融合对比度拉伸的地铁隧道环境图像复原算法

doi:10.19678/j.issn.1000-3428.0066723

摘要/Abstract

摘要：

由于地铁隧道环境图像对比度低且细节模糊，在后续图像处理过程中难以从图像中准确提取关键的环境信息。为了解决该问题，提出一种融合对比度拉伸的模糊图像复原算法。将隧道图像转换到HSV颜色空间，根据V分量的灰度分布特点构建对比度拉伸模型对V分量进行增强，避免了传统低照度增强算法出现的过度增强问题，且能够自适应增强图像对比度。对增强后的V分量进行模糊类型分析，从而确定地铁隧道图像模糊类型，在此基础上将增强后的V分量图像划分为不同的区域，在各个区域分别选择符合条件的刀刃边缘线并对其扩散函数进行估计，同时以点扩散函数作为先验信息，采用非盲去卷积算法对各区域进行去模糊处理。融合H、S、V 3个分量，完成隧道环境低照度图像整体增强与复原。实验结果表明，所提算法能有效提高隧道图像整体及局部对比度，减少由高斯噪声引起的图像模糊，恢复图像中的细节信息。

关键词: 图像复原, 刀刃法, 低照度, 对比度拉伸, 自适应增强

Abstract:

The low contrast and blurred details in subway tunnel images pose challenges in accurately extracting key environmental information during subsequent image processing. To solve this problem, this study proposes a blurred image restoration algorithm based on contrast stretching. First, conversion of the tunnel image to the Hue, Saturation, Value(HSV) color space occurs. Following this, a contrast stretching model tailored for the gray distribution of the V component enhances this component. This circumvents over-enhancement issues found in traditional low-illumination enhancement algorithms and allows for adaptive enhancement of image contrast. Subsequently, an analysis to identify the blur type in the enhanced V component of the tunnel image is conducted. Based on this analysis, division of the enhanced V component image into different regions occurs. In each region, selection of a qualified edge and estimation of its diffusion function takes place. Utilizing the point diffusion function as prior information, a non-blind deconvolution algorithm is applied to deburr each region. In the final step, fusion of the three components H, S, and V occurs, culminating in the overall enhancement and restoration of low-illumination tunnel environment images. Experimental results indicate that the proposed algorithm effectively enhances both overall and local contrast in tunnel images, reduces blur caused by Gaussian noise, and restores detail in the images.

Key words: image restoration, knife edge method, low-illumination, contrast stretching, adaptive enhancement

杨康, 任愈, 吴学杰. 融合对比度拉伸的地铁隧道环境图像复原算法[J]. 计算机工程, 2024, 50(1): 224-231.

Kang YANG, Yu REN, Xuejie WU. Subway Tunnel Environment Image Restoration Algorithm Fusing Contrast Stretching[J]. Computer Engineering, 2024, 50(1): 224-231.

http://www.ecice06.com/CN/Y2024/V50/I1/224

图/表 13

图1 图像退化与复原原理

Fig.1 Principles of image degradation and restoration

图2 融合对比度拉伸的图像复原算法流程

Fig.2 Procedure of image restoration algorithm fusing contrast stretching

图3 V分量灰度直方图

Fig.3 V component grayscale histogram

图4 对比度拉伸函数

Fig.4 Function of contrast stretching

图5 增强效果示例

Fig.5 Example of enhancement effect

图6 模糊图像频谱

Fig.6 Spectrum of blurred images

图7 刀刃法原理

Fig.7 Principle of knife edge method

图8 区域划分示意图

Fig.8 Schematic diagram of the regional division

图9 刀刃区域的处理

Fig.9 Processing of the knife edge region

图10 点扩散函数计算过程

Fig.10 Procedure of PSF calculation

图11 不同算法增强结果

Fig.11 Enhancement results of different algorithms

图12 不同算法实验结果及局部边缘检测图

Fig.12 Experimental results and local edge detection diagrams of different algorithms

