Research and application of video fog concentration detection and real-time fog removal method in underground coal mine
-
摘要: 煤矿井下工作面在生产状态下由于喷雾除尘操作引起雾气浓度动态变化，导致视频图像画面模糊不清，严重影响煤矿井下可视化远程干预性采煤控制及操作。针对上述问题，提出了一种煤矿井下工作面视频雾浓度检测及实时去雾方法。首先利用颜色衰减先验计算含雾视频图像亮度值和饱和度值差异，实现雾浓度检测，进一步识别含雾图像和无雾图像。其次，使用颜色衰减先验和场景变化概率模型对视频时间连续代价函数进行矫正，减小视频相邻帧之间的透射率误差，减轻去雾后视频图像画面的闪烁影响。最后，分别利用煤矿井下工作面视频雾浓度检测及实时去雾方法和Kim方法对煤矿井下场景有雾视频进行处理。实验结果表明：① 雾浓度检测方法可以准确地计算出图像场景中的雾浓度分布，提取到的雾浓度最大连通区域占总图像像素的38.693%，大于雾浓度阈值20%，为含雾图像。根据含雾图像识别结果自动忽略无雾图像，有选择性地对有雾图像进行去雾处理。② 采用煤矿井下工作面视频雾浓度检测及实时去雾方法对煤矿井下工作面的不同区域（支架区域和煤壁区域）及不同雾浓度（中等雾浓度和较高雾浓度）的生产视频进行去雾处理，去雾后视频图像对比度明显增强，视觉效果也更加明亮清晰。③ 实时去雾方法的均方误差曲线在Kim方法的均方误差曲线下方，说明其对连续场景视频去雾后，视频相邻帧的均方误差值减小，有效抑制了去雾视频的闪烁现象；使用对比度代价函数和颜色信息损失代价函数估计含雾图像的透射率值，可在变化场景取得理想的去雾效果。④ 实时去雾方法的去雾视频相邻帧之间的均方误差均值较Kim实时方法减小4.26，提高了相邻帧之间的相似性，进一步抑制了相邻帧之间的图像闪烁现象。在运行时间方面，实时去雾方法每帧处理时间较Kim方法增加了2 ms，但是其每帧处理时间小于40 ms，满足实时性要求。Abstract: The dynamic change of scene fog concentration caused by spraying dust removal operation in the production state of underground coal mine working face leads to the blurred visual video image. This seriously affects the visual remote intervention coal mining control and operation in underground coal mine. Aiming at the above problems, a method of video fog concentration detection and real-time fog removal underground in coal mine working face is proposed. Firstly, the difference between the brightness value and the saturation value of the fog-containing video image is calculated based on the color attenuation prior to realize the fog concentration detection. The fog-containing image and the non-fog image are further identified. Secondly, the color attenuation prior and the scene change probability model are used for correcting the video time continuous cost function. The transmittance error between adjacent frames of the video is reduced. The flicker influence of the fog removed video image is reduced. Finally, the video fog concentration detection and real-time fog removal method in the coal mine working face and the Kim method are respectively used to process the foggy video of the coal mine underground scene. The experimental results show the following points. ① The fog concentration distribution in the image scene can be accurately calculated by the fog concentration detection method. The maximum connected region of the extracted fog concentration accounts for 38.693% of the total image pixels. The fog-containing images are those greater than 20% of the fog concentration threshold. According to the recognition result of the fog-containing image, the non-fog image is automatically ignored. The fog-containing image is selectively fog removed. ② The video fog concentration detection and real-time fog removal method in coal mine working face is used to remove the fog of the production video of different areas (a support area and a coal wall area) and different fog concentrations (a medium fog concentration and a higher fog concentration). The contrast of the video image is obviously enhanced after fog removal, and the visual effect is brighter and clearer. ③ The mean square error curve of the real-time fog removal method is lower than that of Kim method. This indicates that the mean square error value of the adjacent frames of the video is reduced after the continuous scene video is fog removed. The flicker phenomenon of the real-time fog removal result video is effectively suppressed. The contrast cost function and the color information loss cost function are used to estimate the transmittance of the fog-containing image. The ideal fog removal effect is obtained when the scene changes. ④ The mean value of mean square error between the adjacent frames of the fog removal result of the real-time fog removal method is reduced by 4.26 compared with the Kim method. The similarity between the adjacent frames is improved. The image flicker phenomenon between the adjacent frames is further suppressed. In the aspect of running time, the processing time of each frame of the real-time fog removal method is increased by 2 ms compared with the Kim method. However, the processing time of each frame is less than 40 ms, which meets the real-time requirement.
-
-
![]()
图 1 不同含雾浓度像素块的亮度值与饱和度值差异
Figure 1. The difference of brightness values and saturation values of different fog densty
![]()
图 3 输出
$ J\left( x \right) $ 和输入$ I\left( x \right) $ 之间的线性关系Figure 3. Linear relationship between output
$ J\left( x \right) $ and input$ I\left( x \right) $ ![]()
图 4 含雾图像和无雾图像的雾浓度检测结果
Figure 4. Fog concentration detection results for fog-containing images and non-fog images
![]()
图 5 煤矿井下工作面支架区域、煤壁区域不同含雾浓度场景去雾前后效果
Figure 5. The effect before and after fog removal in different fog concentration scenarios in underground working face support area and coal wall area
![]()
图 6 2种视频去雾方法的结果图像和原始含雾图像的相邻帧之间的均方误差曲线
Figure 6. Mean square error curves between adjacent frames of two video defogging result images and original fog-containing images
表 1 不同去雾方法去雾后的视频相邻帧均方误差值和消耗时间统计
Table 1 Statistics of mean square error value and consumption time of adjacent video frames after different fog removal methods
去雾方法 均方误差均值 每帧处理时间/ms Kim方法 17.29 33 本文方法 13.03 35 
下载: 导出CSV
-
[1] LIU Zhigang,CAO Anye,GUO Xiaosheng,et al. Deephole water injection technology of strong impact tendency coal seam-a case study in Tangkou Coal Mine[J]. Arabian Journal of Geosciences,2018,11(2):1-9.
