Mine infrared image enhancement algorithm based on dual domain and ILoG-CLAHE
-
摘要: 针对矿井复杂作业环境导致的红外图像降质，现有红外图像增强算法在实现信噪比和对比度提升的同时易丢失场景细节信息或造成目标边缘模糊的问题，提出了一种基于双域分解耦合改进的高斯−拉普拉斯（ILoG）算子和对比度受限自适应直方图均衡化（CLAHE）（ILoG−CLAHE）的矿井红外图像增强算法。首先，利用双域分解模型将矿井红外图像分解为包含高频信息的细节子图和低频信息的基础子图；其次，利用CLAHE算法对基础子图的亮度、对比度和清晰度进行提升，用以突出监视场景的概貌特征，采用构造的ILoG算子对细节子图进行噪声抑制和边缘锐化，并消除梯度反转现象；然后，通过重构处理后的基础子图和细节子图得到了图像质量改善后的重构图像；最后，设计了一种灰度重分布的Gamma校正函数，对重构图像进行亮度调整，进而得到矿井红外增强图像。通过主观视觉和客观指标对算法进行了性能分析，结果表明：经基于双域和ILoG−CLAHE的矿井红外图像增强算法增强后的矿井红外图像，整体视觉效果和客观指标均得到了较大提升，综合增强性能和鲁棒性更好。相较于原矿井红外图像和6种对比算法（CLAHE算法、双边滤波器（BF）分解与基础子图的CLAHE增强（BF−CLAHE）算法、BF分解与Gamma变换（BF−Gamma）算法、引导滤波与Gamma变换（GF−Gamma）算法、自适应直方图均衡化（AHE）耦合拉普拉斯变换（AHE−LP）算法、基于反锐化掩膜（UM）的图层融合（LF−UM）算法），该算法的综合评价指标值分别提高了0.28，0.11，0.23，0.38，0.57，0.04，0.10，图像亮度、清晰度和对比度均得到了较大提升，并且实现了噪声抑制和边缘锐化，表明该算法适用于矿井复杂作业环境中红外图像的增强处理。Abstract: The complex working environment of mine leads to the degradation of the infrared image. The existing infrared image enhancement algorithm is easy to lose the scene details or causes the target edge blur while improving the signal-to-noise ratio and contrast. In order to solve the above problems, a mine infrared image enhancement algorithm based on dual domain decomposition coupling improved Gaussian Laplacian (ILoG) factor and contrast limited adaptive histogram equalization (CLAHE) (ILoG-CLAHE) is proposed. Firstly, the dual domain decomposition model is used to decompose the mine infrared image into a detailed sub-images containing high-frequency information and a basic sub-images containing low-frequency information. Secondly, the CLAHE algorithm is used to improve the brightness, contrast and definition of the basic sub-images to highlight the general features of the monitoring scene. The constructed ILoG operator is used to suppress noise and sharpen edges of detail sub-images and eliminate gradient inversion. Thirdly, the reconstructed image with improved image quality is obtained through the basic sub-image and detail sub-image after reconstruction processing. Finally, a Gamma correction function of gray level redistribution is designed to adjust the brightness of the reconstructed image. The mine infrared-enhanced image is obtained. The performance of the algorithm is analyzed by subjective vision and objective indicators. The results show that the overall visual effect and objective index of the mine infrared image enhanced by the mine infrared image enhancement algorithm based on dual domain and ILoG-CLAHE have been greatly improved. The comprehensive enhancement performance and robustness are better. Compared with the original mine infrared image and the six comparison algorithms, the comprehensive evaluation index values of this algorithm are increased by 0.28, 0.11, 0.23, 0.38, 0.57, 0.04, and 0.10 respectively. The six algorithms include CLAHE algorithm, bilateral filtering(BF) decomposition and CLAHE enhancement of basic sub-images (BF-CLAHE) algorithm, BF decomposition and Gamma transform (BF-Gamma) algorithm, guided filtering and Gamma transform (GF-Gamma) algorithm, adaptive histogram equalization(AHE) coupled Laplacian transform (AHE-LP) algorithm, and un-sharp mask(UM) based layer fusion (LF-UM) algorithm. The brightness, clarity and contrast of images are greatly improved, and noise suppression and edge sharpening are realized. It shows that the algorithm is suitable for the enhancement of infrared images in the complex working environment of mine.
