Brief analysis of common statistical pattern recognition methods Structural pattern recognition is the main method of early Chinese character recognition research. The main starting point is the composition of Chinese characters. In terms of the composition of Chinese characters, Chinese characters are composed of strokes (dots, etc.) and radicals; it can also be considered that Chinese characters are composed of smaller structural primitives. These structural primitives and their mutual relations can accurately describe Chinese characters, just as an article is composed of words, words, phrases and sentences according to grammatical rules. So this method is also called syntactic pattern recognition. When recognizing, use the above structural information and syntactic analysis method to recognize, similar to a logical reasoner. The statistical pattern recognition of Chinese characters regards the character lattice as a whole, and the features used are obtained from this whole through a lot of statistics. The characteristics of statistical features are strong anti-interference, and the matching and classification algorithms are simple and easy to implement. The disadvantage is that the subdivision ability is weak, and the ability to distinguish similar words is weaker. Common statistical pattern recognition methods are: (1) Use the method of transforming features. Carrying out binary transformation (such as Walsh, Hardama transformation) or more complicated transformation (such as Karhunen-Loeve, Fourier, Cosine, Slant transformation, etc.) on the character image greatly reduces the dimension of the transformed feature. However, these transformations are not rotation-invariant, so there is a large deviation in the recognition of obliquely deformed characters. Although the calculation of binary transformation is simple, the transformed features have no obvious physical meaning. Although the KL transform is optimal from the perspective of the minimum mean square error, the amount of calculation is too large to be practical. In short, the computational complexity of transforming features is relatively high and has certain weaknesses. (2) Template matching. Template matching does not require a feature extraction process. The image of the character is directly used as the feature. Compared with the template in the dictionary, the template class with the highest similarity is the recognition result. This method is simple and easy, and can be processed in parallel; however, a template can only recognize characters of the same size and the same font, and has no good adaptability to slanting and strokes becoming thicker and thinner. Alcohol Sketch Markers Pens,24Color Alchohol Markers,Skin Marker Pens,Acrylic Paint Marker Pens Zhoushan Shenglan Trade Co., Ltd.  , https://www.seamiart.com