摘要
书画印章图像识别过程中,没有对书画印章图像进行降噪处理,导致识别正确率较低,很难有效识别出图像信息,为此提出基于小波分析的书画印章图像识别方法。利用小波分析方法对书画印章图像进行降噪处理,根据5个形状特征参数,提取降噪处理后的书画印章图像纹理特征,并在小波神经网络分类器中实现对书画印章图像的识别。通过实验论证的方式进行对比和分析,实验结果表明,基于小波分析的识别方法对书画印章图像识别,其识别正确率均达到了90%以上,该方法可以更好地应用于书画印章图像识别领域。
In the process of calligraphy and painting seal image recognition, no noise reduction process is applied to the calligraphy and painting seal image, which leads to low recognition correct rate and difficult to recognize the image information effectively, for this reason, we propose a calligraphy and painting seal image recognition method based on wavelet analysis. By using wavelet analysis method to noise reduction processing of calligraphy and painting seal images, the texture features of calligraphy and painting seal images after noise reduction processing are extracted according to five shape feature parameters, and the recognition of calligraphy and painting seal images is realized in wavelet neural network classifier. The comparison and analysis are carried out by means of experimental demonstration, and the experimental results show that the recognition method based on wavelet analysis for calligraphy and painting seal image recognition has a correct recognition rate of more than 90% in all cases, and the method can be better applied to the field of calligraphy and painting seal image recognition.
作者
汪伊函
WANGYihan(Xidian University,Xi’an Shaanxi 710126,China)
出处
《信息与电脑》
2022年第14期89-91,共3页
Information & Computer
关键词
小波分析
书画印章
图像识别
识别正确率
wavelet analysis
calligraphy and painting seals
image recognition
recognition accuracy
