Extracting the VLP number from a VLP image is considered to be one of the most important stages of vehicle license plate recognition (VLPR) system, it greatly influence the overall recognition accuracy and processing speed of the whole system. This paper presents an algorithm to extract the VLP number from the VLPs. The distorted VLPs are also studied to develop a transformation to calibrating the distorted VLPs into normalized VLPs. Using the distortion calibration technique, the VLP number in a distorted state can also be extracted more correctly. The experiment results show that the presented algorithm can effectively extract the VLP number even from a distorted VLP.