Automatic license plate recognition technology, as a key component of intelligent transport system, is playing a more and more important role in many application fields in our life. Vehicle license plate location, character segmentation and character recognition are three key steps of license plate recognition, of which the vehicle license plate location is the core part. The vehicle license plate location and character segmentation of complex background digital image are well researched. The main contributions of this thesis are as follows:1. A new vehicle license plate location algorithm based on multi-characteristics is introduced under the complex background vehicle image taken by highway digital cameral.This algorithm makes full use of the color, texture, geometric characteristics and locates the license plate by three steps: the raw locating of license plate, the incline angle detection and correction, the precise locating of license plate. The experimental result shows that this algorithm has a fast, efficient performance of locating vehicle license plate under the high-pixel complex background vehicle image taken by highway digital cameral.2. A new license plate incline angle detection algorithm base on radon transform is introduced . This algorithm first applies canny edge detection to the license plate region and gets a binary edge image of license plate region. Then projection of this binary edge through different angles is taken and carefully analyzed. The license plate incline angle is deduced from the results of the projection. The experimental result shows that this algorithm has a fast, efficient performance of vehicle license plate incline angle detection, regardless of whether the vehicle license plate has a clear boundaries or not.3. According to the rules of the characters of a vehicle license plate and the geometrical features of the characters, an approach for character segmentation of vehicle license plates is proposed which is based on the location of the second and third character. Firstly, the nail and horizontal boundary are removed by the horizontal projection of the license plate. Secondary the second and third characters of the plate are located in the image of the plate location. And then the character regions will be divided and merged. As a result, the characters of the plate are segmented accurately. The experimental result shows that this approach is a good approach for character segmentation of vehicle license plate, which can be accomplished with a grate result in a short time. |