-
-
![]()
图 2 基于双域和ILoG−CLAHE的矿井红外图像增强算法实现原理
Figure 2. Implementation principle of mine infrared image enhancement algorithm based on dual domain and improved Gaussian Laplacian (ILoG) and contrast limited adaptive histogram equalization (CLAHE)
![]()
图 3 采煤工作面回风巷道的红外图像增强
Figure 3. Infrared image enhancement of the return air roaduay of coal mining face
![]()
图 4 井下辅运大巷的红外图像增强
Figure 4. Infrared image enhancement of underground auxiliary transportation lane
![]()
图 6 不同算法的综合评价指标值折线图
Figure 6. Line chart of comprehensive evaluation indicator of different algorithms
![]()
图 7 本文算法综合评价指标值相较于红外图像f和对比算法的相对提高值
Figure 7. Relative improvement of the comprehensive evaluation indicator value of the proposed algorithm compared to the infrared image f and the comparison algorithms
表 1 实验1中不同算法的客观评价指标值
Table 1 Objective evaluation indicator values of different algorithms in experiment 1
图像 评价指标 Mean MLMSE AG MLIE SSIM 红外图像 42.10 17.39 8.78 2.34 — CLAHE算法处理后图像 80.08 135.50 87.42 2.66 0.50 BF−CLAHE算法处理后图像 74.80 98.38 72.48 2.41 0.57 BF−Gamma算法处理后图像 51.15 12.41 3.06 1.41 0.96 GF−Gamma算法处理后图像 17.90 16.62 4.75 1.90 0.71 AHE−LP算法处理后图像 82.76 155.92 108.96 2.69 0.37 LF−UM算法处理后图像 71.48 146.07 107.14 2.58 0.52 本文算法处理后图像 94.31 91.36 65.30 2.64 0.58 
下载: 导出CSV
表 2 实验2中不同算法的客观评价指标值
Table 2 Objective evaluation indicator values of different algorithms in experiment 2
图像 评价指标 Mean MLMSE AG MLIE SSIM 红外图像 46.24 17.50 10.87 2.12 — CLAHE算法处理后图像 77.45 109.10 83.89 2.51 0.69 BF−CLAHE算法处理后图像 71.43 65.58 38.20 1.92 0.73 BF−Gamma算法处理后图像 54.92 15.78 6.22 1.23 0.98 GF−Gamma算法处理后图像 21.74 15.73 7.35 1.67 0.74 AHE−LP算法处理后图像 77.53 128.72 71.15 2.63 0.67 LF−UM算法处理后图像 77.69 113.94 60.20 2.56 0.68 本文算法处理后图像 109.51 89.79 35.35 2.61 0.70
下载: 导出CSV
表 3 实验3中不同算法的客观评价指标值
Table 3 Objective evaluation indicator values of different algorithms in experiment 3
图像 评价指标 Mean MLMSE AG MLIE SSIM 红外图像 21.97 22.39 5.92 1.45 — CLAHE算法处理后图像 46.84 55.63 40.86 1.70 0.64 BF−CLAHE算法处理后图像 40.53 35.10 18.70 1.13 0.72 BF−Gamma算法处理后图像 27.53 19.65 3.77 0.81 0.97 GF−Gamma算法处理后图像 9.43 27.05 4.69 0.97 0.72 AHE−LP算法处理后图像 50.80 81.73 46.80 2.59 0.55 LF−UM算法处理后图像 46.17 71.15 59.27 1.78 0.64 本文算法处理后图像 61.90 57.41 22.28 2.63 0.66
下载: 导出CSV
表 4 不同算法的综合评价指标值
Table 4 Comprehensive evaluation indictor values of different algorithms
场景 综合评价指标值 f CLAHE算法 BF−CLAHE算法 BF−Gamma算法 GF−Gamma算法 AHE−LP算法 LF−UM算法 本文算法 实验1 0.52 0.71 0.62 0.34 0.24 0.71 0.70 0.72 实验2 0.49 0.63 0.42 0.33 0.13 0.64 0.60 0.65 实验3 0.41 0.51 0.35 0.32 0.13 0.66 0.59 0.68 随机实验 0.45 0.62 0.50 0.35 0.16 0.69 0.63 0.73
下载: 导出CSV
-
[1] 孙继平,范伟强. 矿井红外热成像远距离测温误差分析与精确测温方法[J]. 煤炭学报,2022,47(4):1-14. SUN Jiping,FAN Weiqiang. Error analysis and accurate temperature measurement method of infrared thermal imaging long-distance temperature measurement in underground mine[J]. Journal of China Coal Society,2022,47(4):1-14.
