Research on low illumination video enhancement technology in coal mine heading face
-
摘要:
针对煤矿掘进工作面视频光照较低、亮度不均、纹理模糊、噪声较多等问题，提出一种煤矿掘进工作面低照度视频增强算法。首先，利用卷积的可分离性将视频图像进行一维水平卷积与垂直卷积，再利用完美反射法实现视频图像自动白平衡，并使用图像混合增强技术提高视频图像整体亮度。然后，基于大气散射模型与暗通道先验方法，通过递归分割将图像分割为高光区、中间调和暗调区，并求取对应区间通道像素最大值，将其3者均值作为大气光照估计值，引入调节因子对透射率进行调整优化，并使用拉普拉斯锐化操作，增加图像高频成分、抑制图像低频成分，提高图像对比度。最后，基于改进的大气散射模型对掘进工作面低照度视频进行去雾处理。实验结果表明，视频增强算法能够对煤矿掘进工作面低照度视频进行实时增强、去雾处理，避免了视频图像暗淡、失真、模糊和突变等问题。相较于Retinex算法、ALTM算法和暗通道先验算法，视频增强算法大幅度提高了视频图像的信息熵、标准差和平均梯度，且具有较好的实时处理速度，能够为掘进工作面视频的目标识别、目标跟踪、目标监测和图像分割等后续处理提供优质、可靠的支撑。
Abstract:Aiming at overcoming low illumination, uneven brightness, blurry texture and more noise in the video of coal mine heading face, a low illumination video enhancement algorithm was proposed for coal mine heading face. Firstly, the separability of convolution was utilized to carry out one-dimensional horizontal and vertical convolution of video images, then the perfect reflection method was used to achieve the automatic white balance, and the image hybrid enhancement technology was utilized to improve the overall brightness of the video images. Then, the image was divided into the highlight area, the middle tone area and the dark tone area by recursive segmentation based on the atmospheric scattering model and the dark channel prior method, and the maximum channel pixel of the corresponding interval was obtained. Besides, the mean value of the three maximum pixel values was taken as the estimation value of atmospheric illumination, and the transmittance was adjusted and optimized by introducing the adjustment factor. Meanwhile, the Laplacian sharpening process was used to increase the high frequency component and suppress the low frequency component of the image to increase the image contrast. Finally, the low-illumination video of heading face was dehazed based on the improved atmospheric scattering model. The experimental results show that the proposed video enhancement algorithm could enhance and dehaze the low-illumination video of coal mine heading face in real time, which avoids the problems of dimness, distortion, blurring and mutation of video images. Compared with Retinex algorithm, ALTM algorithm and dark channel prior algorithm, the proposed video enhancement algorithm significantly improves the information entropy, standard deviation and average gradient of the video image, and has a higher real-time processing speed, which can provide high-quality and reliable support for subsequent processings such as video target recognition, target tracking, target monitoring and image segmentation of heading face video.
-
-
![]()
图 2 可分离高斯滤波前后效果对比
Fig. 2 Effect comparison before and after separable Gaussian filtering
表 1 不同算法评价指标结果
Table 1 Evaluation results of different algorithms
算法 图像序列 信息熵 标准差 平均梯度 处理时间/ms Retinex算法 1 5.53 32.62 4.37 374 2 5.82 33.14 5.51 365 3 5.78 33.74 4.98 352 4 6.62 36.92 6.47 353 5 6.66 37.25 5.92 457 ALTM算法 1 5.69 33.03 4.34 992 2 5.93 33.63 5.46 992 3 5.90 34.23 4.92 985 4 6.62 36.92 6.47 353 5 6.66 37.25 5.92 457 暗通道
先验算法1 5.15 24.45 3.39 2943 2 5.17 23.26 4.02 2931 3 5.19 24.01 3.74 2915 4 6.16 28.55 5.60 2951 5 6.16 28.69 4.79 3777 本文算法 1 6.24 39.48 4.60 343 2 6.02 37.79 6.05 344 3 6.04 38.06 5.66 302 4 6.97 41.07 6.81 304 5 6.79 41.74 6.38 316 
下载: 导出CSV
-
[1] 王国法. 加快煤矿智能化建设 推进煤炭行业高质量发展[J]. 中国煤炭,2021,47(1):2−10. DOI: 10.3969/j.issn.1006-530X.2021.01.002 WANG Guofa. Speeding up intelligent construction of coal mine and promoting high−quality development of coal industry[J]. China Coal,2021,47(1):2−10. DOI: 10.3969/j.issn.1006-530X.2021.01.002
[2] 王国法,任怀伟,赵国瑞,等. 煤矿智能化十大“痛点”解析及对策[J]. 工矿自动化,2021,47(6):1−11. DOI: 10.13272/j.issn.1671-251x.17808 WANG Guofa,REN Huaiwei,ZHAO Guorui,et al. Analysis and countermeasures of ten“pain points”of intelligent coal mine[J]. Industry and Mine Automation,2021,47(6):1−11. DOI: 10.13272/j.issn.1671-251x.17808
