Circular mesh-based shape and margin descriptor for object detection

Abstract

In many engineering and medical imaging fields, shape, and margin descriptors play an important role in challenging pattern recognition and classification problems. This paper presents a circular mesh-based shape and margin descriptor (CMSMD) for object recognition and classification. This shape descriptor is the first to have the functions of both structural and global contour-based descriptors. In the proposed descriptor, object contours are embedded in a circular mesh and labelled using circular mesh-based border labelling. New features can also be derived using the proposed descriptor. Further, an algorithm that employs the CMSMD characteristics can be utilised to identify the convexity and concavity of the embedded contours in linear complexity. The effectiveness of the proposed descriptor was demonstrated by performing lesion detection using dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) results. The obtained accuracy of 94.69% shows that the proposed descriptor is superior to the existing shape descriptors used for lesion detection in DCE-MRI.