摘要
针对当前图像增强算法中灰度归一化重新分配时产生均值漂移,难以有效保留图像细节信息和亮度保护等问题,提出了一种基于PSO约束优化与直方图均衡化的图像增强算法。引入最大类间方差法(Otsu),将输入图像分割为目标子图像和背景子图像两部分,并对二者分别均衡化,以提高目标与背景子图像的对比度;并根据阈值加权约束分别计算目标和背景子图像的约束;并引入粒子群优化(PSO)算法,对目标和背景子图像的约束主要参数进行优化,确定最优约束值;联合目标子图像与背景子图像的最优约束值完成图像增强。实验结果表明,与当前增强方法相比,所提出的算法具有更好的亮度保护和对比度增强效果,较好地保留了输入图像的细节信息。
Since the mean shift may generate in the current image enhancement algorithm when the grayscale normalizationis redistributed,and it is difficult to effectively preserve the image detailed information and protect the brightness,a new imageenhancement algorithm based on PSO constraint optimization and histogram equalization is proposed. The Otsu method is intro?duced to segment the input image into the object subimage and background subimage,and the two subimages are respectivelybalanced to improve the contrast of object subimage and background subimage. And then the constraints of the object subimageand background subimage are computed according to the threshold weighted constraint. In this paper,the particle swarm optimi?zation(PSO)algorithm is introduced to optimize the main constraint parameters of object subimage and background subimage,and then determine the optimal constraints. The image was enhanced in combination with the optimal constraint value of the ob?ject subimage and background subimage. The experiment results show that,in comparison with the available image enhancementmethod,the proposed algorithm has better brightness protection and contrast enhancement effect,and can greatly preserve thedetailed information of input image.
作者
周冰
李聪
邓娟
ZHOU Bing;LI Cong;DENG Juan(City College,Wuhan University of Science and Technology,Wuhan 430000,China)
出处
《现代电子技术》
北大核心
2016年第15期32-37,共6页
Modern Electronics Technique
基金
湖北省教育科学"十二五"规划2014年度课题(2014A080)
2014年国家级大学生创新创业训练计划项目(201413235003)
武汉科技大学城市学院教研项目(2014CYYBJY002)
