一种基于双边滤波的红外图像细节增强方法

摘要

When viewing on the conventional monitor,high dynamic range infrared images always appeare low con-trast,blurred details.To solve this problem,a method of infrared image nonlinear detail enhancement and dynamic range compression based on bilateral filter was proposed.Firstly a statistic method based on mean value and standard deviation of the image scan lines was used to detect the defective lines and restore them by interpolation.Then the high dynamic range infrared image was decomposed into base component and detail component by using bilateral fil-ter.The base component has a wide dynamic range,so its dynamic range was compressed.The detail component was enhanced by using a nonlinear gain function.Finally the base component and the detail component were recombined and quantized to 8 -bit low dynamic range image for display.The experimental results show that this method can com-press the dynamic range of the image and can effectively enhance the detail information of the image.%为了解决高动态红外图像在常规显示设备上显示时对比度低、细节模糊等问题，提出了一种基于双边滤波的红外图像非线性细节增强及动态范围压缩方法。首先利用基于扫描行的均值及标准差的统计方法对图像中的条纹噪声进行检测及插值补偿。然后对高动态红外图像进行双边滤波，得到基频分量及细节分量。针对基频分量动态范围大的特点，压缩其动态范围，对细节分量则进行非线性增强。最后将压缩后的基频分量和非线性增强后的细节分量合成，并将灰度范围调整到8位，得到适合显示的低动态范围图像。实验结果表明，提出的方法能在压缩图像动态范围的同时有效实现图像细节信息的增强。