Analyst]

@Sai praneeth Boddapati, why region growing specifically?

See attached.

Answer 1

[Image Analyst]

For those images, maybe you can just threshold, call bwareafilt to get the largest blob, call bwlabel, and then call regionprops asking for the centroid.

grayImage = imread(filename);
binaryImage = grayImage > someThreshold;
binaryImage = bwareafilt(binaryImage, 1);
labeledImage = bwlabel(binaryImage);
measurements = regionprops(labeledImage, 'Centroid');
xCenter = round(measurements.Centroid(1));
yCenter = round(measurements.Centroid(2));

Answer 2

[khansa sanabal]

These errors are occuring in above example

Error in bwpropfilt>parse_inputs (line 119)

validateattributes(bw, {'logical'}, {'nonsparse', '2d'}, mfilename, 'BW', 1);

Error in bwpropfilt (line 58)

[bw, I, attrib, p, direction, conn] = parse_inputs(args{:});

Error in bwareafilt (line 34)

bw2 = bwpropfilt(bw, 'area', p, direction, conn);

Error in findseed (line 5)

binaryImage = bwareafilt(binaryImage, 1);

[Image Analyst]

You forgot to attach your image, and your findseed.m file. You need to do so it you want an answer.