Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography (한국측량학회지)

Korean Society of Surveying, Geodesy, Photogrammetry and Cartography (한국측량학회)
권호 표지
DOI QR코드

DOI QR Code

Road Extraction from Images Using Semantic Segmentation Algorithm

영상 기반 Semantic Segmentation 알고리즘을 이용한 도로 추출

  • Received : 2022.06.10
  • Accepted : 2022.06.21
  • Published : 2022.06.30
Download PDF
⟨ Previous Next ⟩

Abstract

Cities are becoming more complex due to rapid industrialization and population growth in modern times. In particular, urban areas are rapidly changing due to housing site development, reconstruction, and demolition. Thus accurate road information is necessary for various purposes, such as High Definition Map for autonomous car driving. In the case of the Republic of Korea, accurate spatial information can be generated by making a map through the existing map production process. However, targeting a large area is limited due to time and money. Road, one of the map elements, is a hub and essential means of transportation that provides many different resources for human civilization. Therefore, it is essential to update road information accurately and quickly. This study uses Semantic Segmentation algorithms Such as LinkNet, D-LinkNet, and NL-LinkNet to extract roads from drone images and then apply hyperparameter optimization to models with the highest performance. As a result, the LinkNet model using pre-trained ResNet-34 as the encoder achieved 85.125 mIoU. Subsequent studies should focus on comparing the results of this study with those of studies using state-of-the-art object detection algorithms or semi-supervised learning-based Semantic Segmentation techniques. The results of this study can be applied to improve the speed of the existing map update process.

현대에는 급속한 산업화와 인구 증가로 인해 도시들이 더욱 복잡해지고 있다. 특히 도심은 택지개발, 재건축, 철거 등으로 인해 빠르게 변화하는 지역에 해당한다. 따라서 자율주행에 필요한 정밀도로지도와 같은 다양한 목적을 위해 빠른 정보 갱신이 필요하다. 우리나라의 경우 기존 지도 제작 과정을 통해 지도를 제작하면 정확한 공간정보를 생성할 수 있으나 대상 지역이 넓은 경우 시간과 비용이 많이 든다는 한계가 있다. 지도 요소 중 하나인 도로는 인류 문명을 위한 많은 다양한 자원을 제공하는 중추이자 필수적인 수단에 해당한다. 따라서 도로 정보를 정확하고 신속하게 갱신하는 것이 중요하다. 이 목표를 달성하기 위해 본 연구는 Semantic Segmentation 알고리즘인 LinkNet, D-LinkNet 및 NL-LinkNet을 사용하여 광주광역시 도시철도 2호선 공사 현장을 촬영한 드론 정사영상에서 도로를 추출한 다음 성능이 가장 높은 모델에 하이퍼 파라미터 최적화를 적용하였다. 그 결과, 사전 훈련된 ResNet-34를 Encoder로 사용한 LinkNet 모델이 85.125 mIoU를 달성했다. 향후 연구 방향으로 최신 Semantic Segmentation 알고리즘 또는 준지도 학습 기반 Semantic Segmentation 기법을 사용하는 연구의 결과와의 비교 분석이 수행될 것이다. 본 연구의 결과는 기존 지도 갱신 프로세스의 속도를 개선하는 데 도움을 줄 수 있을 것으로 예상된다.

Keywords

Acknowledgement

이 논문은 2021년도 정부(과학기술정보통신부)의 재원으로 한국연구재단의 지원을 받아 수행된 연구임(No. 2021R1C1C1012785). 또한 본 논문은 행정안전부 "극한재난대응기반기술개발사업(20017423)"의 지원을 받아 작성되었음.