[3] 付燕,李瑶,严斌斌. 一种煤矿井下视频图像增强算法[J]. 工矿自动化,2018,44(7):80−83. DOI: 10.13272/j.issn.1671-251x.2017120014 FU Yan,LI Yao,YAN Binbin. An underground video image enhancement algorithm[J]. Industry and Mine Automation,2018,44(7):80−83. DOI: 10.13272/j.issn.1671-251x.2017120014
[4] 袁明道,谭彩,李阳,等. 基于图像融合和改进阈值的管道机器人探测图像增强方法[J]. 煤田地质与勘探,2019,47(4):178−185. DOI: 10.3969/j.issn.1001-1986.2019.04.027 YUAN Mingdao,TAN Cai,LI Yang,et al. A pipeline robot detection image enhancement method based on image fusion and improved threshold[J]. Coal Geology & Exploration,2019,47(4):178−185. DOI: 10.3969/j.issn.1001-1986.2019.04.027
[5] 智宁,毛善君,李梅. 基于照度调整的矿井非均匀照度视频图像增强算法[J]. 煤炭学报,2017,42(8):2190−2197. DOI: 10.13225/j.cnki.jccs.2017.0048 ZHI Ning,MAO Shanjun,LI Mei. Enhancement algorithm based on illumination adjustment for non–uniform illumination video images in coal mine[J]. Journal of China Coal Society,2017,42(8):2190−2197. DOI: 10.13225/j.cnki.jccs.2017.0048
[6] GUO Xiaojie,LI Yu,LING Haibin. LIME:Low–light image enhancement via illumination map estimation[J]. IEEE Transactions on Image Processing,2017,26(2):982−993. DOI: 10.1109/TIP.2016.2639450
[7] 董静薇,赵春丽,海博. 融合同态滤波和小波变换的图像去雾算法研究[J]. 哈尔滨理工大学学报,2019,24(1):66−70. DOI: 10.15938/j.jhust.2019.01.011 DONG Jingwei,ZHAO Chunli,HAI Bo. Image research on image de–fog algorithm based on fusion homomorphic filtering and wavelet transform[J]. Journal of Harbin University of Science and Technology,2019,24(1):66−70. DOI: 10.15938/j.jhust.2019.01.011
[8] 龚云, 颉昕宇. 一种改进同态滤波的井下图像增强算法[J/OL]. 煤炭科学技术, 2022: 1–8[2022-11-22]. DOI: 10.13199/j.cnki.cst.2021-0774. GONG Yun, XIE Xinyu. A downhole image enhancement algorithm based on improved homomorphic filtering[J/OL]. Coal Science and Technology, 2022: 1–8[2022-11-22]. DOI: 10.13199/j.cnki.cst.2021-0774.
[9] LI Zhi,JIA Zhenhong,YANG Jie,et al. Low illumination video image enhancement[J]. IEEE Photonics Journal,2020,12(4):1−13.
[10] 郭伶俐,贾振红. 基于大气散射模型的低照度视频增强算法[J]. 激光杂志,2022,43(6):105−110. DOI: 10.14016/j.cnki.jgzz.2022.06.105 GUO Lingli,JIA Zhenhong. Low illumination video enhancement algorithm based on the atmospheric scattering model[J]. Laser Journal,2022,43(6):105−110. DOI: 10.14016/j.cnki.jgzz.2022.06.105
[11] 蔡利梅,向秀华,李紫阳. 自适应HSV空间Retinex煤矿监控图像增强算法[J]. 电视技术,2017,41(4/5):11−15. DOI: 10.16280/j.videoe.2017.h4.003 CAI Limei,XIANG Xiuhua,LI Ziyang. Adaptive Retinex algorithm at HSV space for coal mine monitoring image enhancement[J]. Video Engineering,2017,41(4/5):11−15. DOI: 10.16280/j.videoe.2017.h4.003
[12] HE Kaiming,SUN Jian,TANG Xiaoou. Single image haze removal using dark channel prior[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,2011,33(12):2341−2353. DOI: 10.1109/TPAMI.2010.168
[13] 蔡秀梅,马今璐,吴成茂,等. 基于模糊同态滤波的彩色图像增强算法[J]. 计算机仿真,2020,37(6):342−346. DOI: 10.3969/j.issn.1006-9348.2020.06.070 CAI Xiumei,MA Jinlu,WU Chengmao,et al. Color image enhancement algorithm based on fuzzy homomorphic filtering[J]. Computer Simulation,2020,37(6):342−346. DOI: 10.3969/j.issn.1006-9348.2020.06.070
[14] 程德强,郑珍,姜海龙. 一种煤矿井下图像增强算法[J]. 工矿自动化,2015,41(12):31−34. DOI: 10.13272/j.issn.1671-251x.2015.12.009 CHENG Deqiang,ZHENG Zhen,JIANG Hailong. An image enhancement algorithm for coal mine underground[J]. Industry and Mine Automation,2015,41(12):31−34. DOI: 10.13272/j.issn.1671-251x.2015.12.009
[15] 汪凤林,周扬,叶绿,等. 基于机器视觉的飞轮齿圈缺陷和尺寸检测方法[J]. 中国测试,2020,46(5):31−38. DOI: 10.11857/j.issn.1674-5124.2020020005 WANG Fenglin,ZHOU Yang,YE Lyu,et al. Method for fault defection and size measurement for flywheel ring gear based on machine vision[J]. China Measurement & Testing Technology,2020,46(5):31−38. DOI: 10.11857/j.issn.1674-5124.2020020005
[16] 方建荣,苏畅,周晓方,等. 一种CMOS图像传感器信号处理自动白平衡算法[J]. 计算机工程,2015,41(9):245−250. DOI: 10.3969/j.issn.1000-3428.2015.09.045 FANG Jianrong,SU Chang,ZHOU Xiaofang,et al. An algorithm of automatic white balance for CMOS image sensor signal processing[J]. Computer Engineering,2015,41(9):245−250. DOI: 10.3969/j.issn.1000-3428.2015.09.045
[17] NAYAR S K, NARASIMHAN S G. Vision in bad weather[C]//Proceedings of the IEEE International Conference on Computer Vision, 1999: 820–827.
