基于神经网络和DSP技术的离线数字识别系统

本文提出了一种基于神经网络的离线数字识别技术,采用神经网络来对图中的数字进行识别,在分析图像预处理对图像特征提取和识别影响的基础上,提出了一种基于目标的图像增强算法,该算法计算量小、实时性好,有效地解决了图像背景颜色差异和环境光线差异对图像特征提取和识别造成的影响。最后给出了系统基于DSP的硬件实现。

Offline digital recognition system based on neural network and DSP

The paper gives up a kind of offline digital recognition technology based on neural network, neural network was used to recognize digits in this project, a target enhancement algorithm is put forward in this project. It is proved that it is much easier to extract image feature by employing target enhancement algorithm.