In this paper, a practical method of stereo vision, “subtraction stereo” is proposed. A huge number of studies have been carried out for stereo vision until now, and several practical stereo vision systems have been reported. However, what is called the correspondence problem that stereo matching becomes difficult and not robust for weak textures or recurrent patterns is inevitable for stereo vision. Subtraction stereo realizes robust measurement of range images by detecting moving regions with each camera first and then applying stereo matching for the detected moving regions. Detection of moving regions is carried out with a subtraction process. Concept and fundamental algorithm of subtraction stereo are introduced. Then measurement of three-dimensional position, height and width of a target object using the subtraction stereo is discussed. The basic algorithm is implemented on a commercially available stereo camera and the effectiveness of the subtraction stereo is verified by several experiments using the stereo camera. Although objects are restricted to moving ones, subtraction stereo gives sufficient information robustly for many applications such as surveillance.