英语语法分析

In this paper, the word out of vocabulary and misrecognition are rejected to cut down the recognition faults, improve the system recognition ratio and reduce the false acceptance ratio.

本论文针对词库外的词汇和错误识别进行拒识，以期达到减少识别错误，提高系统的识别率，降低虚警率的目的。

The problems of reading defect characters of images and misrecognition of similar characters can be solved by a kind of sifting-connection algorithm and the application of BP neural network technique.

提出过筛连通法并结合运用BP神经网络技术，解决了读数图像缺陷字符和相似字符的误识问题。