An enhanced correlation-based method for stereo correspondence with subpixel accuracy

Abstract

The invariance of the similarity measure in photometric distortions as well as its capability in producing subpixel accuracy are two desired and often required features in most stereo vision applications. In this paper we propose a new correlation-based measure which incorporates both mentioned requirements. Specifically, by using an appropriate interpolation scheme in the candidate windows of the matching image, and using the classical zero mean normalized cross correlation function, we introduce a suitable measure. Although the proposed measure is a nonlinear function of the sub-pixel displacement parameter, its maximization results in a closed form solution, resulting in reduced complexity for its use in matching techniques. Application of the proposed measure in a number of benchmark stereo pair images reveals its superiority over existing correlation-based techniques used for sub-pixel accuracy.

Geometrically constrained sub-pixel disparity estimation from stereo images of the retinal fundus

“…In [9] the author proposes a similarity measure based on the correlation coefficient that is called Enhanced Normalized Cross Correlation (ENCC). This measure is based on a linear interpolation scheme.…”

“…This improves the precision of the depth estimation process and thus the quality of the synthesized images for multi-viewpoint visualization. In [9] the authors propose a new similarity measure which is based on the correlation coefficient called Enhanced Normalized Cross Correlation (ENCC). More specifically, by using an appropriate linear interpolation scheme on the intensities of two adjacent candidate windows, a new similarity measure is introduced.…”

The April 2007 Swarm in Trichonis Lake Using Data From a Microseismic Network

“…Finally, the computation of the time differences between events of the same cluster, is based on the maximization of a similarity measure called Enhanced Normalized Cross Correlation (ENCC) (Psarakis and Evangelidis, 2005), which constitutes an extension of the sample level correlation coefficient. The above mentioned similarity measure is a continuous function of the time lag parameter, thus leading to sub -sample accuracy.…”

Study on Inspection of Circular Parts Used in Mechatronics

“…For low contrast components and complex shapes, vision systems encounter a number of difficulties and the best solution has not yet been implemented. From a scientific point of view, there are a number of solutions that offer a certain type of lighting or multi-camera systems [5].…”