Design and Enhancement of License Plate Images Based on Kernel Estimation using Adaptive Filter

Abstract

The speed of vehicles is uncovered because of hit and run accidents are occurred. Generally, the fast moving vehicle image is captured by the surveillance camera. The images that are observed by this camera consist of low resolution and the image will be in the blur format. Because of this the information will be loss. In this paper, to overcome this issue with the design and enhancement of license plate images based on kernel estimation using adaptive filter. Here the information patches are selected from the given images. From these images the edge prediction is performed. It means here it will determine the angle and length of the observations. After this kernel estimation operation is performed. Hence the proposed system can evaluate the images of real world and handle the motion of images when the license plate is unrecognizable. At last the proposed system gives effective output compared to other systems.