This article describes the configuration and the implementation of a system that servos a hand-eye robot using visual information from two cameras. Using vision for robot operation presents several issues with respect to sensing, control, and system configuration. Some of these issues are presented in this paper in detail, such as image segmentation, recognition, calibration and visual servoing. Especially, We combine the edge detection and color segmentation to segment images. This method can extract the object from background in images efficiently. Finally, we present experimental results showing the application of the hand-eye robot