计算机科学 ›› 2021, Vol. 48 ›› Issue (11A): 420-423.doi: 10.11896/jsjkx.210200072
张鼎, 蒋慕蓉, 黄亚群
ZHANG Ding, JIANG Mu-rong, HUANG Ya-qun
摘要： 能见度检测是计算机视觉与交通视频图像处理的热点问题。针对传统检测方法存在硬件成本高、适用范围小、检测效率低等不足,给出一种利用透射率和场景深度获取单幅图像能见度的检测方法。首先根据Koschmieder定律和ICAO推荐的对比阈值推导出能见度检测公式,然后根据大气衰减模型得到消光系数,利用暗通道先验理论获取透射率值,结合SFS(从阴影恢复形状)和双目模型获取场景深度值,最后通过求解消光系数反演图像的能见度。实验结果验证了该方法的有效性,精确度和检测效率有较大提高,且不需要相机内部参数,也不需要拍摄同一场景的多幅图像,操作简单、适用范围较广。
中图分类号:
| [1]ZENG S E,WANG G L.Visibility observation and its instruments[J].Acta Meteorologica Sinica,1999(2):80-85. [2]GAO L D,XU L.Research on detection method of image visibility detector [C]//Proceedings of the 12th China Intelligent Transportation Annual Conference.2017. [3]NICOLAS H,TAREL J P,LAVENANT J,et al.Automatic fog detection and estimation of visibility distance through use of an onboard camera[J].Machine Vision and Applications,2006,17(1):8-20. [4]HAUTIÈRE N,LABAYRADE,R,AUBERT D.Real-time disparity contrast combination for onboard estimation of the visibility distance[J].IEEE Transactions on Intelligent Transportation System,2006,7(2):201-212. [5]XIANG W S.Research on visibility estimation based on Traffic Video[D].Shanghai:Shanghai Jiaotong University,2014. [6]XU M,ZHANG H Y,WU Y D.Visibility detection algorithm of fog image based on scene depth[J].Automatic instrument,2017,38(9):89-94. [7]WANG J H,LI D,LIU S P.Binocular obstacle detection of fire robot based on improved RANSAC[J].Computer Engineering and Application,2017,53(2):236-240. [8]ZHAO Z Y,JIANG M R,HUANG Y Q,et al.Single image depth estimation algorithm based on SFS and binocular model[J].Computer Science,2019,46(S1):161-164. [9]NEGRU M,NEDEVSCHI S.Image based fog detection and visibility estimation for driving assistance systems [C]//IEEE International Conference on Intellugent Computer Communication and Processing.2013:163-168. [10]XU X,YIN X C,LI Y,et al.Visibility measurement methodbased on image understanding[J].Pattern recognition and artificial intelligence,2013,26(6):543-551. [11]XU H,TAN Y B,LIU B W,et al.Image defogging restoration method based on super pixel and dark channel priori[J].Journal of Ocean University of China (Natural Science Edition),2020,50(10):118-124. [12]ZHANG J.Influence of haze environment on image quality and simulation research[D].Xi'an:Xi'an University of Electronic Science and Technology,2018. [13]JIN Y,WANG N,YE J Y.Real time low visibility assistantdriving system based on dark channel priori[J].Journal of East China University of science and Technology (Natural Science Edition),2019,45(2):310-315. [14]CAO F,ZHU Y K.3D reconstruction and accuracy analysisbased on SFS method[J].Computer Science,2017,44(S1):244-247. [15]JOANNIDIS C.Binocular Vision in the Total Situation[J].American Journal of Psycho-analusis,2018,23(3):86-100. [16]QIU H.Research on image degradation model in foggy weather[D].East China University of Science and Technology,2007. [17]BRONTE S,BERGASA L M,ALCANTARILLA P F.Fog De-tection System Based on Computer Vision Techniques[C]//12th International IEEE Conference on Intelligent Transportation Systems,2009(ITSC '09).IEEE,2009. [18]ZHU W X.Research of Fog Driving Scenarios and VisibilityRecognition Algorithm Based on Video[J].Journal of Image & Signal Processing,2015,4(3):67-77. |
| [1] | 何彦辉, 吴桂兴, 吴志强. 基于域适应的X光图像的目标检测 Domain Alignment Based Object Detection of X-ray Images 计算机科学, 2021, 48(1): 175-181. https://doi.org/10.11896/jsjkx.200200023 |
| [2] | 崔倩男,田小平,吴成茂. 基于引导滤波改进的暗原色去雾算法 Improved Algorithm of Haze Removal Based on Guided Filtering and Dark Channel Prior 计算机科学, 2018, 45(5): 285-290. https://doi.org/10.11896/j.issn.1002-137X.2018.05.049 |
|
||
首页