基于OTSU和区域生长的电厂管道缺陷检测与分割

彭道刚; 尹磊; 戚尔江; 胡捷; 杨晓伟

基于OTSU和区域生长的电厂管道缺陷检测与分割

1.
上海电力大学自动化工程学院，上海 200090
2.
宝山钢铁股份有限公司电厂，上海 201900
3.
上海宝信软件股份有限公司，上海 201203

详细信息

作者简介:
彭道刚（1977-），男，教授，博士，从事智能发电、能源互联网、电力巡检机器人研究。E-mail：pengdaogang@126.com

中图分类号: TP391.41
计量
- 文章访问数: 193
- HTML全文浏览量: 136
- PDF下载量: 28
- 被引次数: 0
出版历程
- 收稿日期: 2020-08-16
- 修回日期: 2020-10-24
- 刊出日期: 2021-05-22

Power Plant Pipeline Defect Detection and Segmentation Based on Otsu's and Region Growing Algorithms

1.
College of Automation Engineering, Shanghai University of Electric Power, Shanghai 200090, China
2.
Power plant of Baoshan Iron and Steel Co., Ltd, Shanghai 201900, China
3.
Shanghai Baosight Software Co., Ltd, Shanghai 201203, China

摘要

摘要: 针对电厂高温管道红外图像背景复杂、干扰较多的特点，结合电厂巡检机器人系统对图像处理算法的需求，提出了基于改进二维最大类间方差法（OTSU）和区域生长法的电厂高温管道缺陷定位与分割方法。将红外图像灰度化后，通过改进二维OTSU进行预分割，提取出管道区域；基于管道区域灰度直方图，结合邻域灰度均值，实现多种子点的自动检测与定位；采用基于生长区域灰度均值和标准差的自适应阈值以及基于Prewitt算子的梯度幅值改进的生长准则完成缺陷区域的分割。实验证明，所提算法不仅能实现电厂高温管道多缺陷自动检测与定位，而且能精确地提取出缺陷区域，准确性高且具有良好的实时性。
- 电厂巡检机器人 /
- 红外图像 /
- 高温管道缺陷分割 /
- 改进二维OTSU /
- 改进区域生长法
Abstract: In this study, we consider the complex background and high interference that adversely affect infrared images of high-temperature pipelines in power plants and the requirements of image processing algorithms for inspection robot systems. We propose a high-temperature pipeline defect detection and extraction method based on an improved two-dimensional Otsu and region growth algorithms. After grayscale conversion, a 2D Otsu method was used to extract the pipeline area. Based on the grayscale histogram of the pipeline region and the average gray value of the neighborhood, automatic detection and positioning of multiple sub-points were realized. The segmentation of the defect area was accomplished using two methods. The adaptive threshold was determined based on the gray mean and standard deviation values of the growth area, while the growth criterion was improved using the gradient amplitude of the Prewitt operator. The experimental results show that the proposed algorithm can not only realize the automatic detection and positioning of various defects in high-temperature pipelines of power plants, but it additionally segments the defect regions more accurately with high accuracy and good real-time performance.
- power plant inspection robot /
- infrared image /
- high temperature pipeline defect detection /
- improved two-dimensional OTSU /
- improved region growing algorithm

HTML全文

图 1 缺陷检测系统流程图

Figure 1. Flow chart of defect detection system

下载: 全尺寸图片幻灯片

图 2 二维直方图

Figure 2. Two-dimensional histogram

下载: 全尺寸图片幻灯片

图 3 区域生长法流程图

Figure 3. Flow chart of region growing algorithm

下载: 全尺寸图片幻灯片

图 4 多种子点自动选取流程图

Figure 4. Flow chart for automatic selection of multiple seed points

下载: 全尺寸图片幻灯片

图 5 像素3×3邻域图

Figure 5. Pixel neighborhood map

下载: 全尺寸图片幻灯片

图 6 电厂高温管道红外图像

Figure 6. Infrared image of high temperature pipeline in a power plant

下载: 全尺寸图片幻灯片

图 7 本文方法实验结果图

Figure 7. The experimental results of this paper's method

下载: 全尺寸图片幻灯片

图 8 3种算法结果对比图

Figure 8. Comparison of three algorithm results

下载: 全尺寸图片幻灯片

表 1 缺陷区域像素点个数表

Table 1. Number of pixels in defect area

Algorithm	Defect image	Defect one	Defect two	Defect three	Defect four
Algorithm of this article	Scene one	929	-	-	-
	Scene two	796	489	-	-
	Scene three	249	129	109	96
Traditional regional growth algorithm	Scene one	930	-	-	-
	Scene two	827	551	-	-
	Scene three	117	79	59	80
Algorithm of literature ^[10]	Scene one	941	-	-	-
	Scene two	776	493	-	-
	Scene three	226	118	106	95

