基于人眼视觉的自适应Riemann-Liouville分数阶红外图像增强.pdfVIP

中国科技论文在线 基于人眼视觉的自适应Riemann-Liouville 分数阶红外图像增强# 张凯，郝群，宋勇，胡摇** 5 （北京理工大学光电学院） 摘要：提出了一种基于人眼视觉特性的自适应Riemann-Liouville （R-L ）分数阶红外图像增 强算法。在R-L 分数阶微积分理论的基础上推导出R-L 分数阶差分公式，结合视网膜Rodieck 感受野模型以及图像局部统计特征与人眼结构感知特性的关系设计了阶次局部自适应可变 10 的R-L 分数阶Rodieck 增强模板，实现对红外图像的自适应纹理边缘增强；依据人眼的亮度 感知特性构造了非线性对比度提升模型调整增强结果的灰度分布，提升红外图像的对比度。 实验结果表明，该算法能有效地增强红外图像的纹理细节、抑制噪声干扰，增强结果的视觉 效果自然，更适合人眼观察。 关键词：红外图像处理；图像增强；分数阶微积分；人眼视觉特性；局部自适应 15 中图分类号：TP391 Adaptive Infrared image enhancement based on Human Visual Properties and Riemann-Liouville Fractional Differential 20 ZHANG Kai, HAO Qun, SONG Yong, HU Yao (School of Optoelectronics,Beijing Institute of Technology) Abstract: In order to make the infrared image more accessible for human eyes, an adaptive infrared image enhancement algorithm based on human visual properties and Riemann-Liouville (R-L) fractional differential is proposed in this paper. R-L fractional order differential equation is 25 deduced on the basis of R-L fractional calculus theory. An local adaptive R-L fractional order Rodieck enhancement mask is designed, combining of the mathematical model of human retinal receptive field and the relationship between local statistical features of the image and the human vision sensitivity properties. The mask is used to enhance the textures and edges of the image. Then, the grayscales of the enhancement result are adjusted using non-linear contrast optimize 30 model which is constructed according to the grayscale resolution capabilities of huma