医用内窥镜图像的高光移除算法

doi:10.11772/j.issn.1001-9081.2022030478

《计算机应用》唯一官方网站 ›› 2023, Vol. 43 ›› Issue (4): 1278-1283.DOI: 10.11772/j.issn.1001-9081.2022030478

所属专题：多媒体计算与计算机仿真

• 多媒体计算与计算机仿真 • 上一篇下一篇

医用内窥镜图像的高光移除算法

池月, 李正平(), 徐超, 冯博

安徽大学集成电路学院，合肥 230601

收稿日期:2022-04-13 修回日期:2022-09-09 接受日期:2022-09-14 发布日期:2023-01-11 出版日期:2023-04-10
通讯作者: 李正平
作者简介:池月（1996—），男，河南南阳人，硕士研究生，主要研究方向：图像处理；
徐超（1962—），男，安徽合肥人，教授，博士，主要研究方向：算法设计、计算复杂性；
冯博（1990—），男，安徽合肥人，博士研究生，主要研究方向：深度学习、图像处理。
基金资助:
国家重点研发计划项目(2019YFC0117800)

Highlight removal algorithm for medical endoscopic images

Yue CHI, Zhengping LI(), Chao XU, Bo FENG

School of Integrated Circuits，Anhui University，Hefei Anhui 230601，China

Received:2022-04-13 Revised:2022-09-09 Accepted:2022-09-14 Online:2023-01-11 Published:2023-04-10
Contact: Zhengping LI
About author:CHI Yue， born in 1996， M. S. candidate. His research interests include image processing.
XU Chao， born in 1962， Ph. D.， professor. His research interests include algorithm design， computational complexity.
FENG Bo， born in 1990， Ph. D. candidate. His research interests include deep learning， image processing.
Supported by:
National Key Research and Development Program of China(2019YFC0117800)

摘要/Abstract

摘要：

现有的内窥镜图像高光移除算法往往存在移除结构不合理、颜色失真等情况，而这会导致病灶识别算法和图像增强算法产生错误的结果。为了解决以上问题，在高光定位方面，提出了一种基于暗区域内生长和Scharr 滤波结合的相对高光定位方法；在高光填充方面，提出了一种改进的Crinminisi算法。首先，通过统计大量数据，限定搜索范围，从而提高填充效率；其次，改进优先权统计范围，以避免重复的无意义的计算；最后，针对不同区域的自适应模板，合理重建纹理。在选取不同人体组织的内窥镜图像数据集上进行实验，相较于基于双色反射模型的方法、自适应鲁棒主成分分析（RPCA）方法、基于热扩散的方法和原始Criminisi算法，所提算法的自然图像质量评估（NIQE）值均为最小；相较于RPCA方法、基于热扩散的方法和原始Crimnisi算法，所提算法的运行时间也均为最少。实验结果表明，所提算法不仅比其他算法具有更好的客观图像指标，而且相较于原始Criminisi算法在效率上有近百倍的提升。

关键词: 内窥镜图像处理, 高光定位, 高光填充, 区域分割, Criminisi算法

Abstract:

The existing endoscopic image highlight removal algorithms often have some problems such as unreasonable removal structure and color distortion， which leads to the wrong results of the focus recognition algorithms and image enhancement algorithms. In order to solve the above problems， in the aspect of highlight localization， a method based on the combination of growth in dark region and Scharr filtering was proposed to locate relative highlight； in the aspect of highlight filling， an improved Crinminisi algorithm was proposed. Firstly， through the statistics on a huge amount of data， the search scope was limited and the filling efficiency was increased. Secondly， the statistical scope of priority was improved to avoid repeated meaningless calculations. Finally， the reasonable reconstruction of texture was performed according to the adaptive templates of different regions. Experiments were carried out on endoscopic image dataset of different human tissues， compared with the dichromatic reflection model based method， the Robust Principle Component Analysis （RPCA） method， the thermal diffusion method and the original Criminisi algorithm， the Natural Image Quality Evaluator （NIQE） value of the proposed algorithm was the lowest. Compared with the RPCA method， the thermal diffusion method and the original Crimnisi algorithm， the running time of the proposed algorithm was the lowest. Experimental results show that the proposed algorithm not only has better objective image indicators than other algorithms， but also has a nearly 100-fold improvement in efficiency compared to the original Criminisi algorithm.

