Open Access Open Access  Restricted Access Subscription or Fee Access

Vanity Plate Image Estimation of Moving Vehicles

Mayuri Vijay Sakure, Supriya O. Rajankar



License plate recognition of automobile is concern and challenging exploration discussion thread from past little duration. It is a very important field in intelligent transportation systems (ITS). Many times, it would be very hard to recognize the proprietor of an automobile who disregards the commercial transportation law and is moving with a fast speed. The motto of license plate recognition (LPR) systems is to trace, segment and identify the vanity plate from the embodiment of the automobile. As notable familiar evidence, registration code of automobile is a solution to reveal over-speed vehicles or the ones included in attempt at assassination hit and run accidents. The snap of over-speed automobile trap by supervision camera is misty and indistinguishable because of speedy movement, which is not recognizable by individual. The extracted vanity plate snapshot which is more often low resolution and endures serious loss of edge data, which radiate a complexity to already illustrated de-blurring techniques. In vanity plate picture, vague is formed by fast movement; the unclear bit can be seen as direct uniform convolution and parametrically demonstrated with edge and length. Nevertheless, this paper gave an idea stand on sparse representation to distinguish the vague section. By probing sparsely represented coefficients of the obscure snapshot, an angle is estimated. After that length is assessed from indistinguishable part of the vanity plate by taking Radon transform in Fourier space. The method explained in this paper can handle large motion blurs even when the vanity plate is unrecognizable by individual.

Keywords: Kernel parameter estimation, license plate de-blurring, linear motion blur, sparse representation

Cite this Article

Mayuri Sakure, Rajankar Supriya O. Vanity Plate Image Estimation of Moving Vehicles. Journal of Open Source Developments. 2017; 4(2): 19–27p.

Full Text:



  • There are currently no refbacks.