基于BP神经网络的单数字字符识别算法实现

提出一个BP神经网络的单数字字符识别算法的总体实现方案,实现单数字字符图像的采集、储存、识别和显示。根据BP神经网络的原理,把识别算法分成工作期算法和学习期算法,并采用VC 6.0软件,运用VC语言分别设计并实现其算法,用MFC设计了系统的显示界面,最终可以在液晶屏幕上看到字符的识别结果。结果显示运用该算法,单数子图像识别率在90%以上,而且学习次数越多,识别率越高。基于BP神经网络的单数字字符识别算法可以广泛应用到数字仪表、车牌识别、卫星定位等很多领域,具有一定的应用价值。

Recognition Algorithm of Single Numeric Character Based on BP Neural Network

Propose an overall realized project that recognition algorithm of single numeric character based on BP neural network, realizes single numeric character image collection, storage, recognition and display. According to the BP neural network's principle,the recognition algorithm is divided into the work algorithm and the study algorithm. Use the software of VC 6.0, respectively design and realize the al-gorithms with the VC language, design system's display interface with MFC. Finally the character recognition results display on the liquid crystal screen. The result showed that utilizes this algorithm, the recognition rate of single numeric character image is above 90%, moreover the more times of study are, the higher recognition rate is. Recognition algorithm of single numeric character based on BP neural network is widely applied in the digital instrument, the car license recognition, the satellite positioning and so on many field, and it has certain appli-cation prospect.