基于数字图像处理的字迹识别技术探讨

Study of Handwriting Recognition Technology Based on Digital Image Processing

针对传统字迹识别方法中存在的识别精度低、识别耗时长等问题,提出基于数字图像处理的字迹识别方法.采用领域平均法对字迹图像进行降噪处理,将字迹图像进行多尺度小波分解,提取字迹图像在不同尺度下的小波系数;通过平滑函数平滑后获取字迹图像的幅角,并对字迹图像进行拉伸处理,完成图像的增强;在采用数字图像处理技术对字迹进行识别时,通过灰度统计直方图对字迹图像进行灰度处理;采用归一化方法对字迹特征进行提取,通过傅里叶变换方法对字迹图像的频谱、相位角以及功率谱进行识别,完成基于数字图像处理的字迹识别.仿真结果表明:采用本文所提方法对字迹进行识别的精度可达98%,且识别速度较快,具有一定可行性.

In order to solve the problems of low recognition accuracy and long recognition time in traditional handwriting recognition methods,a new method based on digital image processing is proposed.The domain average method is used to denoise the handwriting image,the multi-scale wavelet decomposition is used to extract the wavelet coefficients of the handwriting image at different scales,the width of the handwriting image is obtained after smoothing with the smoothing function,and the handwriting image is stretched to complete the image enhancement.When the digital image processing technology is used to recognize the handwriting,the gray level is used,and the statistical histogram is used to process the gray scale of the handwriting image.The normalization method is used to extract the handwriting features,and the Fourier transform method is used to identify the spectrum,phase angle and power spectrum of the handwriting image to complete the handwriting recognition based on the digital image processing.The simulation results show that the accuracy of the proposed method is up to 98%,and the recognition speed is fast,which is feasible.