参考文献 28

1	交通运输部: 9月全国城市轨道交通新增运营里程96.8公里[EB/OL]. [2022-10-25]. http://www.gov.cn/xinwen/2022-10/11/content_5717394.htm.
	Ministry of Transport: 96.8 kilometers of new operating mileage for urban rail transit nationwide in september[EB/OL]. [2022-10-25]. http://www.gov.cn/xinwen/2022-10/11/content_5717394.htm. (in Chinese)
2	李腾飞, 贾振红. 基于照度调整的自适应低照度视频增强算法. 激光杂志, 2021, 42(8): 45- 51.
	LI T F, JIA Z H. Adaptive low-light video enhancement algorithm based on illumination adjustment. Laser Journal, 2021, 42(8): 45- 51.
3	GUO X J, LI Y, LING H B. LIME: low-light image enhancement via illumination map estimation. IEEE Transactions on Image Processing, 2017, 26(2): 982- 993. doi: 10.1109/TIP.2016.2639450
4	张恩齐, 孔令胜, 郭俊达, 等. 基于Retinex理论的低照度图像对比度增强算法. 机电工程技术, 2022, 51(3): 95-98, 144.
	ZHANG E Q, KONG L S, GUO J D, et al. Low illumination images based on Retinex theory contrast enhancement algorithm. Mechanical & Electrical Engineering Technology, 2022, 51(3): 95-98, 144.
5	ZHANG H Y, ZHAO J D. RetinexNet low illumination image enhancement algorithm in HSV space. Laser & Optoelectronics Progress, 2020, 57(20): 201504.
6	SHI Y M, WU X P, ZHU M. Low-light image enhancement algorithm based on Retinex and generative adversarial network[EB/OL]. [2022-10-25]. https://arxiv.org/abs/1906.06027.pdf.
7	HUANG W, ZHU Y F, HUANG R. Low light image enhancement network with attention mechanism and Retinex model. IEEE Access, 2020, 8, 74306- 74314. doi: 10.1109/ACCESS.2020.2988767
8	JEBADASS J R, BALASUBRAMANIAM P. Low light enhancement algorithm for color images using intuitionistic fuzzy sets with histogram equalization. Multimedia Tools and Applications, 2022, 81(6): 8093- 8106. doi: 10.1007/s11042-022-12087-9
9	K S S, JAGADALE B N, BHAT J S. Image enhancement using recursive separate standard intensity deviation based clipped sub image histogram equalization. International Journal of Innovative Technology and Exploring Engineering, 2020, 9(3): 782- 787. doi: 10.35940/ijitee.C8515.019320
10	CHENG H, LONG W, LI Y Y, et al. Two low illuminance image enhancement algorithms based on grey level mapping. Multimedia Tools and Applications, 2021, 80(5): 7205- 7228. doi: 10.1007/s11042-020-09919-x
11	PENG Y T, LIAO H H, CHEN C F. Two-exposure image fusion based on optimized adaptive gamma correction. Sensors, 2021, 22(1): 24. doi: 10.3390/s22010024
12	LIU S G, ZHANG Y. Detail-preserving underexposed image enhancement via optimal weighted multi-exposure fusion. IEEE Transactions on Consumer Electronics, 2019, 65(3): 303- 311. doi: 10.1109/TCE.2019.2893644
13	张聿, 刘世光. 基于分区曝光融合的不均匀亮度视频增强. 计算机辅助设计与图形学学报, 2017, 29(12): 2317- 2322.
	ZHANG Y, LIU S G. Non-uniform illumination video enhancement based on zone system and fusion. Journal of Computer-Aided Design & Computer Graphics, 2017, 29(12): 2317- 2322.
14	曹学影, 王琳, 谭覃燕, 等. 自适应双刃边法模糊红外图像复原. 光学精密工程, 2022, 30(5): 630- 640.
	CAO X Y, WANG L, TAN Q Y, et al. Restoration of blurred infrared image by adaptive double edge method. Optics and Precision Engineering, 2022, 30(5): 630- 640.
15	JIN X, ZHANG L, SHAN C W, et al. Dual prior learning for blind and blended image restoration. IEEE Transactions on Image Processing, 2022, 31, 1042- 1056. doi: 10.1109/TIP.2021.3135482
16	师雪艳. 模糊图像自适应复原算法研究[D]. 西安: 西安电子科技大学, 2015.
	SHI X Y. Research on adaptive restoration algorithm of blurred image[D]. Xi'an: Xidian University, 2015. (in Chinese)
17	SHI J M. Noise data removal and image restoration based on partial differential equation in sports image recognition technology[EB/OL]. [2022-10-25]. https://doaj.org/article/7d1a798fb1ab401c8428d95dbcd89a12.
18	XU B, FANG Y, BIAN Z, et al. Low-light image restoration using discrete cosine transform and conditional generative adversarial network. Ukrainian Journal of Physical Optics, 2021, 22(4): 225- 238. doi: 10.3116/16091833/22/4/225/2021
19	刘迪. 面向夜晚图像的去模糊算法研究[D]. 北京: 华北电力大学, 2021.
	LIU D. Research on night image deblurring algorithm[D]. Beijing: North China Electric Power University, 2021. (in Chinese)
20	CHOI T. IKONOS satellite on orbit Modulation Transfer Function(MTF) measurement using edge and pulse method[D]. Brookings, USA: South Dakota State University, 2002.
21	杨利红. 大视场航天相机遥感图像复原研究[D]. 长春: 中国科学院研究生院, 2012.
	YANG L H. Research on remote sensing image restoration of large field of view space camera[D]. Changchun: Graduate University of Chinese Academy of Sciences, 2012. (in Chinese)
22	YING Z Q, LI G, GAO W. A bio-inspired multi-exposure fusion framework for low-light image enhancement[EB/OL]. [2022-10-25]. https://arxiv.org/abs/1711.00591.pdf.
23	JOBSON D J, RAHMAN Z, WOODELL G A. A multiscale Retinex for bridging the gap between color images and the human observation of scenes. IEEE Transactions on Image Processing, 1997, 6(7): 965- 976. doi: 10.1109/83.597272
24	DONG X A, WANG G A, PANG Y, et al. Fast efficient algorithm for enhancement of low lighting video[C]//Proceedings of 2011 IEEE International Conference on Multimedia and Expo. Washington D. C., USA: IEEE Press, 2011: 1-6.
25	YING Z Q, LI G, REN Y R, et al. A new low-light image enhancement algorithm using camera response model[C]//Proceedings of 2017 IEEE International Conference on Computer Vision Workshops. Washington D. C., USA: IEEE Press, 2017: 3015-3022.
26	GUO C L, LI C Y, GUO J C, et al. Zero-reference deep curve estimation for low-light image enhancement[C]//Proceedings of 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Washington D. C., USA: IEEE Press, 2020: 1-10.
27	WEI K X, FU Y, YANG J L, et al. A physics-based noise formation model for extreme low-light raw denoising[C]//Proceedings of 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Washington D. C., USA: IEEE Press, 2020: 1-8.
28	ABDOLI M, NASIRI F, BRAULT P, et al. Quality assessment tool for performance measurement of image contrast enhancement methods. IET Image Processing, 2019, 13(5): 833- 842. doi: 10.1049/iet-ipr.2018.5520

