基于自适应引导滤波的红外图像细节增强

Adaptive guided filtering based infrared image detail enhancement

为了解决原始红外图像动态范围高、图像对比度低、边缘模糊等问题,提出了一种基于自适应引导滤波的红外图像细节增强算法。首先,利用设计的参数自适应引导滤波将输入图像分成基础层和细节层。然后,利用平台直方图均衡化压缩基础层动态范围,利用掩膜加权和γ变换抑制细节层噪声并增强图像细节。最后,将处理后的基础层和细节层线性加权并拉伸到显示范围。实验结果表明,相比于传统和先进的红外图像细节增强算法,本文算法具有更强的场景适应性,在高动态范围场景、复杂室外场景以及海天场景都获得了更好的增强效果。相比于先进算法,本文算法在平均梯度上平均提高35.3%,在基于感知的图像质量评价指标上平均提高10.7%。本文算法可以有效压缩红外图像动态范围,显著提高图像对比度和清晰度,处理结果更有助于人眼观察。

To solve the problems of high dynamic range,low image contrast and blurred edges of original infrared images,an infrared image detail enhancement algorithm based on adaptive guided filtering is proposed.Firstly,the input image is divided into a base layer and a detail layer using the designed parameteradaptive guided filtering.Then,the plateau histogram equalization is used to compress the dynamic range of the base layer,and mask weighting and gamma transform are used to suppress detail layer noise and enhance the image details.Finally,the processed base and detail layers are linearly weighted and stretched to the display range.The experiment results show that compared with the traditional and advanced infrared image detail enhancement algorithms,the proposed algorithm has stronger scene adaptability,and has obtained better enhancement effects in scenes such as high dynamic range scene,complex outdoor,and sea and sky.Compared with the advanced algorithm,the proposed algorithm has an average improvement of 35.3%in the average gradient and an average of 10.7%in the perception-based image quality evaluator.The proposed algorithm can effectively compresses the dynamic range of infrared images,significantly improves image contrast and sharpness,and the processing results are more helpful for human eyes to observe.