CNN for License Plate Motion Deblurring
This work addresses motion blur in license plate images for traffic surveillance systems, offering a practical, incremental improvement over existing methods.
The paper tackled license plate deblurring in traffic surveillance by training CNNs on artificially blurred data, achieving superior reconstruction quality on real images compared to traditional blind deconvolution methods, with results demonstrated through evaluation of blur direction and length.
In this work we explore the previously proposed approach of direct blind deconvolution and denoising with convolutional neural networks in a situation where the blur kernels are partially constrained. We focus on blurred images from a real-life traffic surveillance system, on which we, for the first time, demonstrate that neural networks trained on artificial data provide superior reconstruction quality on real images compared to traditional blind deconvolution methods. The training data is easy to obtain by blurring sharp photos from a target system with a very rough approximation of the expected blur kernels, thereby allowing custom CNNs to be trained for a specific application (image content and blur range). Additionally, we evaluate the behavior and limits of the CNNs with respect to blur direction range and length.