A mammogram contains two distinctive regions: the exposed breast region and the unexposed air-background region. The background region often contains radiopaque artifacts in the form of identification labels, radiopaque markers, and wedges. The primary motivation for removing such artifacts from mammograms is too lessen their effect on subsequent processing algorithms. For example, accurate segmentation of the breast region is an important pre-processing step in the computerized analysis of mammograms. It allows the search for abnormalities to be limited to the breast region of the mammogram without undue influence from the background. One of the problems with precise segmentation of the breast region is that high-intensity radiopaque artifacts can result in a non-uniform background region, and interfere with deriving an accurate representation of the breast contour. This paper proposes a new approach for removing radiopaque artifacts from the background region of mammograms based on the concept of area morphology. Area morphology uses attributes of structures rather than a fixed shape structuring element as used in classical morphology. This allows radiopaque artifacts to be removed, irrespective of shape.
The process of contrast enhancement refers to the accentuation, or sharpening of image structures to allow for improved
image analysis and interpretation. A mammogram is a x-ray projection of the 3D structures of the breast obtained by
compressing the breast between two plates. Unlike most other x-ray or Computed Tomography images, mammograms
have an inherent "fuzzy" or diffuse appearance. This is due in part to the superimposition of densities from differing
breast tissues, and the differential x-ray attenuation (absorption) characteristics associated with these various tissues.
Much of the difficulty in accurately interpreting a mammogram is related to there being insufficient contrast to
accurately identify potential abnormalities. Recently, morphology-based algorithms based on structuring elements have
been proposed for contrast enhancement. One of the limitations of these approaches from traditional morphology is their
dependence on the shape of structuring elements. In certain circumstances it may be more appropriate to filter an image
using attributes of structures such as their size, irrespective of shape. This paper introduces a novel nonlinear
enhancement technique that is based on the concept of area morphology. Various mammogram structures are enhanced
to illustrate the technique and a comparison is made with enhancement techniques such as “Contrast Limited Adaptive
Histogram Equalization” and classical morphological enhancement.