Mammogram image enhancement using a two-threshold model of shadowed set with gradual representation of cardinality

Breast cancer can cause severe side effects on women's health. According to some recent study it has become the second leading cause of woman's death by cancer. Screening is one of the most successful ways of early-stage detection of breast cancer even before it has started showing any symptom. Among other popular screening tests, mammography plays a significant role in early-stage breast cancer detection. Microcalcification is an important characteristic of a mammogram. However, the identification of calcification in the presence of dense breast tissues is a difficult task, contrast enhancement plays a significant role to improve the visual quality of mammograms, thereby helps to identify microcalcifications. In this regard to deal with the challenges of identifying microcalcification from low contrast mammogram images we have applied a two-threshold shadowed set where cardinality has a gradual representation. The overall process is done in two-steps. Separately, we perform the global processing and local processing on the image then finally we combine the effect of both. Furthermore, subjective and objective measures have also been provided along with a comparative study with the stand alone enhancement approaches as part of experimental analysis.