[2] 孔松涛,谢义,王松,等. 红外热像增强算法发展研究综述[J]. 重庆科技学院学报(自然科学版),2021,23(4):77-83. KONG Songtao,XIE Yi,WANG Song,et al. Review on the development of infrared thermal image enhancement algorithms[J]. Journal of Chongqing University of Science and Technology (Natural Sciences Edition),2021,23(4):77-83.
[3] 杨伟,陈益能,童鑫. 矿山应急救援红外图像SVM分割算法[J]. 陕西煤炭,2022,41(1):82-87,91. YANG Wei,CHEN Yineng,TONG Xin. SVM segmentation algorithm for infrared image of mine emergency rescue[J]. Shaanxi Coal,2022,41(1):82-87,91.
[4] 孙继平,范伟强. MS−ADoG域结合ReNLU与VGG−16的矿井双波段图像融合算法[J]. 光子学报,2022,51(3):13-27. DOI: 10.3788/gzxb20225103.0310002 SUN Jiping,FAN Weiqiang. Mine dual-band image fusion in MS-ADoG domain combined with ReNLU and VGG-16[J]. Acta Photonica Sinica,2022,51(3):13-27. DOI: 10.3788/gzxb20225103.0310002
[5] 刘涛,张炜,何付军,等. 红外热波检测方法图像增强环节研究[J]. 红外与激光工程,2012,41(7):1922-1927. LIU Tao,ZHANG Wei,HE Fujun,et al. Research on image enhancement in infrared thermal waves NDT[J]. Infrared and Laser Engineering,2012,41(7):1922-1927.
[6] 谭宇璇,樊绍胜. 基于图像增强与深度学习的变电设备红外热像识别方法[J]. 中国电机工程学报,2021,41(23):7990-7998. TAN Yuxuan,FAN Shaosheng. Infrared thermal image recognition of substation equipment based on image enhancement and deep learning[J]. Proceedings of the CSEE,2021,41(23):7990-7998.
[7] 陈钱. 红外图像处理技术现状及发展趋势[J]. 红外技术,2013,35(6):311-318. CHEN Qian. The status and development trend of infrared image processing technology[J]. Infrared Technology,2013,35(6):311-318.
[8] 沈磊,苏建忠,郭肇敏,等. 基于反锐化掩模技术的红外图像增强算法设计[J]. 南开大学学报(自然科学版),2019,52(1):29-35. SHEN Lei,SU Jianzhong,GUO Zhaomin,et al. Design of infrared image enhancement algorithm based on unsharp mask technology[J]. Acta Scientiarum Naturalium Universitatis Nankaiensis,2019,52(1):29-35.
[9] 周永康,朱尤攀,曾邦泽,等. 宽动态红外图像增强算法综述[J]. 激光技术,2018,42(5):718-726. ZHOU Yongkang,ZHU Youpan,ZENG Bangze,et al. Review of high dynamic range infrared image enhancement algorithms[J]. Laser Technology,2018,42(5):718-726.
[10] 范永杰,金伟其,刘斌,等. FLIR公司热成像细节增强DDE技术的分析[J]. 红外技术,2010,32(3):161-164. FAN Yongjie,JIN Weiqi,LIU Bin,et al. An analysis of digital detail enhancement (DDE) technology developed by FLIR[J]. Infrared Technology,2010,32(3):161-164.
[11] 程铁栋,卢晓亮,易其文,等. 一种结合单尺度Retinex与引导滤波的红外图像增强方法[J]. 红外技术,2021,43(11):1081-1088. CHENG Tiedong,LU Xiaoliang,YI Qiwen,et al. Research on infrared image enhancement method combined with single-scale Retinex and guided image filter[J]. Infrared Technology,2021,43(11):1081-1088.