[2] 肖军良. 辛置煤矿2−208工作面喷雾降尘技术研究与应用[J]. 煤矿现代化,2021,30(6):46-48. XIAO Junliang. Research and application of spray dust suppression technology in 2-208 working face of Xinzhi Coal Mine[J]. Coal Mine Modernization,2021,30(6):46-48.
[3] 张立亚,郝博南,孟庆勇,等. 基于HSV空间改进融合Retinex算法的井下图像增强方法[J]. 煤炭学报,2020,45(增刊1):532-540. ZHANG Liya,HAO Bonan,MENG Qingyong,et al. Method of image enhancement in coal mine based on improved Retinex fusion algorithm in HSV space[J]. Journal of China Coal Society,2020,45(S1):532-540.
[4] 龚云,杨庞彬,颉昕宇. 结合同态滤波与直方图均衡化的井下图像匹配算法[J]. 工矿自动化,2021,47(10):37-41,61. GONG Yun,YANG Pangbin,JIE Xinyu. Underground image matching algorithm combining homomorphic filtering and histogram equalization[J]. Industry and Mine Automation,2021,47(10):37-41,61.
[5] TAN R. Visibility in bad weather from a single image[C]. IEEE Computer Society Conference on Computer Vision and Pattern Recognition（CVPR）, Anchorage, 2008: 1-8.
[6] FATTAL R. Single image dehazing[J]. ACM TransActions on Graphics,2008,27(3):1-9.
[7] HE Kaiming, SUN Jian, TANG Xiao'ou. Single image haze removal using dark channel prior[C]. 2nd International Conference on Computer Science and Intelligent Communicatio（CSIC）, Xi'an, 2018: 135-141.
[8] TAREL J, HAUTIERE N. Fast visibility restoration from a single color or gray level image[C]. Proceedings of the 12th IEEE International Conference on Computer Vision, Kyoto, 2009: 2201-2208.
[9] TAREL J,HAUTIERE N,CARAFFA L. Vision enhancement in homogeneous and heterogeneous fog[J]. IEEE Intelligent Transportation Systems Magazine,2012,4(2):6-20. DOI: 10.1109/MITS.2012.2189969
[10] ANCUTI C O, ANCUTI C, HERMANS C, et al. A fast semi-inverse approach to detect and remove the haze from a single image[C]. Asian Conference on Computer Vision（ACCV）, Queenstown, 2011: 501-514.
[11] KRATZ L, NISHINO K. Factorizing scene albedo and depth from a single foggy image[C]. Proceedings of the 2009 IEEE International Conference on Vision, Piscataway, 2009: 1701-1708.
[12] MENG Gaofeng, WANG Ying, DUAN Jiangyong, et al. Efficient image dehazing with boundary constraint and contextual regularization[C]. Proceedings of the 2013 IEEE International Conference on Computer Vision, Sydney, 2013: 617-624.
[13] 杨红,崔艳. 基于开运算暗通道和优化边界约束的图像去雾算法[J]. 光子学报,2018,47(6):244-250. YANG Hong,CUI Yan. Image defogging algorithm based on opening dark channel and improved boundary constraint[J]. Acta Photonica Sinica,2018,47(6):244-250.
[14] KIM J H,JANG W D,SIM J Y,et al. Optimized contrast enhancement for real-time image and video dehazing[J]. Journal of Visual Communication and Image Representation,2013,24(3):410-425. DOI: 10.1016/j.jvcir.2013.02.004
[15] YU Jing, LIAO Qingmin. Fast single image fog removal using edge-preserving smoothing[C]. IEEE International Conference on Acoustics, Speech and Signal Processing （ICASSP）, Prague, 2011: 1245-1248.
[16] CARR P, HARTLEY P. Improved single image dehazing using geometry[C]. Digital Imagee Computing: Techniques & Applications, Melbourne, 2009: 103-110.
[17] CAI Bolun,XU Xiangmin,JIA Kuiqing,et al. DehazeNet:an end-to-end system for single image haze removal[J]. IEEE Transactions on Image Processing,2016,25(11):5187-5198. DOI: 10.1109/TIP.2016.2598681
[18] ZHU Qingsong,MAI Jiaming,SHAO Ling. A fast single image haze removal algorithm using color attenuation prior[J]. IEEE Transactions on Image Processing,2015,24(11):3522-3533. DOI: 10.1109/TIP.2015.2446191
[19] HIDE R. Optics of the atmosphere:scattering by molecules and particles[J]. Physics Bulletin,1977,28(11):521.
[20] HE Kaiming,SUN Jian,TANG Xiao'ou. Guided image filtering[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2013,35(6):1397-1409. DOI: 10.1109/TPAMI.2012.213
[21] CHAPMAN B, JOST G, PAS R. Using OpenMP: portable shared memory parallel programming（scientific and engineering computation）[M]. Massachusetts（United Sates）, The MIT Press, 2007.
-
期刊类型引用(0)
其他类型引用(3)
计量
- 文章访问数: 327
- HTML全文浏览量: 74
- PDF下载量: 61
- 被引次数: 3
