Aerial image de-roping based on U-net model

HONG Hanyu;SUN Jianguo;LUAN Ling;WANG Shuo;ZHENG Xinbo

CN Submission Journal Articles Figures Search
Adv. Search

  • Sponsored by:

    Editor-In-Chief:

    ISSN 1002-2082

    　

  • Hosted By:

    Published By: Journal of Applied Optics

    CN 61-1171/O4

Journal of Applied Optics ›› 2019, Vol. 40 ›› Issue (5) : 786-794. DOI: 10.5768/JAO201940.0502001

Aerial image de-roping based on U-net model

  • HONG Hanyu1, SUN Jianguo1, LUAN Ling2, WANG Shuo1, ZHENG Xinbo2
Author information +
History +

Abstract

The Kuang-chi cloud platform is used to pull the airbag with a cable rope. Since the aerial photography system is suspended under the airbag, the captured image inevitably contains the rope information, which affects the image quality and needs to eliminated in the scene analysis and object detection. A rope detection algorithm based on U-net model was proposed, which introduced the separable convolution to improve the detection speed and adopted a weighted cross entropy as a loss function to solve the problem of convergence instability caused by category imbalance. The final model was able to quickly and accurately detect the rope with fewer data in less time, and then the rope image waseliminated using the fast marching method algorithm (FMM). The test results show that the mean intersection over union(mIOU) of the algorithm reaches 62.8% and has obtained good results.

Key words

aerial photography / semantic segmentation / rope detection / U-net

Cite this article

EndNote

Ris (Procite)

Bibtex

Download Citations
HONG Hanyu, SUN Jianguo, LUAN Ling, WANG Shuo, ZHENG Xinbo. Aerial image de-roping based on U-net model. Journal of Applied Optics. 2019, 40(5): 786-794 https://doi.org/10.5768/JAO201940.0502001

References

乔寅骐, 肖健华, 黄银, 等.基于最小二乘修正的随机Hough变换直线检测[J].计算机应用, 2015, 35(11): 3312-3315.
洪汉玉.现代图像图形处理与分析[M].武汉:中国地质大学出版社, 2011.
徐威, 唐振民, 吕建勇.基于图像显著性的路面裂缝检测[J].中国图像图形学报, 2013, 18(1): 69-77.
高凯珺, 孙韶媛, 姚广顺, 等.基于深度学习的无人车夜视图像语义分割[J].应用光学, 2017, 38(3):421-427.
刘哲, 张晓林, 宋余庆, 等.结合改进的U-Net和Morphsnakes的肝脏分割[J].中国图象图形学报, 2018, 23(8):1254-1262.
张红英, 彭启琮.数字图像修复技术综述[J].中国图象图形学报, 2007, 12(1):1-10.
BURNS J B, HANSON A R, RISEMAN E M.Extracting straight lines[J].IEEE Trans. Pattern Analysis and Machine Intelligence, 1986, 8(4): 425-455.
XU L, OJA E, KULTANEN P. A new curve detection method: randomized Hough transform (RHT)[J]. Pattern Recognition Letters, 1990, 11(5): 331-338.
QIAO Yinqi, XIAO Jianhua, HUANG Yin, et al. Random Hough transform line detection based on least squares correction[J]. Journal of Computer Applications, 2015, 35(11): 3312-3315.
HONG Hanyu. Modern image graphics processing and analysis[M]. Wuhan: China University of Geosciences Press, 2011.
XU Wei, TANG Zhenmin, LYU Jianyong. Detection of pavement crack based on image saliency[J]. Journal of Image and Graphics, 2013, 18(1): 69-77.
ALEXANDRU T.An image inpainting technique based on the fast marching method[J]. Journal of Graphics Tools, 2004, 9(1):23-34.
GAO Kaiwei, SUN Yiyuan, YAO Guangshun, et al. Semantic segmentation of night vision images of unmanned vehicles based on deep learning[J]. Journal of Applie

9

Accesses

0

Citation

Detail
Sections
Recommended

/

〈 〉