References

  1. Abdollahi, A., Pradhan, B., Shukla, N., Chakraborty, S., and Alamri, A., (2020), Deep learning approaches applied to remote sensing datasets for road extraction: A state-of-the-art review, Remote Sensing, Vol. 12, No. 9, pp. 1444-1465. https://doi.org/10.3390/rs12091444
  2. Alshehhi, R., Marpu, P. R., Woon, W. L., And Dalla Mura, M. (2017), Simultaneous extraction of roads and buildings in remote sensing imagery with convolutional neural networks, ISPRS Journal of Photogrammetry and Remote Sensing, Vol. 130, No. 2017, pp. 139-149. https://doi.org/10.1016/j.isprsjprs.2017.05.002
  3. Awad, M. M., (2013), A morphological model for extracting road networks from high-resolution satellite images, Journal of Engineering, Vol. 2013, pp. 243021 1-9. https://doi.org/10.1155/2013/243021
  4. Chaurasia, A. and Culurciello, E., (2017), Linknet: Exploiting encoder representations for efficient semantic segmentation, IEEE Visual Communications and Image Processing-2017, 10-13 December, St. Petersburg, FL, USA, pp. 1-4. https://doi.org/10.1109/VCIP.2017.8305148
  5. Demir, I., Koperski, K., Lindenbaum, D., Pang, G., Huang, J., Basu, S., Hughes, F., Tuia, D., and Raskar, R., (2018), Deepglobe 2018: A challenge to parse the earth through satellite images, Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops-2018, 18-22 June, Salt Lake City, UT, USA, pp. 172-181. https://doi.org/10.1109/CVPRW.2018.00031
  6. He, K., Zhang, X., Ren, S., and Sun, J., (2016), Deep residual learning for image recognition, Proceedings of the IEEE conference on computer vision and pattern recognition-2016, 27-30 June, Las Vegas, NV, USA, pp. 770-778. https://doi.org/10.48550/arXiv.1512.03385
  7. Jang, Y.J., Oh, J.H., and Lee, C.N., (2020), Urban Building Change Detection Using nDSM and Road Extraction, Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography, Vol. 38, No. 3, pp. 237-246. https://doi.org/10.7848/ksgpc.2020.38.3.237
  8. Kim, J.Y., Huh, Y, Yu, K.Y., and Kim, J.O., (2013), Automatic Change Detection Based on Areal Feature Matching in Different Network Data-sets, Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography, Vol. 31, No. 6,pp. 483-491. https://doi.org/10.7848/ksgpc.2013.31.6-1.483
  9. Liu, W. and Wang, H., (2008), An interactive image segmentation method based on graph theory, J. Electron. Inf. Technol, Vol. 8, No. 30, pp. 1973-1976.
  10. Long, J., Shelhamer, E., and Darrell, T., (2015), Fully convolutional networks for semantic segmentation, Proceedings of the IEEE conference on computer vision and pattern recognition-2015, Boston, 7-12 June, MA, USA, pp. 3431-3440. https://doi.org/10.48550/arXiv.1411.4038
  11. Ma, R., Wang, W., and Liu, S., (2012), Extracting roads based on Retinex and improved Canny operator with shape criteria in vague and unevenly illuminated aerial images, Journal of applied remote sensing, Vol. 6, No. 1, pp. 063610 1-14. (in Korean with English abstract) https://doi.org/10.1117/1.JRS.6.063610
  12. Mena, J. B. and Malpica, J. A., (2005), An automatic method for road extraction in rural and semi-urban areas starting from high resolution satellite imagery, Pattern recognition letters, Vol. 26, No. 9, pp. 1201-1220. https://doi.org/10.1016/j.patrec.2004.11.005
  13. Miao, Z., Wang, B., Shi, W., and Zhang, H., (2014), A semiautomatic method for road centerline extraction from VHR images, IEEE Geoscience and Remote Sensing Letters, Vol. 11, No. 11, pp. 1856-1860. https://doi.org/10.1109/LGRS.2014.2312000
  14. Mnih, V. and Hinton, G. E., (2010), Learning to detect roads in high-resolution aerial images, European conference on computer vision-2010, 5-11 September, Heraklion, Crete, Greece, pp. 210-223. https://doi.org/10.1007/978-3-642-15567-3_16
  15. Panboonyuen, T., Vateekul, P., Jitkajornwanich, K. and Lawawirojwong, S., (2017), An enhanced deep convolutional encoder-decoder network for road segmentation on aerial imagery, International conference on computing and information technology-2017, 27-29 December, Singapore, Singapore, pp. 191-201. https://doi.org/10.1007/978-3-319-60663-7_18