下载: 导出CSV

表 2 算法性能对比表

Table 2. Algorithm performance comparison table

Algorithm	False detection	Seed point selection	Time/(s/sheet)
Algorithm of this article	No	Auto	0.331
Traditional regional growth algorithm	No	Manual	≥3.5
Algorithm of literature ^[10]	Yes	Auto	0.254

下载: 导出CSV

参考文献(19)

[1]	王丞浩. 基于物联网的电厂智能巡检系统移动端设计与实现[D]. 吉林: 东北电力大学, 2019. WANG Chenghao. Design and Implement of Mobile Terminal Power Plant Intelligent Patrol System Based on IOT[D]. Jilin: Northeast Electric Power University, 2019.
[2]	华志刚, 郭荣, 汪勇. 燃煤智能发电的关键技术[J]. 中国电力, 2018, 51(10): 8-16. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGDL201810004.htm HUA Zhigang, GUO Rong, WANG Yong. Key technologies for intelligent coal-fired power generation[J]. Electric Power, 2018, 51(10): 8-16. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGDL201810004.htm
[3]	张燕东, 田磊, 李茂清, 等. 智能巡检机器人系统在火力发电行业的应用研发及示范[J]. 中国电力, 2017, 50(10): 1-7. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGDL201710001.htm ZHANG Yandong, TIAN Lei, LI Maoqing, et al. Application and development of intelligent inspection robot system in thermal power plant[J]. Electric Power, 2017, 50(10): 1-7. https://www.cnki.com.cn/Article/CJFDTOTAL-ZGDL201710001.htm
[4]	徐蔚波, 刘颖, 章浩伟. 基于区域生长的图像分割研究进展[J]. 北京生物医学工程, 2017, 36(3): 317-322. doi: 10.3969/j.issn.1002-3208.2017.03.16. XU Weibo, LIU Ying, ZHANG Haowei. Research progress in image segmentation based on region growing[J]. Beijing Biomedical Engineering, 2017, 36(3): 317-322. doi: 10.3969/j.issn.1002-3208.2017.03.16.
[5]	Jain Preetha M M S, Padmasuresh L, Bosco M J. Firefly based region growing and region merging for image segmentation[C]//2016 International Conference on Emerging Technological Trends (ICETT), 2016: 1-9.
[6]	彭双, 肖昌炎. 结合区域生长与模糊连接度的肺气管树分割[J]. 计算机工程与应用, 2016, 52(13): 201-205. doi: 10.3778/j.issn.1002-8331.1407-0623 PENG Shuang, XIAO Changyan. Segmentation of pulmonary airway tree by combining region growing and fuzzy connectedness[J]. Computer Engineering and Applications, 2016, 52(13): 201-205. doi: 10.3778/j.issn.1002-8331.1407-0623
[7]	Senthilkumar B, Umamaheswari G, Karthik J. A novel region growing segmentation algorithm for the detection of breast cancer[C]//2010IEEE International Conference on Computational Intelligence and Computing Research, 2010: 1-4.
[8]	SONG L, LV Y, YANG B, et al. Segmentation of breast masses using adaptive region growing[C]//Ulaanbaatar, Ifost, 2013: 77-81.
[9]	倪豪, 郑慧峰, 王月兵, 等. 基于自动种子区域生长的超声B图像缺陷分割方法[J]. 计量学报, 2018, 39(6): 878-883. doi: 10.3969/j.issn.1000-1158.2018.06.24 NI Hao, ZHEN Huifeng, WANG Yuebing, et al. Ultrasonic B image defect segmentation Method Based on automatic seeded region growing[J]. Acta Metrologica Sinica, 2018, 39(6): 878-883. doi: 10.3969/j.issn.1000-1158.2018.06.24
