融合暗通道与直方图均衡化的去雾算法及应用
Image Defogging Algorithm Based on Dark Channel Prior and Adaptive Histogram Equalization and Applications
韩超 1房建东 1赵于东2
作者信息
- 1. 内蒙古工业大学信息工程学院 呼和浩特 010080;内蒙古自治区感知技术与智能系统重点实验室 呼和浩特 010080
- 2. 内蒙古自治区感知技术与智能系统重点实验室 呼和浩特 010080
- 折叠
摘要
针对可见光图像在采集过程,由于环境温差的变化导致镜头起雾所引起采集的可见光图像清晰度下降的情况,设计一种暗通道先验和自适应直方图均衡化生成图像进行加权融合的去雾算法,并将其用于植物采收机器人的去雾复原处理上.使用暗通道先验理论,得到大气光强和透射率,完成暗通道方法的去雾复原处理,再对原图进行限制对比度的自适应直方图均衡化,将两幅图像进行加权融合,使得生成的图像更加清晰.实验结果表明,该方法能够有效地完成图像去雾工作,与经典算法相比能得到更多的信息,图像更加清晰可见,具备一定的优异性.
Abstract
To address the situation that the clarity of the acquired visible images is degraded due to the fogging of the lens caused by the change of ambient temperature during the acquisition process of visible images,a defogging algorithm with dark chan-nel prior and adaptive histogram equalization to generate images for weighted fusion is designed and applied to the defogging recov-ery process of the plant harvesting robot.By using the dark channel a priori theory,the atmospheric light intensity and transmittance are obtained to complete the defogging recovery process by the dark channel method,and then the original image is subjected to adaptive histogram equalization with restricted contrast,and the two images are weighted and fused to make the generated image clearer.The experimental results show that the method can effectively complete the image defogging work,and get more information compared with the classical algorithm,the images are clear and visible,and has certain superiority.
关键词
暗通道先验/图像去雾/图像加权融合/自适应直方图均衡化Key words
dark channel prior/image dehazing/image weighted fusion/adaptive histogram equalization引用本文复制引用
基金项目
内蒙古自治区自然科学基金项目(2019MS06023)
内蒙古自治区应用技术研究与开发资金计划项目(2019GG334)
内蒙古自治区应用技术研究与开发资金计划项目(2019GG337)
出版年
2024
国家科技期刊平台
NSTL
万方数据