A license plate recognition system can be divided into three steps: license plate detection, character segmentation and character recognition. License plate detection will be inclined, light, pollution and so on, algorithm complexity. In recent years, put forward a lot of license plate detection algorithm, but most of them are inadequate, not in a complex environment, achieve a good license plate recognition. Edge detection algorithm with filming the impact of changes, when there is no clear license plate edge can be detected in the plate. Combination of color and texture feature detection accuracy is better, but by the influence of illumination is very obvious, and the algorithm complexity is high, no effect. The morphology of license plate detection algorithm, it is easy to be affected by class of license plate character effects, detection speed is slow. These detection algorithm has shortcomings because of lack of adaptation algorithm. This paper presents the Boosting algorithm in the AdaBoost algorithm, this is a kind of structure accurately classifier adaptive learning algorithm, it combines many features to construct accurate classifier, finally complete license plate detection.
This paper studies the key technologies in license plate recognition, license plate location, introduces in detail the character segmentation, feature extraction and recognition method of license plate location, and with the AdaBoost algorithm based on license plate recognition is discussed, the results show that the AdaBoost algorithm can evidently improve the stability and precision of license plate detection of.（望采纳）
嘿嘿～～