基于剪切波域改进Gamma校正的医学图像增强算法

Medical image enhancement method based on improved Gamma correction in Shearlet domain

为了解决医学图像在采集和传输过程中引入噪声和干扰导致图像质量恶化从而严重影响医学诊断的问题,提出一种基于剪切波(shearlet)域改进Gamma校正的图像增强方法。首先,通过剪切波变换,把图像分解成高频部分和低频部分;其次,用改进的Gamma校正处理剪切波分解后的低频部分以调整图像的整体对比度,采用改进的自适应阈值函数对高频部分进行去噪;最后,把剪切波反变换的重构图像进行模糊对比增强,以突出图像的细节信息。实验结果表明,本文算法的峰值信噪比(PSNR)、结构相似度(SSIM)和绝对均值差(MAE)优于其他对比算法,尤其是PSNR的提升更加明显。这些客观指标说明,本文算法不仅能有效地抑制噪声,而且能明显改善增强对比度。从主观方面观察,本文算法与其他算法相比,能获得更好的视觉效果。

Noises and artifacts are introduced to medical images in the process of acquisit ion and transmission,which causes image degradation and further seriously affect s the clinical diagnoses.Therefore,in order to solve this problem,a medical imag es enhancement method based on shearlet transformation and improved gamma correctio n is proposed in this paper.First,the original image is decomposed into the shearlet domain with low-frequency component and high-frequency component.Then,the improved gamma correction is adopted for the low-frequency component to improve the global contrast of the image,and an adap tive threshold method is used for the removal of high-frequency image noise.Finally,the improved fu zzy contrast is used to enhance the details of the reconstruct image which is obtained by the sh earlet inverse transformation.The experimental results show that the proposed method is super ior to other comparative methods in peak signal to noise ratio （PSNR） ,structural similarity （SSIM） and mean absolute error （MAE）,especially the PSNR is increased observably.These objective criterions indicate that the proposed method not only can remove image noise efficiently, but also significantly improve the contrast of image.In the subjective aspect, the proposed method can get better visual effect than other methods.