An Efficient Method Based on Orientation Field for Detection of License Plates

License plate area contains rich texture and edge information. In our method, the car image first is divided into set of 5×5 non-overlapping black and a local orientation is defined for each black. The M×N car image is converted to an orientation image and a vertical edges image of size (M×N)/25. It reduces much processing time. The orientation and the vertical edges of black are calculated using Sobel operator. We remove most of the background and noise regions in the orientation and vertical edges image by an effective algorithm and segment the plate out from the original car image. Experimental results demonstrate the great robustness and efficiency of our method.