[1]	杨瑞君, 秦晋京, 程燕. 基于生成对抗网络的自然场景低照度增强模型[J]. 计算机工程, 2024, 50(1): 279-288.
[2]	杨谢柳, 门国文, 梁文峰, 王丹, 谢正义, 范慧杰. 水下图像增强复原对深度学习目标检测精度的影响研究[J]. 计算机工程, 2023, 49(11): 247-256.
[3]	庄子杰, 范之国, 金海红, 宫凯强. 基于水体衰减系数反演的水下图像复原方法[J]. 计算机工程, 2023, 49(1): 258-269.
[4]	高海韬, 李丹宁, 王彬, 唐鑫鑫. 运动模糊图像PSF参数估计方法改进及图像复原[J]. 计算机工程, 2022, 48(9): 197-203,212.
[5]	徐超越, 余映, 何鹏浩, 李淼, 马玉辉. 基于U-Net的多尺度低照度图像增强网络[J]. 计算机工程, 2022, 48(8): 215-223.
[6]	丁泳钧, 黄山. 一种用于图像去雾的改进生成对抗网络[J]. 计算机工程, 2022, 48(6): 207-212.
[7]	缪斯, 祝永新. 针对图像盲去模糊的可微分神经网络架构搜索方法[J]. 计算机工程, 2021, 47(9): 313-320.
[8]	刘文, 杨梅芳, 聂江天, 章阳, 杨和林, 熊泽辉. 基于局部生成对抗网络的水上低照度图像增强[J]. 计算机工程, 2021, 47(5): 16-23.
[9]	甄诚, 杨永胜, 李元祥, 钟娟娟. 基于多尺度生成对抗网络的大气湍流图像复原[J]. 计算机工程, 2021, 47(11): 227-233.
[10]	韩梦妍, 李良荣, 蒋凯. 基于光照图估计的Retinex低照度图像增强算法[J]. 计算机工程, 2021, 47(10): 201-206.
[11]	王小元,张红英,吴亚东,刘言. 基于视觉感知的低照度图像增强算法[J]. 计算机工程, 2016, 42(8): 249-254.
[12]	卢来,王军民,范锐. 基于自适应增强的图像二值描述子[J]. 计算机工程, 2016, 42(6): 230-234,240.
[13]	潘磊,袁小珂,周欢. 单幅图像梯度域实时去雾算法[J]. 计算机工程, 2016, 42(10): 266-270.
[14]	方帅,范东,于磊,曹风云. 一种基于L0正则化的模糊复原算法[J]. 计算机工程, 2016, 42(1): 254-259.
[15]	汪畅，顾行发，余涛，谢勇，谢燕华，李小英. 资源三号卫星在轨MTF测量与图像复原[J]. 计算机工程, 2014, 40(4): 237-241,246.

选择文件类型/文献管理软件名称

选择包含的内容