医学图像增强系统设计

为了改善医学图像的视觉效果,提高图像的清晰度,使之更适合于机器的分析处理以及人的视觉特性,并突出病灶点,为病理学诊断和临床诊断提供可靠依据。设计了一个对医学图像十分具有针对性的图像增强系统。针对CT图像的电子噪声提出了基于修正维纳滤波的小波包去噪算法;针对B型超声图像的散斑噪声提出了基于脉冲耦合神经网络(PCNN)模型的小波自适应斑点噪声滤除算法;针对医学图像对比度低,边缘信息模糊等特点,提出了基于小波变换的医学图像增强算法。当噪声方差为0.01时,基于脉冲耦合神经网络(PCNN)模型的小波自适应斑点噪声滤除算法获得的PSNR比经Wiener滤波方法获得的PSNR高出9 dB。系统能快速找到噪声点进行定点去噪,能有效提高医学图像的对比度,增强边缘细节信息,突出病灶点的位置,从而达到较好的处理效果,为医疗工作者观察病症提供更加清晰准确的依据。

System design of medical image enhancement

In order to improve the visual effects of the medical image and improve image clarity,a targeted image enhancement system for medical images is designed. For the electronic noise of CT images,a wavelet packet denoising algorithm based on amendments Wiener filtering is proposed; For speckle noise of B-mode ultrasound image,a wavelet adaptive speckle reduction algorithm based on the pulse coupled neural network (PCNN)model is proposed. For medical images with low contrast and blur edge,a wavelet-based medical image enhancement algorithm is proposed. The peak signal to noise ratio (PSNR)from the PCNN method is 9dB higher than that from the Wiener filtering when the noise variance is 0.01. Noise points can be quickly found,and fixed point denoising is carried out,the medical image contrast can be effectively improved,edge details are enhanced,the location of the lesion point is highlighted. It provides a more clear and accurate basis for symptoms observing of health workers.