摘要
为了提高低照度图像的可视性和清晰度,提出一种基于梯度策略的DCT域低照度图像增强算法。该算法将图像分为入射分量和反射分量,建立灰度线性增强模型,对入射分量进行灰度线性增强;分别将入射分量和反射分量分别分块同时变换到离散余弦变换域(Discrete Co-sine Trans-form,DCT)中,进行DCT多聚焦融合,这里采用的是局部梯度融合准则,而不是目前常用的基于方差的融合策略,并采用一致性判别对融合的图像进行校验。最后将融合结果进行逆DCT变换得到最终增强的图像。实验结果表明,该算法对于低照度图像阴暗区域的细节可视性有明显的提升效果。
In order to improve the visibility and clarity of low illumination images,a low illumination image enhancement algorithm based on gradient strategy in DCT domain is proposed.The image is divided into the incident component and the reflected component,and the grayscale linear enhancement model is established to linearly increase the incident component.The incident component and the reflected component are respectively divided into blocks and simultaneously transformed into the discrete cosine transform domain DCT,DCT multi-focus fusion is performed.The local gradient fusion criterion is used here instead of the commonly used variance-based fusion strategy,and the fusion image is verified by consistency discrimination.The fusion result is inverse DCT transformed to obtain the final enhanced image.The experimental results show that the algorithm has obvious improvement effect on the detail visibility of the dark area of low illumination image.
作者
谢有庆
陈泽辉
黎建健
周瑞鹏
黄浩衡
何涛
XIE Youqing;CHEN Zehui;LI Jianjian;ZHOU Ruipeng;HUANG Haoheng;HE Tao(Shenzhen Nanshan Power Supply Bureau Co.,Ltd.,Shenzhen 518000,China)
出处
《电子设计工程》
2021年第2期168-173,178,共7页
Electronic Design Engineering
基金
深圳供电局有限公司项目(090000KK52180035)。
关键词
梯度
线性增强
DCT域
多聚焦融合
可视性
gradient
linear enhancement
DCT domain
multi-focus
fusion visibility
