摘要
提出了一种应用单目机器视觉技术并结合Hough变换消失线检测和边界检测技术来构建距离深度图的方法,并应用衰减-空气光去雾模型对图像进行了恢复。结果表明,采用该方法恢复的道路图像色彩自然,模型与实际道路完全匹配,对于不同距离的景物都具有良好的恢复效果。
In order to remove weather effects from images,monocular camera machine vision,Hough transform and edge detection are combined to construct the scene depth map. It is assumed that the brightness of each pixel at the receiver is composed of scene point attenuation irradiance and airlight illumination scattering intensity,by removing airlight illumination scattering and removing particles to restore the image degraded in foggy day,the color of image restored by this method is natural,the model and real road are matched accurately,the different distance scene has satisfied effect.
出处
《光学技术》
CAS
CSCD
北大核心
2010年第4期554-559,共6页
Optical Technique
基金
863计划(2007AA11Z238)资助项目
关键词
摄像机标定
消失线
衰减-空气光模型
单目视觉
HOUGH变换
边界检测
去雾
camera calibration
vanishing line
attenuation-airlight model
monocular camera machine vision
Hough transform
edge detection
defog
