摘要
针对传统方法在增强低对比度指纹图像时,效果不佳的状况,研究了一种利用快速曲波变换增强低对比度指纹图像的方法。图像经曲波分解后,对于低频分量引入基于SMQT的非线性函数调整低频子带系数,以提高指纹图像的对比度;对于高频分量使用阈值去噪增强图像细节。实验表明,该方法优于直方图均衡化方法和基于小波变换的增强方法。该算法在增强低对比度指纹图像的同时降低了噪声。增强后的图像取得了较为理想的视觉效果。
To overcome the shortcoming of the traditional method used to improve the low contrast of fingerprint images,a novel method is proposed for enhancing the low contrast of fingerprint images based on Fast Discrete Curvelet Transforms（FDCT） . After the input image is decomposed,to enhance the contrast of the fingerprint image,this paper uses nonlinear function based on Successive Mean Quantization Transform（SMQT） to adjust the coefficient of low frequency sub-band in low frequency component. It uses threshold denoising to enhance the details of the image in high frequency component. The simulation results show that the proposed method is better than histogram equalization and the method based on wavelet transform. The proposed algorithm enhances the contrast of low contrast fingerprint image and reduces noise at the same time. The images have satisfactory visual quality after enhancement.
出处
《计算机系统应用》
2010年第4期58-61,共4页
Computer Systems & Applications
基金
国家自然科学基金(60574051)
