A Comparative Study of Correlation Based Stereo Matching Algorithms: Illumination and Exposure

Stereo matching is one of the most active research areas in computer vision for an accurate estimation of disparity. Many algorithms for computing stereo algorithm have been proposed, but there has been very little work on experimentally evaluating algorithm performance, especially using real-time scenario. Many researchers have been undergone past from many decades to find an accurate disparity, but still it is not an easy task to choose an appropriate algorithm for the required real-time application. To overcome from this problem, we proposed an experimental comparison of several different cross-correlation-based stereo algorithms and also introduce an objective that evaluates a set of six known correlation-based stereo algorithms. An evaluation of correlation-based stereo matching algorithm results will be very useful for selecting the appropriate stereo algorithms for a given application. Here, we make use of two stereo pairs: Aloe and Cloth from Middlebury stereo datasets. This work mainly focuses on the evaluation of robustness to change in illumination and exposure.