摘要
为避免光照、天气以及物体本身反射性质等因素导致图像清晰度低、色彩饱差、细节信息缺失等问题,提出基于频率分解的平面图像色彩增强方法.首先将平面图像的RGB色彩空间转化成HSV空间,得到色度、饱和、亮度分量;然后利用自适应二维经验模式分解方法分解各个分量,得到区域Retinex入射光分量;最后使用Retinex算法计算照射分量,校正入射光图像、反射光图像,修正全局HSV色彩空间,从而实现平面图像色彩增强.实验结果证明,所提方法可获取详细信息的低频分量以及高频细节突出的高频分量;图像增强后的灰度直方图中每一个像素点在合理范围内分布,不会减少灰度层次,图像均衡化效果较好;增强后的平面图像色彩、清晰度、细节信息都得到很好的改善和保持.
In order to avoid the problems of low image definition,poor color saturation and lack of detail information caused by illumination,weather and the reflection property of the object itself,a color enhancement method of plane image based on frequency decomposition is proposed.Firstly,the RGB color space of the plane image is transformed into HSV space to obtain the chroma,saturation and brightness components.Then,each component is decomposed by adaptive two dimensional empirical mode decomposition method to obtain the regional Retinex incident light component.Finally,Retinex algorithm is used to calculate the illumination component,correct the incident light image and reflected light image,and correct the global HSV color space.The color enhancement of plane image is realized by this way.Experimental results show that the proposed method can obtain the low-frequency component of detailed information and the high frequency component with prominent high-frequency details.After image enhancement,each pixel in the gray histogram is distributed within a reasonable range,the gray level will not be reduced,and the image equalization effect is good.The color,clarity and detail information of the enhanced plane image have been well improved and maintained.
作者
胡奡
HU Ao(Department of Commerce and Electronic Information,Tongcheng Teachers College,Tongcheng Anhui 231400)
出处
《宁夏师范学院学报》
2021年第10期98-106,共9页
Journal of Ningxia Normal University
基金
安徽省质量工程项目(2020jxtd264)
安徽省质量工程项目(2020mooc502)
安徽省自然科学重点项目(KJ2020A0892)
安徽省质量工程项目(2020jyxm1996).
关键词
频率分解
平面图像
色彩空间
入射光
反射光
视觉特征
Frequency decomposition
Plane image
Color space
Incident light
Reflected light
Visual feature
