2018/02/17 by P. S. Prashanth Rao, Rao, P. S. Prashanth, Muthu, Rajesh Kumar · 1 citation
Computer Science · Engineering · #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Image and Object Detection Techniques #Medical Image Segmentation Techniques #Vehicle License Plate Recognition
paper · pdf · doi:10.48550/arxiv.1802.06214
openalex publication_date 2018/02/17 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Recognizing a license plate clearly while seeing a surveillance camera\nsnapshot is often important in cases where the troublemaker vehicle(s) have to\nbe identified. In many real world situations, these images are blurred due to\nfast motion of the vehicle and cannot be recognized by the human eye. For this\nkind of blurring, the kernel involved can be said to be a linear uniform\nconvolution described by its angle and length. We propose a new de-blurring\ntechnique in this paper to parametrically estimate the kernel as accurately as\npossible with emphasis on the length estimation process. We use a technique\nwhich employs Hough transform in estimating the kernel angle. To accurately\nestimate the kernel length, a novel approach using the cepstral transform is\nintroduced. We compare the de-blurred results obtained using our scheme with\nthose of other recently introduced blind de-blurring techniques. The\ncomparisons corroborate that our scheme can remove a large blur from the image\ncaptured by the camera to recover vital semantic information about the license\nplate.\n