摘要
为解决有雾环境下由于图像退化造成的视觉SLAM系统的稳定性问题,对含雾连续图像的雾浓度变化特性进行研究,提出一种针对未知雾浓度环境的自适应图像去雾算法。使用自适应PI调节方式调节去雾参数,使用去雾图像中的阻塞点占比表示图像去雾程度,实现对去雾参数的实时校正,保证算法对不同雾浓度环境下图像的去雾效果。实验结果表明,该算法相较于现有去雾算法具有更好的去雾效果,使用该算法作为前端的视觉SLAM系统在雾天环境下具有更好的鲁棒性。
To solve the problem of instability of visual SLAM system in foggy environment caused by image degradation,the characteristics of foggy video were studied,and an adaptive image defogging algorithm for unknown foggy environment was proposed.The proportion of blocking points in the defogging image was used to judge the defogging degree of the image.Accor-ding to the defogging degree of the image,an adaptive PI algorithm was used to correct the parameters of the defogging algorithm,and the defogging effect of the algorithm in the unknown foggy environment was ensured.Experimental results indicate that the adaptive defogging algorithm has better quality and stability than the traditional defogging algorithm,and the visual SLAM system using the proposed algorithm as the front-end has better robustness in foggy environment.
作者
刘自若
万熠
侯嘉瑞
梁西昌
孙尧
LIU Zi-ruo;WAN Yi;HOU Jia-rui;LIANG Xi-chang;SUN Yao(Key Laboratory of High-Efficiency and Clean Mechanical Manufacture,School of Mechanical Engineering,Shandong University,Jinan 250061,China;National Demonstration Center for Experimental Mechanical Engineering Education,Shandong University,Jinan 250061,China)
出处
《计算机工程与设计》
北大核心
2020年第9期2447-2453,共7页
Computer Engineering and Design
基金
山东省重点研发计划基金项目(2017CXGC0917)
山东大学基本科研业务费专项基金项目(2017JC041)。