A vehicle license plate recognition algorithm in night based on HSV

In night, it is difficult to locate vehicle license plate for vehicle recognition. In this paper, the algorithm uses colorful component in HSV space depending on the character of vehicle images in night, can divide the vehicle- plate region and many noises including vehicle- lamp and it´s reflection light and road- line, and can delete the region without similar color of vehicle plate but with higher saturation value.And the paper stresses the improvement on “maximum & minimum optimization of entropy thresholding method” as well as the prospects of regional character, the background distribution in the traditional character segmentation algorithm. The experimental result shows that the method is simple and fast, and it decreases the conglutination between the characters but also reduce the isolated noise point. The recognition quality of system is as good as what we expected.