Key words: endoscopic image processing, highlight localization, highlight filling, region segmentation, Criminisi algorithm

中图分类号:

TP391

池月, 李正平, 徐超, 冯博. 医用内窥镜图像的高光移除算法[J]. 计算机应用, 2023, 43(4): 1278-1283.

Yue CHI, Zhengping LI, Chao XU, Bo FENG. Highlight removal algorithm for medical endoscopic images[J]. Journal of Computer Applications, 2023, 43(4): 1278-1283.

图/表 10

参考文献 27

1	MORO C， ŠTROMBERGA Z， RAIKOS A， et al. The effectiveness of virtual and augmented reality in health sciences and medical anatomy［J］. Anatomical Sciences Education， 2017， 10（6）： 549-559. 10.1002/ase.1696
2	BERNHARDT S， NICOLAU S A， SOLER L， et al. The status of augmented reality in laparoscopic surgery as of 2016［J］. Medical Image Analysis， 2017， 37： 66-90. 10.1016/j.media.2017.01.007
3	BUCHS N C， VOLONTÉ F， PUGIN F， et al. Augmented environments for the targeting of hepatic lesions during image-guided robotic liver surgery［J］. Journal of Surgical Research， 2013， 184（2）： 825-831. 10.1016/j.jss.2013.04.032
4	WHITE S， KALKOFEN D， SANDOR C. Visualization in mixed reality environments［C］// Proceedings of the 2011 IEEE International Symposium on Mixed and Augmented Reality - Arts， Media and Humanities. Piscataway： IEEE， 2011： 1-1. 10.1109/ismar.2011.6092361
5	KERSTEN-OERTEL M， JANNIN P， COLLINS D L. The state of the art of visualization in mixed reality image guided surgery［J］. Computerized Medical Imaging and Graphics， 2013， 37（2）： 98-112. 10.1016/j.compmedimag.2013.01.009
6	OH J， HWANG S， LEE J， et al. Informative frame classification for endoscopy video［J］. Medical Image Analysis， 2007， 11（2）： 110-127. 10.1016/j.media.2006.10.003
7	卢桂荣，汤景凡，姜明. 基于快速双边滤波的图像高光去除研究［J］. 计算机工程与应用， 2014， 50（10）： 176-179， 207. 10.3778/j.issn.1002-8331.1310-0359
	LU G R， TANG J F， JIANG M. Research on image highlight removal based on fast bilateral filtering［J］. Computer Engineering and Applications， 2014， 50（10）： 176-179， 207. 10.3778/j.issn.1002-8331.1310-0359
8	REN W H， TIAN J D， TANG Y D. Specular reflection separation with color-lines constraint［J］. IEEE Transactions on Image Processing. 2017， 26（5）： 2327-2337. 10.1109/tip.2017.2675204
9	YANG Q X， TANG J H， AHUJA N. Efficient and robust specular highlight removal［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence， 2015， 37（6）： 1304-1311. 10.1109/tpami.2014.2360402
10	SHEN H L， ZHANG H G， SHAO S J， et al. Chromaticity-based separation of reflection components in a single image［J］. Pattern Recognition， 2008， 41（8）：2461-2469. 10.1016/j.patcog.2008.01.026
11	KHAN H A， THOMAS J B， HARDEBERG J Y. Analytical survey of highlight detection in color and spectral images［C］// Proceedings of the 2017 International Workshop on Computational Color Imaging， LNCS 10213. Cham： Springer， 2017：197-208.
12	JOSPIN L V， BAECHLER G， SCHOLEFIELD A. Embedded polarizing filters to separate diffuse and specular reflection［C］// Proceedings of the 2018 Asian Conference on Computer Vision， LNCS 11362. Cham： Springer， 2019：3-18.
13	SUO J L， AN D S， JI X Y， et al. Fast and high quality highlight removal from a single image［J］. IEEE Transactions on Image Processing， 2016， 25（11）： 5441-5454. 10.1109/tip.2016.2605002
