Automatic and flexible geometry measurement of the weld pool surface can help better understand the complex welding processes and even provide feedback to better control this process. Most of existing imaging systems use an additional source of illumination to remove the light interference coming from the welding arc but it is usually costly. This paper introduces a novel low-cost optical-sensor-based monitoring system working under passive mode to monitor the wire + arc additive manufacture process, particularly for plasma arc welding. Initially, configurations and parameters of camera are investigated to achieve good visualization of weld pool. A novel camera calibration methodology using the nozzle of a computer numerical control (CNC) machine is then proposed for this imaging system, allowing estimation of the camera position with respect to the inspecting surface and its orientation in an easy-to-use approach. The verification tests show that the average error of the calibration is less than 1 pixel. As a case study, an image analysis routine is proposed to measure the width of the bead during the welding process. The results show that the proposed system is effective to measure the dimension of weld pool. Index Terms-Additive manufacturing (AM), camera calibration, edge detection, plasma arc welding (PAW), wire + arc additive manufacture (WAAM). I. INTRODUCTION A DDITIVE manufacturing (AM) is a process of threedimensional (3-D) printing in which parts are built by adding materials layer by layer. Compared to traditional subtrac