作者:张兆茹，邓彩霞，岳欣华
Authors:ZHANG Zhao-ru,DENG Cai-xia,YUE Xin-hua摘要:摘要:为了使指纹识别技术更加准确和识别效率更高，提出了改进的阈值函数对指纹图像进行去噪。首先针对指纹图像的特点，构造了一个改进的阈值函数，该阈值函数与传统软、硬阈值函数及某些现有改进阈值函数相比，具有很好的可调性，并且是处处可导的，其更好地逼近软阈值函数，且在阈值点处的图像更加光滑，在对图像去噪时能够更多的保留其上的真实信息，同时有效的滤除噪声。仿真实验表明，改进的阈值函数处理后的图像峰值信噪比和结构相似度高、均方根误差和扭曲程度小且相关系数大，能够更加接近于原图像，因此改进的阈值函数具有很好的应用价值。
Abstract:Abstract:In order to make fingerprint identification technology more accurate and efficient, an improved threshold function is proposed to denoise fingerprint images.First, according to the characteristics of fingerprint images, an improved threshold function is constructed.Compared with traditional soft and hard threshold functions and some existing improved threshold functions, this threshold function has good adjustability, and is adjustable everywhere. It is better to approximate the soft threshold function, and the image at the threshold point is smoother.When denoising the image, it can retain more real information on it and effectively filter out the noise-Simulation experiments show that the improved threshold function has high peak signal-to-noise ratio and structure similarity, small root mean square error and distortion degree, and better correlation coefficient, and can be closer to the original image.Therefore, the improved threshold function has good application value.

PDF全文下载地址:

可免费Download/下载PDF全文