融合暗通道先验与粒子群算法的去雾改进算法
Image Defogging Algorithm Based on Dark Channel Prior and Particle Swarm Optimization
田昊 1王小玉1
作者信息
- 1. 哈尔滨理工大学 计算机科学与技术学院,哈尔滨150006
- 折叠
摘要
为了解决雾霾条件下传统固定值暗通道先验算法导致的去雾图像质量低、颜色失真等问题,提出了一种暗通道先验与粒子群融合的去雾改进算法.利用粒子群优化算法的特性,对每个平均亮度范围内的保留因子进行优化,然后将其代入暗通道先验算法中.同时,在求解大气光值时,采用中值滤波算法替代原有的两次最小值滤波算法.实验结果表明,相较于传统固定值暗通道先验算法,所提算法在图像的去雾处理上既在主观视觉效果上有所提升,也在客观评价标准上具有更好的表现.同时,该算法的运行速度也提升了约14.3％.
Abstract
In the case of haze, aiming at the shortcomings of traditional fixed-value dark channel prior algorithm,such as low image quality and color distortion,a defogging improve ment algorithm integrating dark channel prior and particle swarm optimization is proposed.According to the characteristics of the particle swarm optimization algorithm, the best value of the retention factor in each average brightness range is optimized and brought into the dark channel prior algorithm.At the same time, the median filtering algorithm is used to replace the original two minimum filtering algorithms when solving the atmospheric light value.The experimental results show that the proposed algorithm has better subjective visual effects and objective evaluation criteria in image defogging compared with the traditional fixed value dark channel prior algorithm,and the operation speed of the algorithm is improved by about 14.3％.
关键词
图像去雾/暗通道先验算法/粒子群优化算法/中值滤波算法Key words
images-defogging/dark channel prior/particle swarm optimization/median filtering algorithm引用本文复制引用
出版年
2024
NETL
NSTL
万方数据