[18] DONG Xuan, WANG Guan, PANG Yi, et al. Fast efficient algorithm for enhancement of low lighting video[C]//IEEE International Conference on Multimedia and Expo, 2011: 1–6.
[19] 贾海鹏,张云泉,龙国平,等. 基于OpenCL的拉普拉斯图像增强算法优化研究[J]. 计算机科学,2012,39(5):271−277. DOI: 10.3969/j.issn.1002-137X.2012.05.065 JIA Haipeng,ZHANG Yunquan,LONG Guoping,et al. Research on Laplace image enhancement algorithm optimization based on OpenCL[J]. Computer Science,2012,39(5):271−277. DOI: 10.3969/j.issn.1002-137X.2012.05.065
[20] 王小东,冯筠,鲁定国,等. 基于先验知识的肝脏轮廓线提取算法研究[J]. 计算机应用研究,2014,31(1):281−284. DOI: 10.3969/j.issn.1001-3695.2014.01.066 WANG Xiaodong,FENG Jun,LU Dingguo,et al. Liver contour extraction based on prior knowledge model[J]. Application Research of Computers,2014,31(1):281−284. DOI: 10.3969/j.issn.1001-3695.2014.01.066
[21] 王媛彬,韦思雄,段誉,等. 基于自适应双通道先验的煤矿井下图像去雾算法[J]. 工矿自动化,2022,48(5):46−51. DOI: 10.13272/j.issn.1671-251x.2021110053 WANG Yuanbin,WEI Sixiong,DUAN Yu,et al. Defogging algorithm of underground coal mine image based on adaptive dual−channel prior[J]. Industry and Mine Automation,2022,48(5):46−51. DOI: 10.13272/j.issn.1671-251x.2021110053
[22] 张英俊,雷耀花,潘理虎. 基于暗原色先验的煤矿井下图像增强技术[J]. 工矿自动化,2015,41(3):80−83. DOI: 10.13272/j.issn.1671-251x.2015.03.020 ZHANG Yingjun,LEI Yaohua,PAN Lihu. Enhancement technique of underground image based on dark channel prior[J]. Industry and Mine Automation,2015,41(3):80−83. DOI: 10.13272/j.issn.1671-251x.2015.03.020
-
期刊类型引用(8)
1. 张旭辉，解彦彬，杨文娟，张超，万继成，董征，王彦群，蒋杰，李龙. 煤矿井下采掘工作场景非均质图像去雾与增强技术. 煤田地质与勘探. 2025(01): 245-256 . 
本站查看
2. 张铁聪，陈华州，赵俊杰，王利景，贾冬冬. 基于数字孪生技术的煤矿掘进机自动截割方法研究. 中国煤炭. 2024(01): 93-100 . 百度学术
3. 贾澎涛，靳路伟，王斌，郭风景，李娜. 采煤机截割部低照度图像的边缘检测技术. 煤田地质与勘探. 2024(04): 172-178 . 本站查看
4. 张旭辉，王悦，杨文娟，陈鑫，张超，黄梦瑶，刘彦徽，杨骏豪. 基于改进最佳缝合线的矿井图像拼接方法. 工矿自动化. 2024(04): 9-17 . 百度学术
5. 汪进超，韩增强，王益腾，王超，张国华. 基于像素空间信息的孔内低照度图像孔隙结构量化方法研究. 岩石力学与工程学报. 2024(S1): 3175-3186 . 百度学术
6. 丁序海，张侯，陈录平，党国杰. 基于多频无线电坑透技术的煤矿地质综合勘探研究. 能源与环保. 2024(06): 82-87 . 百度学术
7. 肖耀猛. 目标背景对比度优化下高煤尘低照度环境主动成像技术. 矿山机械. 2024(09): 60-65 . 百度学术
8. 陶荣颖，王守军，李南，朱伟. 煤矿副井口全景智能识别技术的研究与应用. 内蒙古煤炭经济. 2024(22): 154-156 . 百度学术
其他类型引用(2)
计量
- 文章访问数: 534
- HTML全文浏览量: 16
- PDF下载量: 53
- 被引次数: 10