[12] 吕侃徽,张大兴. 基于自适应直方图均衡化耦合拉普拉斯变换的红外图像增强算法[J]. 光学技术,2021,47(6):747-753. LYU Kanhui,ZHANG Daxing. Infrared image enhancement algorithms based on adaptive histogram equalization coupled with Laplace transform[J]. Optical Technology,2021,47(6):747-753.
[13] 路陆,姜鑫,杨锦程,等. 基于自适应引导滤波的红外图像细节增强[J]. 液晶与显示,2022,37(9):1182-1189. DOI: 10.37188/CJLCD.2022-0024 LU Lu,JIANG Xin,YANG Jincheng,et al. Adaptive guided filtering based infrared image detail enhancement[J]. Chinese Journal of Liquid Crystals and Displays,2022,37(9):1182-1189. DOI: 10.37188/CJLCD.2022-0024
[14] 葛朋,杨波,毛文彪,等. 基于引导滤波的高动态红外图像增强处理算法[J]. 红外技术,2017,39(12):1092-1097. GE Peng,YANG Bo,MAO Wenbiao,et al. High dynamic range infrared image enhancement algorithm based on guided image filter[J]. Infrared Technology,2017,39(12):1092-1097.
[15] 汪伟,许德海,任明艺. 一种改进的红外图像自适应增强方法[J]. 红外与激光工程,2021,50(11):419-427. WANG Wei,XU Dehai,REN Mingyi. An improved infrared image adaptive enhancement method[J]. Infrared and Laser Engineering,2021,50(11):419-427.
[16] 魏亮,王炎,胡文浩,等. 基于双域分解的夜间车辆红外图像研究[J]. 激光与红外,2021,51(11):1538-1544. WEI Liang,WANG Yan,HU Wenhao,et al. Research on infrared image of vehicle at night based on dual domain decomposition[J]. Laser & Infrared,2021,51(11):1538-1544.
[17] 孙继平,孙雁宇,范伟强. 基于可见光和红外图像的矿井外因火灾识别方法[J]. 工矿自动化,2019,45(5):1-5,21. SUN Jiping,SUN Yanyu,FAN Weiqiang. Mine exogenous fire identification method based on visible light and infrared image[J]. Industry and Mine Automation,2019,45(5):1-5,21.
[18] UMRI B K, UTAMI E, KURNIAWAN M P. Comparative analysis of CLAHE and AHE on application of CNN algorithm in the detection of Covid-19 patients[C]. 4th International Conference on Information and Communications Technology （ICOIACT）, Yogyakarta, 2021: 203-208.
[19] 黄勇. 基于双边滤波和改进CLAHE算法的低照度图像增强研究[D]. 湘潭: 湘潭大学, 2019. HUANG Yong. Low illumination image enhancement based on bilateral filtering and improved CLAHE algorithm[D]. Xiangtan: Xiangtan University, 2019.
[20] YU Yongbin,YANG Nijing,YANG Chenyu,et al. Memristor bridge-based low pass filter for image processing[J]. Journal of Systems Engineering and Electronics,2019,30(3):448-455. DOI: 10.21629/JSEE.2019.03.02
[21] VENETIS J. An analytic exact form of the unit step function[J]. Mathematics and Statistics,2014,2(7):235-237. DOI: 10.13189/ms.2014.020702
[22] FAN Weiqiang,HUO Yuehua,LI Xiaoyu. Degraded image enhancement using dual-domain-adaptive wavelet and improved fuzzy transform[J]. Mathematical Problems in Engineering,2021(3):1-12.
[23] BRANCHITTA F,DIANI M,CORSINI G,et al. New technique for the visualization of high dynamic range infrared images[J]. Optical Engineering,2009,48(9):096401.DOI: 10.1117/1.3216575.
[24] 范伟强,刘毅. 基于自适应小波变换的煤矿降质图像模糊增强算法[J]. 煤炭学报,2020,45(12):4248-4260. FAN Weiqiang,LIU Yi. Fuzzy enhancement algorithm of coal mine degradation image based on adaptive wavelet transform[J]. Journal of China Coal Society,2020,45(12):4248-4260.
[25] AMANDEEP K,CHANDAN S. Contrast enhancement for cephalometric images using wavelet-based modified adaptive histogram equalization[J]. Applied Soft Computing,2016,51(2):180-191.
计量
- 文章访问数: 292
- HTML全文浏览量: 49
- PDF下载量: 28
