Identification of the breast boundary in digital mammograms is one of the most important preprocessing steps of the Computer Aided Detection (CAD) and Computer Aided Diagnosis (CADi) systems. Accurate border detection increases CAD system accuracy and reliability and reduces program execution time and capacity needs in image storage. The breast contour lies in radiographically similar regions, the soft-tissue and the non-breast region. This feature makes the segmentation of the breast-region from the back-ground process difficult. In last decade, there have been lots of methods presented for breast border detection but nevertheless most of them are complicated and consumes high execution time. In this study, pixel intensity based a novel approach Cross - Border Detection Method is proposed, for accurate breast region segmentation in digital mammograms.