14	SHEN H L， ZHENG Z H. Real-time highlight removal using intensity ratio［J］. Applied Optics， 2013， 52（19）： 4483-4493. 10.1364/ao.52.004483
15	LI R Y， PAN J J， SI Y Q， et al. Specular reflections removal for endoscopic image sequences with adaptive-RPCA decomposition［J］. IEEE Transactions on Medical Imaging， 2020， 39（2）： 328-340. 10.1109/tmi.2019.2926501
16	BOBROW T L， MAHMOOD F， INSERNI M， et al. DeepLSR： a deep learning approach for laser speckle reduction［J］. Biomedical Optics Express， 2019， 10（6）：2869-2882. 10.1364/boe.10.002869
17	KUDVA V， PRASAD K， GURUVARE S. Detection of specular reflection and segmentation of cervix region in uterine cervix images for cervical cancer screening［J］. IRBM， 2017， 38（5）： 281-291. 10.1016/j.irbm.2017.08.003
18	SAINT-PIERRE C A， BOISVERT J， GRIMARD G， et al. Detection and correction of specular reflections for automatic surgical tool segmentation in thoracoscopic images［J］. Machine Vision and Applications， 2011， 22（1）： 171-180. 10.1007/s00138-007-0099-6
19	LIANG Z S， YANG G B， DING X L， et al. An efficient forgery detection algorithm for object removal by exemplar-based image inpainting［J］. Journal of Visual Communication and Image Representation， 2015， 30： 75-85. 10.1016/j.jvcir.2015.03.004
20	QURESHI M A， DERICHE M， BEGHDADI A， et al. A critical survey of state-of-the-art image inpainting quality assessment metrics［J］. Journal of Visual Communication and Image Representation， 2017， 49： 177-191. 10.1016/j.jvcir.2017.09.006
21	BERTALMIO M， BERTOZZI A L， SAPIRO G. Navier-stokes， fluid dynamics， and image and video inpainting［C］// Proceedings of the 2001 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2001： I-355-I-362. 10.1109/cvpr.2001.990916
22	CRIMINISI A， PÉREZ P， TOYAMA K. Region filling and object removal by exemplar-based image inpainting［J］. IEEE Transactions on Image Processing， 2004， 13（9）： 1200-1212. 10.1109/tip.2004.833105
23	BUYSSENS P， DAISY M， TSCHUMPERLÉ D， et al. Exemplar-based inpainting： technical review and new heuristics for better geometric reconstructions［J］. IEEE Transactions on Image Processing， 2015， 24（6）： 1809-1824.
24	王凤随，刘正男，付林军. 基于信息熵和梯度因子的改进Criminisi图像修复方法［J］. 激光与光电子学进展， 2020， 57（22）： No.221006. 10.3788/lop57.221006
	WANG F S， LIU Z N， FU L J. An improved Criminisi image inpainting method based on information entropy and gradient factor［J］. Progress in Lasers and Optoelectronics， 2020， 57（22）： No.221006. 10.3788/lop57.221006
25	KUMAR S， SRIVASTAVA A K， JHA A， et al. Implementation of linear structuring element in OpenCV for blood vessel segmentation from color fundus images［C］// Proceedings of the 10th International Conference on Computing， Communication and Networking Technologies. Piscataway： IEEE， 2019：1-5. 10.1109/icccnt45670.2019.8944610
26	MITTAL A， SOUNDARARAJIAN R， BOVIK A C. Making a “completely blind” image quality analyzer［J］. IEEE Signal Processing Letters， 2013， 20（3）： 209-212. 10.1109/lsp.2012.2227726
27	TELEA A. An image inpainting technique based on the fast marching method［J］. Journal of Graphics Tools， 2004， 9（1）： 23-34. 10.1080/10867651.2004.10487596

