摘要
利用CNN处理器的并行图像处理原理和通用编程的理论,提出基于CNN二值图像字母识别的通用编程方法,说明了用该方法作实时图像处理与识别的操作过程,解决了传统计算机串行数据处理的瓶颈问题.
Principles of parallel image processing and universal programming via CNN universal machine are introduced at first. Then an example of letter recognition via CNN universal programming is given so as to show its ability of real time image processing and recognition. It is proved to be a good alternative for serial data processing of traditional computer.
出处
《集美大学学报(自然科学版)》
CAS
2006年第2期142-145,共4页
Journal of Jimei University:Natural Science
基金
福建省自然科学基金资助项目(A0210019)
关键词
细胞神经网络(CNN)
通用编程
字母识别
cellular neural networks (CNN)
universal programming
letter recognition