Disassembly is the first step for remanufacturing and recycling end-of-life (EoL) products. Advanced technologies such as robotics, artificial intelligence and machine vision, have been increasingly employed to achieve automated and smart disassembly. Machine vision systems enable robots with a capability to recognize and locate disassembly targets. In this paper, a 3D vision system has been investigated for the robotic disassembly of EoL products. A 3D depth camera was integrated into a collaborative robot. The coordinate system transformations between the camera and the robot were developed. Scene point clouds were captured to recognize and locate a target object to be dismantled. Experiments were carried out to validate the effectiveness of the 3D vision system.