一种基于阈值分割的自适应逆光图像处理算法  

An Adaptive Backlight Image Processing Algorithm Based on Threshold Sementation

作　　者：田文启 李震 段新涛[1] 张润泽 TIAN Wenqi;LI Zhen;DUAN Xintao;ZHANG Runze(School of Computer and Information Engineering,Henan Normal University,Xinxiang 453007;School of Computer and Information Engineering,Zhengzhou Finance and Trade School,Zhengzhou 450000)

机构地区：[1]河南师范大学计算机与信息工程学院,新乡453007 [2]郑州市财贸学校,郑州450000

出　　处：《计算机与数字工程》2020年第10期2465-2470,共6页Computer & Digital Engineering

基　　金：河南师范大学国家大学生创新创业训练计划项目(编号:S201810476032);河南省高校重点科研项目(编号:16A520058)资助。

摘　　要：针对现有算法对非逆光区亮度提升过高所造成虚化的问题本文基于现有的自适应逆光处理算法,提出了一种基于阈值分割的自适应调逆光图像处理算法,并验证了Otsu算法对逆光图像分割的适用性。算法首先将图像转换到HSV色彩空间,采用Otsu分割算法将图像的亮度分量分为逆光区和非逆光区,对逆光区采用了改进型自适应对数变换,对非逆光区进行了非线性叠加亮度提升,通过调整亮度值来适应逆光处理后的逆光区图像达到很好的图像融合效果。该方法在保证逆光图像的恢复质量的前提下,解决了图像的虚化问题,且有效地减少了自适应逆光恢复算法的复杂度。In order to solve the problem of the existing algorithms that enhance the brightness of non backlight area too much.In this paper,based on the existing adaptive backlighting algorithm,an adaptive backlighting image processing algorithm based on threshold segmentation is proposed,and the applicability of Otsu algorithm for backlighting image segmentation is verified.Firstly,the algorithm transforms the image into HSV color space.Using Otsu segmentation algorithm,the brightness components of the image are divided into reverse and non-reverse regions.The improved adaptive logarithmic transform is used in the backlight region,and the simple non-linear superposition brightness lifting transform is used in the non-backlight region.By adjusting the brightness value to adapt to the image of the backlight area after backlight processing,a good image fusion effect can be achieved.On the premise of guaranteeing the quality of image restoration,this method solves the problem of image virtualization and effectively reduces the complexity of adaptive image algorithm.

关 键 词：阈值分割 图像增强 逆光恢复 信息熵 

分 类 号：TP317.4[自动化与计算机技术—计算机软件与理论]