[10]	李小磊, 曾曙光, 郑胜, 等. 基于滑动滤波和自动区域生长的陶瓷瓦表面裂纹检测[J]. 激光与光电子学进展, 2019, 56(21): 49-55. https://www.cnki.com.cn/Article/CJFDTOTAL-JGDJ201921006.htm LI Xiaolei, ZENG Shuguang, ZHENG Sheng, et al. Surface crack detection of ceramic tile based on sliding filter and automatic region growth[J]. Laser & Optoelectronics Progress, 2019, 56(21): 49-55. https://www.cnki.com.cn/Article/CJFDTOTAL-JGDJ201921006.htm
[11]	施兢业, 刘俊. 基于改进区域生长法的电力设备红外图像分割[J]. 光学技术, 2017, 43(4): 381-384. https://www.cnki.com.cn/Article/CJFDTOTAL-GXJS201704019.htm SHI Jingye, LIU Jun. Metation based on modified region growing algorithm[J]. Optical Technique, 2017, 43(4): 381-384. https://www.cnki.com.cn/Article/CJFDTOTAL-GXJS201704019.htm
[12]	胡淋波, 姚建刚, 孔维辉, 等. 基于红外图像的高压绝缘子串自动定位方法[J]. 红外技术, 2015, 37(12): 1047-1051. http://hwjs.nvir.cn/article/id/hwjs201512011 HU Linbo, YAO Jiangang, KONG Weihui, et al. High voltage insulator string automatic location method based on infrared image[J]. Infrared Technology, 2015, 37(12): 1047-1051. http://hwjs.nvir.cn/article/id/hwjs201512011
[13]	宋银龙. 基于二维Otsu和模糊聚类的图像分割的研究及应用[D]. 合肥: 合肥工业大学, 2012. SONG Yinlong. Research and application of image segmentation based on two-dimensional otsu and fuzzy clustering[D]. Hefei: Hefei University of Technology, 2012.
[14]	倪伟传, 许志明, 刘少江, 等. 复杂环境下的自适应红外目标分割算法[J]. 红外技术, 2019, 41(4): 357-363. http://hwjs.nvir.cn/article/id/hwjs201904010 NI Weichuan, XU Zhiming, LIU Shaojiang, et al. Adaptive Infrared Target Segmentation Algorithm in Complex Environment[J]. Infrared Technology, 2019, 41(4): 357-363. http://hwjs.nvir.cn/article/id/hwjs201904010
[15]	SHAO L, ZHANG Y, LI J, et al. Research on High Temperature Region of Infrared Pipeline Image Based on Improved Two-Dimensional-Otsu[J]. Spectroscopy and Spectral Analysis, 2019, 39(5): 1637-1642.
[16]	彭启伟, 罗旺, 冯敏, 等. 改进二维Otsu法和果蝇算法结合的图像分割方法[J]. 计算机应用, 2017, 37(S2): 193-197. https://www.cnki.com.cn/Article/CJFDTOTAL-JSJY2017S2047.htm PENG Qiwei, LUO Wang, FENG Min, et al. Novel method for image segmentation based on improved two-dimensional Otsu and fruit fly algorithm[J]. Journal of Computer Applications, 2017, 37(S2): 193-197. https://www.cnki.com.cn/Article/CJFDTOTAL-JSJY2017S2047.htm
[17]	周云燕, 杨坤涛, 黄鹰. 基于最小类内离散度的改进Otsu分割方法的研究[J]. 华中科技大学学报: 自然科学版, 2007, 35(2): 101-103. https://www.cnki.com.cn/Article/CJFDTOTAL-HZLG200702030.htm ZHOU Yunyan, YANG Kuntao, HUANG Ying. Improved Otsu thresholding based on minimum inner-cluster variance[J]. Journal of Huazhong University of Science and Technology: Natural Science Edition, 2007, 35(2): 101-103. https://www.cnki.com.cn/Article/CJFDTOTAL-HZLG200702030.htm
[18]	HUANG C, LIU Q, LI X. Color image segmentation by seeded region growing and region merging[C]//2010 Seventh International Conference on Fuzzy Systems and Knowledge Discovery, 2010: 533-536.
[19]	YANG L, WU X, ZHAO D, et al. An improved Prewitt algorithm for edge detection based on noised image[C]//20114th International Congress on Image and Signal Processing, 2011: 1197-1200.