---

Detection of defects in the internal surfaces of pipes due to the inherent feature of these surfaces which is inaccessibility is always a troublesome process. In this study, a novel method has been designed for detection of defect locations on the internal surfaces of pipes and accurate estimation of defect geometrical parameters such as length and average material loss depth. The way that this method works is that a band heater is located on the external surface of the pipe in inspection segment, and specified heat flux is applied to this surface for a short time, and the temperature of these sensors located on rear of the band heater is measured during and after of applying thermal heat flux. The local temperature rise on the section of the external surface of the pipe indicates a defect on its internal surface. In this case, a defect with unknown parameters is supposed on the internal surface of the pipe, and by using the inverse heat conduction method, an iterative numerical simulation procedure continues until the unknown geometrical parameters of defects are estimated in a way to minimize the difference between the measured and simulated temperature in the location of sensors.