dear all, I am working in detection of microcalcification in mammogram images but i cannot able to remove the artifacts present in the image. Please help me
If you want to filter out quantum or Gaussian noise related artifacts, you may apply multi-resolution approaches so that the diagnostic details are not compromised.
Further, for improvement of contrast you may refer non-linear filtering approaches.
Refer:
A. Pandey, A. Yadav and V. Bhateja, “Contrast Improvement of Mammographic Masses Using Adaptive Volterra Filter," Proc. of (Springer) 4th International Conference on Signal and Image Processing (ICSIP 2012), Coimbatore, India, vol. 2, pp. 583-593, December 2012.
V. Bhateja, M. Misra, S. Urooj and A. Lay-Ekuakille, “A Robust Polynomial Filtering Framework for Mammographic Image Enhancement from Biomedical Sensors,” IEEE Sensors Journal, vol. 13, no. 11, pp. 4147-4156, November 2013.
These papers are also included in my profile here.
Before going into the computer software method of reducing artefacts as suggested by other scholars, you may wish to try improving your technique of image acquisition first.
If using a film-screen system then ensure that the fluorescent screen in the cassette is not stained or dusty. If using computed radiography the same goes for the photo-stimulable plates.
A little speck of dust on the plates or fluorescent screen can mimic microcalcification.
Once an artefact is there, the starting point would be to decode that it is an artefact.
The attached file may be a little useful.
Thank you.
Article WAJR: Greyscale Appearance of Film-screen Radiographic Artef...