feature extraction on specified cordinate on image

Hi I want to extract some features for train data on the specified part of the image I don't know how can i go to that part and run my code for the features?for example the malignant mass is located at 538 681 29 as is indicated in the ground truth data and I want the features at that region. Thank you in advance to guide me.

Answer 1

[Jing]

For extracting features of a part of the image, you can just index into the image or create a new image of that part to do the extraction.

[bahar h]

Thanks

Answer 2

[Image Analyst]

See my image segmentation tutorial: http://www.mathworks.com/matlabcentral/fileexchange/?term=authorid%3A31862

If you still need help, post your image somewhere, and tell us where you posted it and let us know what kind of help you need. http://www.mathworks.com/matlabcentral/answers/7924-where-can-i-upload-images-and-files-for-use-on-matlab-answers

[bahar h]

Thank you so much

[bahar h]

Hi Image Analyst, thank you for the tutorial but I couldn't find my answer within it,actually I have some mammogram images that I know the right position of malignant and benign masses on them,as I said before the position are written in the database for example like this: mdb023 G CIRC M 538 681 while: G - Fatty-glandular and M - Malignant now I want to extract some features like GLRLM,GLDM,morphological and so on,at the mentioned positions to use as my train data for svm classifier and I don't know how can I go to the exact mentioned ROI to extract the features.the image is from MIAS database and I don't know if I am allowed to share it on dropbox or not?I will be appreciated if you help me. Thank you

[Image Analyst]

That's a subject that people have been working on for decades so it's certainly not going to be any 5 minute demo that I can whip up for you. Please go to VisionBib for a list of papers on the subject. http://iris.usc.edu/Vision-Notes/bibliography/contentsmedical.html#Medical%20Applications,%20CAT,%20MRI,%20Ultrasound,%20Heart%20Models,%20Brain%20Models