紧急求计算机毕业论文摘要翻译！急急急！要人翻译的

[智能交通灯毕业设计摘要翻译求高手翻译下]The ages is an automation ages nowadays and transportation light control's waiting a lot of equipmentses of professions all is closely related with calculator.T...+阅读

This paper describes the handwritten character recognition of the three processes, including image pre-processing, extraction and Signature BP neural network classifiers. First image of the pretreatment technologies, including image data read, the gray-scale image, the image of the value of the gradient sharpening, noise points separated the removal, images tilt adjustment, the characters split, in the one treatment, Tightening process, and so are followed by a comparative analysis of the structural characteristics of extraction methods and statistics feature extraction methods on the basis of directions given strokes feature extraction method, the extraction will feature coding as a neural network input to the character encoding as 8421 Output value of the network training, to be stable after the network as a knowledge base, character recognition and finally, under the direction of strokes feature extraction method designed to BP neural network classifiers, handwritten digit identification. Experimental results show that the strokes used in the direction of neural network coding method of handwriting in the application of digital identification is feasible. Finally completed the handwritten digital identification system can be used alone, can also identify it as a core system software application to license plate recognition system to.