基于去尘估计和多重曝光融合的煤矿井下图像增强方法

郝博南

doi:10.13272/j.issn.1671-251x.2023080105

基于去尘估计和多重曝光融合的煤矿井下图像增强方法

doi: 10.13272/j.issn.1671-251x.2023080105

郝博南^{1, 2, 3,}

1.
煤炭科学技术研究院有限公司，北京　100013
2.
煤矿应急避险技术装备工程研究中心，北京　100013
3.
北京市煤矿安全工程技术研究中心，北京　100013

基金项目: 国家自然科学基金青年基金项目（42201386）；天地科技股份有限公司科技创新创业资金专项项目（2023-TD-ZD005-005，2022-2-TD-ZD001，2022-TD-ZD001）。

详细信息

作者简介:
郝博南（1993—），男，河北唐山人，助理研究员，硕士，现从事矿井视频分析技术的相关研究工作，E-mail：haobonan@ccrise.cn

中图分类号: TD67
计量
- 文章访问数: 172
- HTML全文浏览量: 24
- PDF下载量: 45
- 被引次数: 0
出版历程
- 收稿日期: 2023-08-28
- 修回日期: 2023-11-15
- 网络出版日期: 2023-11-27

Coal mine underground image enhancement method based on dust removal estimation and multiple exposure fusion

HAO Bonan^{1, 2, 3
,}

1.
CCTEG China Coal Research Institute, Beijing 100013, China
2.
Engineering Research Center for Technology Equipment of Emergency Refuge in Coal Mine, Beijing 100013, China
3.
Beijing Engineering and Research Center of Mine Safe, Beijing 100013, China

摘要

摘要: 煤矿井下粉尘和暗光等因素导致采集的图像质量低，而现有图像增强方法存在图像细节丢失、局部特征不清晰、无法消除噪声、去尘效果不理想等问题。针对上述问题，提出了一种基于去尘估计和多重曝光融合的煤矿井下图像增强方法。该方法通过尘化图像简易模型及暗原色理论，并引入自适应衰减系数估算出图像透射率，再根据透射率分布，通过尘化图像简易模型复原物体的原始图像，将煤矿井下图像中的粉尘去除；利用多重曝光融合算法为曝光不足的原始图像生成一组不同曝光比的图像，并引入权值矩阵将这些不同曝光比的图像与原始图像进行融合，有效提升暗光图像质量。实验结果表明：相较于直方图均衡法、带色彩恢复的Retinex（MSRCR）方法、改进Retinex方法，该方法在去尘及暗光增强方面效果较好，颜色还原度较高，白边和过曝等现象得到抑制，且增强后的图像平均对比度分别提升了169.00%，42.50%，10.88%，平均图像熵分别提升了51.80%，16.45%，8.99%，平均亮度顺序误差（LOE）分别降低了31.01%，16.94%，7.83%，同时该方法运算耗时最短。
- 图像增强 /
- 去尘 /
- 多重曝光融合 /
- 暗光增强 /
- 透射率 /
- 曝光比
Abstract: Factors such as dust and dim light in coal mines lead to low quality of collected images. The existing image enhancement methods have problems such as loss of image details, unclear local features, inability to eliminate noise, and unsatisfactory dust removal effects. In order to solve the above problems, a coal mine underground image enhancement method based on dust removal estimation and multiple exposure fusion is proposed. This method uses a simplified model of dust image and dark primary color theory, and introduces an adaptive attenuation coefficient to estimate the image transmittance. Based on the transmittance distribution, the original image of the object is restored using the simplified model of dust image to remove dust from the coal mine underground image. The method uses a multiple exposure fusion algorithm to generate a set of images with different exposure ratios for underexposed original images, and introduces a weight matrix to fuse these images with the original image, effectively improving the quality of dim light images. The experimental results show that compared to the histogram equalization method, the multiple-scale Retinex with color restoration method (MSRCR), and the improved Retinex method, this method has better results in dust removal and dim light enhancement, with higher color restoration, suppressed white edges and overexposure. The average contrast of the enhanced images has increased by 62.78%, 29.82%, 9.8%, and the average image entropy has increased by 34.13%, 14.12%, and 8.25%, respectively. The average lightness order error (LOE) has been reduced by 40.9%, 20.39%, and 8.5%, respectively. This method has the shortest computational time.
- image enhancement /
- dust removal /
- multiple exposure fusion /
- dim light enhancement /
- transmittance /
- exposure ratio

