Enhanced point descriptors

We are focused on how to describe a common image point distinctively, make its descriptor concise and invariant to general image transformations. We use neighborhood pixel characteristics, including HSV color space, Gaussian-weighted gradient magnitudes and orientations, sampled in specific window around interest point to enhance the description. The enhanced point descriptor (EPD) is a covariance matrix of the above-mentioned image characteristics. Experimental results show that, the performance of EPD, in the distinctiveness and invariance aspects, is as good as now popular local descriptors (SIFT and SURF), while the time cost of descriptor construction and matching is far less than them. Moreover, in comparison with SIFT and SURF, the EPD combines more image characteristics, which makes it be able to describe common image points, but not limited to the image extreme points. These advantages make the EPD finding new applications in the field of dense stereo matching.