论文部分内容阅读
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。
This paper presents an overall implementation scheme of single digital character recognition algorithm based on BP neural network to realize the collection, storage, identification and display of single digital character images. According to the principle of BP neural network, the recognition algorithm is divided into working-period algorithm and learning-period algorithm. VC ++ 6.0 software is used to design and implement the algorithm respectively using VC language. The system’s display interface is designed with MFC, Screen to see the character recognition results. The results show that using this algorithm, the recognition rate of single-sub-image is above 90%, and the more learning times, the higher the recognition rate. The single digital character recognition algorithm based on BP neural network can be widely used in many fields such as digital instruments, license plate recognition, satellite positioning and so on, and has certain application value.