摘要
针对雾霾天气下拍摄的图像对比度低、细节信息丢失、严重影响户外视觉系统分析工作等问题,利用基于暗原色先验的引导滤波算法对雾天图像进行初步复原,并通过改进经典的伽马变换和滚动引导滤波算法分别对图像对比度的增强和多尺度细节补偿问题开展研究。结果表明:所提方法不仅能够更好地复原场景结构信息、自适应提高图像对比度,同时还能够有效丰富图像多尺度纹理细节,主观视觉效果与客观评价指标验证了算法的有效性。
Outdoor images captured in smoggy weather are prone to have low contrast and lose details,which adversely affects outdoor visual system analysis.To solve these problems,a guided filtering algorithm based on dark channel prior was used to restore the initial defogging image.Furthermore,a study was conducted on enhancing image contrast and multi-scale detail compensation by the improved classical gamma transform and rolling-guided filtering algorithm,respectively.The experimental results show that the proposed method can recover the scene structure information,improve the image contrast adaptively,and enrich the multi-scale texture details effectively.The effectiveness of the algorithm is verified by subjective visual effect and objective evaluation index.
作者
陈耀欣
翟艺书
高智慧
刘月圆
CHEN Yaoxin;ZHAI Yishu;GAO Zhihui;LIU Yueyuan(School of Science,Tianjin University of Technology and Education,Tianjin 300222,China)
出处
《天津职业技术师范大学学报》
2023年第4期42-48,共7页
Journal of Tianjin University of Technology and Education
关键词
图像去雾
引导滤波
改进的伽马变换
滚动引导滤波
多尺度细节
image defogging
guided filtering
improved gamma transform
rolling guided filtering
multi-scale detail
作者简介
陈耀欣(1999—),女,硕士研究生,研究方向为图像去雾;通信作者:翟艺书(1979—),女,副教授,硕士生导师,研究方向为图像增强、图像去雾,51762041@qq.com.