  16. Park, Y.K., Kang, W.P., Choi, J.E., and Kim, B.J., (2019), A study on the Evaluation of Real-Time Map Update Technology for Automated Driving, Journal of the Korean Association of Geographic Information Studies, Vol. 22, No. 3, pp. 146-154. (in Korean with English abstract) https://doi.org/10.11108/kagis.2019.22.3.146
  17. Shi, W., Miao, Z., and Debayle, J., (2013), An integrated method for urban main-road centerline extraction from optical remotely sensed imagery, IEEE Transactions on Geoscience and Remote Sensing, Vol. 52, No. 6, pp. 3359-3372. https://doi.org/10.1109/TGRS.2013.2272593
  18. Tao, W. B. and Jin, H., (2007), A novel method of image threshold segmentation based on graph theory, Chinese Journal of Computer, Vol. 1, pp. 110-119.
  19. Unsalan, C. and Sirmacek, B., (2012), Road network detection using probabilistic and graph theoretical methods, IEEE Transactions on Geoscience and Remote Sensing, Vol. 50, No. 11, pp. 4441-4453. https://doi.org/10.1109/TGRS.2012.2190078
  20. Varia, N., Dokania, A. & Senthilnath, J., (2018), DeepExt: A convolution neural network for road extraction using RGB images captured by UAV, IEEE Symposium Series on Computational Intelligence-2018, 18-21 Novemver, Bengaluru, India,pp. 1890-1895. https://doi.org/10.1109/SSCI.2018.8628717
  21. Vosselman, G. and Knecht, J. D., (1995), Road tracing by profile matching and Kaiman filtering, Automatic extraction of man-made objects from aerial and space images, Springer, pp. 265-274. https://doi.org/10.1007/978-3-0348-9242-1_25
  22. Wang, J., Qin, Q., Yang, X., Wang, J., Ye, X., and Qin, X., (2014), Automated road extraction from multi-resolution images using spectral information and texture, IEEE Geoscience and Remote Sensing Symposium-2014, 13-18 July, Quebec City, QC, Canada, pp. 533-536. https://doi.org/10.1109/IGARSS.2014.6946477
  23. Wang, J., Song, J., Chen, M., and Yang, Z., (2015), Road network extraction: A neural-dynamic framework based on deep learning and a finite state machine, International Journal of Remote Sensing, 36, 3144-3169. https://doi.org/10.1080/01431161.2015.1054049
  24. Wang, X., Girshick, R., Gupta, A., and He, K., (2018), Nonlocal neural networks, Proceedings of the IEEE conference on computer vision and pattern recognition, 7794-7803. https://doi.org/10.1109/CVPR.2018.00813
  25. Wang, Y., Seo, J., and Jeon, T., (2021), NL-LinkNet: Toward lighter but more accurate road extraction with nonlocal operations, IEEE Geoscience and Remote Sensing Letters, Vol. 19, pp. 1-5. https://doi.org/10.1109/LGRS.2021.3050477
  26. Wei, Y., Wang, Z., and Xu, M., (2017), Road structure refined CNN for road extraction in aerial image, IEEE Geoscience and Remote Sensing Letters, Vol. 14, No. 5, pp. 709-713. https://doi.org/10.1109/LGRS.2017.2672734
  27. Yun, B.Y., Moon, D.Y., and Hong, S.H., (2006), A Study on Updating of Digital Map using Beacon GPS, Journal of the Korean Geophysical Society, Vol. 9, No. 4, pp. 387-395. (in Korean with English abstract)
  28. Zhang, C., Murai, S., and Baltsavias, E. P., (1999), Road network detection by mathematical morphology, ISPRS Workshop 3D Geospatial Data Production: Meeting Application Requirements, 7-9 April, Paris, France, pp. 185-200. https://doi.org/10.3929/ethz-a-004334280
  29. Zhou, L., Zhang, C., and Wu, M., (2018), D-LinkNet: LinkNet with pretrained encoder and dilated convolution for high resolution satellite imagery road extraction, Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops-2018, 18-22 June, Salt Lake City, UT, USA, pp. 182-186. https://doi.org/10.1109/CVPRW.2018.00034
  30. Zhu, C., Shi, W., Pesaresi, M., Liu, L., Chen, X., and King, B., (2005), The recognition of road network from high-resolution satellite remotely sensed data using image morphological characteristics, International Journal of Remote Sensing, Vol. 26, No. 24, pp. 5493-5508. https://doi.org/10.1080/01431160500300354