HTML全文

图 1 基于去尘估计和多重曝光融合的煤矿井下图像增强方法流程

Figure 1. Process of coal mine underground image enhancement method based on dust removal estimation and multiple exposure fusion

下载: 全尺寸图片幻灯片

图 2 多重曝光融合算法原理

Figure 2. Principle of multiple exposure fusion algorithm

下载: 全尺寸图片幻灯片

图 3 3种场景下不同方法增强效果

Figure 3. Enhancement effect of different methods in three scenarios

下载: 全尺寸图片幻灯片

图 4 不同方法运算耗时

Figure 4. Operation time of different methods

下载: 全尺寸图片幻灯片

表 1 消融实验结果

Table 1. Results of ablation experiment

图像	平均对比度	平均图像熵	平均LOE
原始图像	15.11	3.80	1 600
去尘处理后图像	32.86	5.30	1 209
暗光增强处理后图像	29.73	5.17	1 180
本文方法处理后图像	70.86	7.19	814

下载: 导出CSV

表 2 不同方法图像增强客观评价结果

Table 2. Objective evaluation results of image enhancement by different methods

方法	RESIDE			HazeRD			VV			MEF
方法	对比度	图像熵	LOE	对比度	图像熵	LOE	对比度	图像熵	LOE	对比度	图像熵	LOE
直方图均衡法	23.85	4.97	1 071	26.74	5.02	1 324	25.12	4.75	1 146	28.37	4.88	1 103
MSRCR方法	49.39	6.12	892	52.93	6.08	1 203	51.16	6.24	993	42.74	6.11	879
改进Retinex方法	61.55	6.52	801	69.32	6.43	1 089	64.50	6.71	914	56.82	6.58	771
本文方法	70.24	7.08	739	78.35	7.11	1 015	72.58	7.43	843	67.66	6.99	702

下载: 导出CSV

表 3 不同方法图像增强客观评价结果平均值

Table 3. Objective evaluation average results of image enhancement by different methods

方法	平均对比度	平均图像熵	平均LOE
直方图均衡法	26.02	4.71	1 161
MSRCR方法	49.06	6.14	992
改进Retinex方法	63.05	6.56	894
本文方法	69.91	7.15	824

下载: 导出CSV

参考文献(21)

