A rapid stereo matching algorithm based on disparity interpolation

Aiming at the constraint of consistency and smoothness in stereo matching, a rapid stereo matching algorithm based on disparity interpolation is proposed in this paper. Firstly, a gradient based cost function of matching similarity is defined to ensure the consistency of image gray pixels and the smooth in the neighborhood; then Lorentzian function is adopted to reduce the effects of the noise in the stereo matching process; secondly, the disparity and the matching cost are calculated based on stereo matching result by using the left and right image as the base image, the disparity in this two kind of situations are also optimized, then the matching area and un-matching area can be further marked; lastly, the disparity interpolation process is done started from the edge of un-matching area and stop until all the pixel in the un-matching area are finished, with the dense and smooth disparity obtained. Experimental results and analysis shows that the dense and smooth disparity can be required and the time complexity can be reduced, and the effect of the new algorithm is better than traditional stereo matching algorithm.