In this paper an improved K-means algorithm is presented to cut character out of the license plate images. Although there are many existing commercial LPR systems, with poor illumination conditions and moving vehicle the accuracy impaired. After examination and comparison of different image segmentation approaches, the K-means algorithm based method gave better image segmentation results. The K-means algorithm was modified by introducing automatic cluster number determination by filtering SIFT key points. After modification it efficiently detects the local maxima that represent different clusters in the image. The process is successful by getting a clean license plate image. While testing by the OCR software, the experimental results show a high accuracy of image segmentation and significantly higher recognition rate. The recognition rate increased from about 86.6% before our proposed process to about 94.03% after all unwanted non-character areas are removed. Hence, the overall recognition accuracy of LPR was improved.