基于Contourlet变换和模糊理论的红外图像增强算法

红外图像具有噪声大、对比度低等特点，针对该特点，提出了一种基于Contourlet变换与模糊理论的红外图像增强算法。首先对图像进行Contourlet变换，得到多尺度多方向的低通子带和带通子带。对低通子带，进行基于子带系数最大最小值的线性变换，提高图像的整体对比度；对于带通子带，先估计噪声阈值，对子带系数进行抑制噪声处理，然后通过模糊增强算法，对高频系数进行非线性增强，增强目标边缘纹理的特征，抑制背景信号。最后经过Contourlet逆变换得到对比度增强，噪声被抑制的图像。经过算法仿真，与几种现有的图像增强算法相比，该算法更能有效地抑制噪声，增强图像的对比度，突出图像的边缘与细节纹理信息。

Novel scheme for infrared image enhancement based on contourlet transform and fuzzy theory

Low contrast and large noise are two main characteristics of infrared image.Considering this characteristics,a novel scheme for infrared image enhancement based on Contourlet transform and fuzzy operator is proposed.Firstly,Contourlet transform is performed on the original infrared image,we can obtain low-pass subband and band-pass subband at different scales and directions.At low-pass subband,the linear transformation based on the maximal and minimal coefficient of subband is applied to enhance the global contrast.At band-pass subband,the noise threshold is estimated and denoising is carried out,then the fuzzy nonlinear enhancement operator is used to enhance the high frequency subband coefficient,the objective characteristics of the edge texture is enhanced,but the background is suppressed.Finally,the inverse transform of Contourlet is applied to produce enhanced infrared image.Through computer simulation and compareing with other common image enhancement scheme,the proposed scheme can suppress noise,enhance contrast and highlight the edge of the image texture information and details effectively.