A method to enhance the performance of stereo matching is presented. The position of the specular light reflection on an object surface varies due to the change in the position of the camera, light source, object or all combined. Additionally, there may be situations exhibiting a colour shift owing to a change in the light source chromaticity or camera white balance settings. These variations cause misleading results when stereo matching algorithms are applied. In this reported work, a single-imagebased statistical method is used to normalise source images. This process effectively eliminates non-saturated specularities regardless of their positions on the object. The effect of specularity removal is tested on stereo image pairs.Introduction: Stereo matching is an active research area and a vast number of algorithms for finding correspondences in multi-frame images have been presented. A comprehensive performance analysis of these methods is given by Sharstein and Szeliski [1].The main goal of stereo methods is to generate a disparity map from image pairs. Specular reflections, especially in cases where the position of the specular reflection changes drastically, cause unwanted results. This position change is mostly due to a change in the relative orientation of the object surface with respect to the camera. The light source position may also change independently.In [2], a stereo-enhanced colour processing method which deals with the unwanted effect of specular variations on surfaces is given by Lin et al. This method uses an image sequence to remove specularity with a strict assumption that all scene points present diffuse reflection in at least one of the views.In [3], Bhat and Nayar embedded a physically based model of reflected light into a stereo matching method to obtain a better stereo matching result in the specular regions of images, but their analysis required controlled lighting conditions and fully calibrated camera positions.In this Letter, the proposed method does not have restrictive assumptions. Furthermore, it does not require a complicated experimental setup. The only assumption in the method proposed is that the pixel values in the images are not saturated. This assumption holds in most cases where exposure settings avoid saturation and objects do not contain mirror-like highly reflective parts.