[1]	程德强,钱建生,郭星歌,等. 煤矿安全生产视频AI识别关键技术研究综述[J]. 煤炭科学技术,2023,51(2):349-365. CHENG Deqiang,QIAN Jiansheng,GUO Xingge,et al. Review on key technologies of AI recognition for videos in coal mine[J]. Coal Science and Technology,2023,51(2):349-365.
[2]	孔二伟,张亚邦,李佳悦,等. 面向煤矿井下低光照图像的增强方法[J]. 工矿自动化,2023,49(4):62-69,85. KONG Erwei,ZHANG Yabang,LI Jiayue,et al. An enhancement method for low light images in coal mines[J]. Journal of Mine Automation,2023,49(4):62-69,85.
[3]	龚云,颉昕宇. 基于同态滤波方法的煤矿井下图像增强技术研究[J]. 煤炭科学技术,2023,51(3):241-250. GONG Yun,XIE Xinyu. Research on coal mine underground image recognition technology based on homomorphic filtering method[J]. Coal Science and Technology,2023,51(3):241-250.
[4]	JIANG He,CAI Huangkai,YANG Jie. Learning in-place residual homogeneity for image detail enhancement[C]. IEEE International Conference on Acoustics,Speech and Signal Processing,Calgary,2018:1428-1432.
[5]	智宁,毛善君,李梅. 基于双伽马函数的煤矿井下低亮度图像增强算法[J]. 辽宁工程技术大学学报(自然科学版),2018,37(1):191-197. ZHI Ning,MAO Shanjun,LI Mei. An enhancement algorithm for coal mine low illumination images based on bi-gamma function[J]. Journal of Liaoning Technical University(Natural Science),2018,37(1):191-197.
[6]	冯玮,姚顽强,蔺小虎,等. 顾及图像增强的煤矿井下视觉同时定位与建图算法[J]. 工矿自动化,2023,49(5):74-81. FENG Wei,YAO Wanqiang,LIN Xiaohu,et al. Visual simultaneous localization and mapping algorithm of coal mine underground considering image enhancement[J]. Journal of Mine Automation,2023,49(5):74-81.
[7]	纪平,胡学友,张瑞琦. 基于直方图均衡算法的图像增强技术研究[J]. 蚌埠学院学报,2021,10(2):40-43. JI Ping,HU Xueyou,ZHANG Ruiqi. Research on image enhancement technology based on histogram equalization algorithm[J]. Journal of Bengbu University,2021,10(2):40-43.
[8]	ZHAO Lijun,BAI Huihui,LIANG Jie,et al. Local activity-driven structural-preserving filtering for noise removal and image smoothing[J]. Signal Processing,2019,157:62-72. doi: 10.1016/j.sigpro.2018.11.012
[9]	魏华良,王金祥. 运动模糊数字图像边缘锐化增强仿真研究[J]. 计算机仿真,2020,37(7):459-462,497. WEI Hualiang,WANG Jinxiang. Simulation research on edge sharpening enhancement of motion blurred digital image[J]. Computer Simulation,2020,37(7):459-462,497.
[10]	贺元恺. 基于FPGA的实时图像去雾系统研究与设计[D]. 西安:西安科技大学,2020. HE Yuankai. Research and design of real-time image defogging system based on FPGA[D]. Xi'an:Xi'an University of Science and Technology,2020.
[11]	吴开兴,张琳,李丽宏. 煤矿井下雾尘图像清晰化算法[J]. 工矿自动化,2018,44(3):70-75. WU Kaixing,ZHANG Lin,LI Lihong. Sharpening algorithm for underground images with fog and dust[J]. Industry and Mine Automation,2018,44(3):70-75.
[12]	ZHAN Yutong,GAO Kun,WANG Junwei,et al. Single-image dehazing using extreme reflectance channel prior[J]. IEEE Access,2021,9:87826-87838. doi: 10.1109/ACCESS.2021.3090202
[13]	曹虎晨,姚善化,王仲根. 基于边界约束的煤矿井下尘雾图像去雾算法[J]. 工矿自动化,2022,48(6):139-146. CAO Huchen,YAO Shanhua,WANG Zhonggen. Defogging algorithm of underground coal mine dust and fog image based on boundary constraint[J]. Journal of Mine Automation,2022,48(6):139-146.
[14]	FENG Xiaomei,LI Jinjiang,HUA Zhen,et al. Low-light image enhancement based on multi-illumination estimation[J]. Applied Intelligence,2021(51):5111-5131.
[15]	MERIANOS I,MITIANOUDIS N. Multiple-exposure image fusion for HDR image synthesis using learned analysis transformations[J]. Journal of Imaging,2019,5(3). DOI: 10.3390/jimaging5030032.
[16]	CHEN Xiaobo,DU Hu,ZHAN Jinkai,et al. A self-adaptive multiple exposure image fusion method for highly reflective surface measurements[J]. Machines,2022(10). DOI: 10.3390/machines10111004.
[17]	黄子蒙,徐望明,但愿. 基于对称亮度映射和虚拟多曝光融合的非均匀光照图像增强[J]. 液晶与显示,2022,37(12):1580-1589. doi: 10.37188/CJLCD.2022-0172 HUANG Zimeng,XU Wangming,DAN Yuan. Non-uniform illumination image enhancement via symmetric brightness mapping and virtual multi-exposure fusion[J]. Chinese Journal of Liquid Crystals and Displays,2022,37(12):1580-1589. doi: 10.37188/CJLCD.2022-0172
[18]	WANG Xiaocheng,HU Ruimin,XU Xin. Single low-light image brightening using learning-based intensity mapping[J]. Neurocomputing,2022,508:315-323. doi: 10.1016/j.neucom.2022.08.042
[19]	刘卫华,马碧燕. 基于图像全序列特征权重的多曝光图像融合方法[J]. 激光与光电子学进展,2022,59(8):289-299. LIU Weihua,MA Biyan. Multiexposure image fusion method based on feature weight of image sequence[J]. Laser & Optoelectronics Progress,2022,59(8):289-299.
[20]	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
[21]	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. doi: 10.13225/j.cnki.jccs.2020.0514 ZHANG Liya,HAO Bonan,MENG Qingyong,et al. Method of image enhancement in coal mine based on improved retex fusion algorithm in HSV space[J]. Journal of China Coal Society,2020,45(S1):532-540. doi: 10.13225/j.cnki.jccs.2020.0514