算法	食道1	食道2	咽部	腹腔1	腹腔2	腹腔3	腹腔4	鼻部	子宫1	子宫2
原图	7.992 8	5.261 3	7.111 6	6.926 3	6.950 3	6.443 1	5.959 0	7.235 7	3.699 5	6.765 2
双色反射模型^［10］	7.861 9	5.200 5	7.060 5	6.761 9	6.585 0	6.463 5	6.885 3	7.606 7	3.763 9	6.735 7
RPCA^［15］	7.162 1	5.083 3	6.617 5	6.614 4	6.378 8	5.829 7	6.647 8	7.253 3	3.771 7	6.733 9
热扩散^［27］	6.908 7	5.120 8	6.453 6	6.705 3	6.359 0	5.967 9	5.610 7	7.350 5	3.868 9	6.757 9
Criminisi^［22］	6.543 3	5.092 7	6.282 5	6.606 0	6.128 1	5.654 0	5.795 3	7.233 5	3.745 3	6.884 0
本文算法	6.164 2	5.032 3	5.931 5	6.581 8	5.514 6	5.464 2	5.535 0	7.204 5	3.683 9	6.723 7

算法	食道1	食道2	咽部	腹腔1	腹腔2	腹腔3	腹腔4	鼻部	子宫1	子宫2
原图	7.992 8	5.261 3	7.111 6	6.926 3	6.950 3	6.443 1	5.959 0	7.235 7	3.699 5	6.765 2
双色反射模型^［10］	7.861 9	5.200 5	7.060 5	6.761 9	6.585 0	6.463 5	6.885 3	7.606 7	3.763 9	6.735 7
RPCA^［15］	7.162 1	5.083 3	6.617 5	6.614 4	6.378 8	5.829 7	6.647 8	7.253 3	3.771 7	6.733 9
热扩散^［27］	6.908 7	5.120 8	6.453 6	6.705 3	6.359 0	5.967 9	5.610 7	7.350 5	3.868 9	6.757 9
Criminisi^［22］	6.543 3	5.092 7	6.282 5	6.606 0	6.128 1	5.654 0	5.795 3	7.233 5	3.745 3	6.884 0
本文算法	6.164 2	5.032 3	5.931 5	6.581 8	5.514 6	5.464 2	5.535 0	7.204 5	3.683 9	6.723 7

样图	分辨率	高光占比/%	运行时间/s
样图	分辨率	高光占比/%	RPCA^［15］	Criminisi^［22］	本文算法
picture1	863×1 027	2.60	42.657	1 755.800	21.072
picture2	626×721	2.54	18.022	434.800	11.531
picture3	823×1 062	1.89	33.131	1 303.080	13.314
picture4	343×338	1.21	3.207	15.741	2.271
picture5	422×527	1.66	7.864	82.974	4.151
picture6	847×1 001	3.67	33.067	2 142.800	30.423
picture7	492×475	2.31	8.349	112.633	5.769
picture8	527×549	2.26	12.240	88.946	11.478

样图	分辨率	高光占比/%	运行时间/s
样图	分辨率	高光占比/%	RPCA^［15］	Criminisi^［22］	本文算法
picture1	863×1 027	2.60	42.657	1 755.800	21.072
picture2	626×721	2.54	18.022	434.800	11.531
picture3	823×1 062	1.89	33.131	1 303.080	13.314
picture4	343×338	1.21	3.207	15.741	2.271
picture5	422×527	1.66	7.864	82.974	4.151
picture6	847×1 001	3.67	33.067	2 142.800	30.423
picture7	492×475	2.31	8.349	112.633	5.769
picture8	527×549	2.26	12.240	88.946	11.478

[1]	汤慧, 周明全, 耿国华. 基于区域分割的低覆盖点云配准算法[J]. 计算机应用, 2019, 39(11): 3355-3360.
[2]	魏震宇, 文畅, 谢凯, 贺建飚. 光流估计下的移动端实时人脸检测[J]. 计算机应用, 2018, 38(4): 1146-1150.
[3]	李梦雪, 翟东海, 孟红月, 曹大命. 划分特征子区域的图像修复算法[J]. 计算机应用, 2017, 37(12): 3541-3546.
[4]	翟东海肖杰鱼江李同亮. 基于自适应模板的图像修复算法[J]. 计算机应用, 2013, 33(10): 2891-2894.
[5]	傅沈文. 复杂环境下的运动车辆检测[J]. 计算机应用, 2012, 32(06): 1581-1584.
[6]	王海洋车生兵舒旭. 基于信息熵与动态区域分割的半脆弱数字水印算法[J]. 计算机应用, 2011, 31(08): 2169-2173.
[7]	刘涛张登福何宜宝. 基于区域分割和非下采样Contourlet变换的多聚焦图像融合算法[J]. 计算机应用, 2010, 30(10): 2805-2807.
[8]	周强锋田铮李小斌刘丙涛. 基于Gomory-Hu算法有效实现的图像区域分割[J]. 计算机应用, 2008, 28(3): 671-673.

医用内窥镜图像的高光移除算法

Highlight removal algorithm for medical endoscopic images

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 10

参考文献 27

相关文章 8

编辑推荐

Metrics