KIPS Transactions on Computer and Communication Systems (정보처리학회논문지:컴퓨터 및 통신 시스템)

Korea Information Processing Society (한국정보처리학회)
권호 표지
DOI QR코드

DOI QR Code

Semantic Segmentation Intended Satellite Image Enhancement Method Using Deep Auto Encoders

심층 자동 인코더를 이용한 시맨틱 세그멘테이션용 위성 이미지 향상 방법

  • ;
  • 이효종 (전북대학교 컴퓨터공학부)
  • Received : 2022.12.20
  • Accepted : 2023.06.20
  • Published : 2023.08.31
Download PDF
⟨ Previous Next ⟩

Abstract

Satellite imageries are at a greatest importance for land cover examining. Numerous studies have been conducted with satellite images and uses semantic segmentation techniques to extract information which has higher altitude viewpoint. The device which is taking these images must employee wireless communication links to send them to receiving ground stations. Wireless communications from a satellite are inevitably affected due to transmission errors. Evidently images which are being transmitted are distorted because of the information loss. Current semantic segmentation techniques are not made for segmenting distorted images. Traditional image enhancement methods have their own limitations when they are used for satellite images enhancement. This paper proposes an auto-encoder based image pre-enhancing method for satellite images. As a distorted satellite images dataset, images received from a real radio transmitter were used. Training process of the proposed auto-encoder was done by letting it learn to produce a proper approximation of the source image which was sent by the image transmitter. Unlike traditional image enhancing methods, the proposed method was able to provide more applicable image to a segmentation model. Results showed that by using the proposed pre-enhancing technique, segmentation results have been greatly improved. Enhancements made to the aerial images are contributed the correct assessment of land resources.

위성 이미지는 토지 표면 조사에서 매우 중요하다. 따라서 위성에서 지상국으로 이미지를 전송하기 위해 다양한 방법을 사용하고 있다. 그러나 전송 시스템의 품질 저하로 인해 이미지는 왜곡에 취약하고 올바른 데이터를 제공하지 못하고 있다. 그러한 이미지의 세그먼트 결과는 토지 표면 데이터를 올바르게 분류할 수 없다. 본 논문에서는 위성영상에 대한 자동인코더 기반의 영상 전처리 방법을 제안한다. 실험결과 사전 향상 기술을 사용하여 세그멘테이션 결과도 크게 향상될 수 있음을 보여주었다. 또한 본 논문에서 적용한 항공 이미지 향상기법은 토지 자원의 정확한 평가에 이바지할 수 있음을 확인하였다.

Keywords

Acknowledgement

This research was funded by the project for Joint Demand Technology R&D of Regional SMEs funded by the Korea Ministry of SMEs and Startups in 2023(Project No. RS-2023-00207672).

References

  1. A. Boguszewski, D. Batorski, N. Ziemba-Jankowska, T. Dziedzic, and A. Zambrzycka, "LandCover.ai: Dataset for automatic mapping of buildings, woodlands, water and roads from aerial imagery," Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops, June, 2021, Pages 1102, 1110., 2022. https://doi.org/10.48550/ arXiv.2005.02264
  2. F. Gerard et al., "Land cover change in Europe between 1950 and 2000 determined employing aerial photography," Progress in Physical Geography: Earth and Environment, Vol.34, No.2, pp.183-205, 2010. https://doi.org/10.1177/0309133309360141
  3. S. M. Azimi, E. Vig, R. Bahmanyar, M. Korner, and P. Reinartz, "Towards multi-class object detection in unconstrained remote sensing imagery," In Asian Conference on Computer Vision, Springer, pp.150-165, 2018.
  4. W. Zhou, G. Huang, and M. L. Cadenasso, "Does spatial configuration matter? Understanding the effects of land cover pattern on land surface temperature in urban landscapes," Landscape and Urban Planning, Vol.102, No.1, pp.54-63, 2011. https://doi.org/10.1016/j.landurbplan.2011.03.009
  5. A. Gupta, S. Watson, and H. Yin, "Deep Learning-based Aerial Image Segmentation with Open Data for Disaster Impact Assessment," Neurocomputing, Vol.439, pp.22-33, 2021. https://doi.org/10.1016/j.neucom.2020.02.139
  6. J. Kim, S. Song, and S.-C. Yu, "Denoising Auto-encoder based image enhancement for high resolution sonar image," 2017 IEEE Underwater Technology (UT).
  7. Q. Xiang and X. Pang, "Improved Denoising Auto-encoders for image denoising," 2018 11th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI 2018).
  8. V. Badrinarayanan, A. Kendall, and R. Cipolla, "Segnet: A deep convolutional encoder-decoder architecture for image segmentation," arXiv preprint arXiv:1511.00561, 2015.
  9. O. Ronneberger, P. Fischer, and T. Brox, "U-Net: Convolutional networks for biomedical image segmentation," Medical Image Computing and Computer-Assisted Intervention - MICCAI, pp.234-241, 2015.
  10. G. Sethi, B. S. Saini, and D. Singh, "Segmentation of cancerous regions in liver using an edge-based and phase congruent region enhancement method," Computers & Electrical Engineering, Vol.53, pp.244-262, 2016. https://doi.org/10.1016/j.compeleceng.2015.06.025.
  11. K. Wu and D. Zhang, "Robust tongue segmentation by fusing region-based and edge-based approaches," Expert Systems with Applications, Vol.42, No.21, pp.8027-8038, 2015. https://doi.org/10.1016/j.eswa.2015.06.032.
  12. P. Vijay and N. C. Patil, "Gray scale image segmentation using OTSU thresholding optimal approach," Journal for Research, Vol.2, No.5, pp.20-24, 2016.
  13. S. Aja-Fernandez, A. H. Curiale, and G. Vegas-SanchezFerrero, "A local fuzzy thresholding methodology for multi-region image segmentation," Knowledge-Based Systems, Vol.83, pp.1-12, 2015. https://doi.org/10.1016/j.knosys.2015.02.029.
  14. N. M. Zaitoun and M. J. Aqel. "Survey on image segmentation techniques," Procedia Computer Science, Vol.65, pp.797-806, 2015. https://doi.org/10.1016/j.procs.2015.09.027.
  15. S. Niu, Q. Chen, L. de Sisternes, Z. Ji, Z. Zhou, and D. L. Rubin, "Robust noise region-based active contour model via local similarity factor for image segmentation," Pattern Recognition, Vol.61, pp.104-119, 2017. https://doi.org/10.1016/j.patcog.2016.07.022.
  16. E. Anjna and E. R. Kaur. "Review of image segmentation technique," International Journal, Vol.8, No.4, pp.36-39, 2017.
  17. K. Simonyan and A. Zisserman, "Very deep convolutional networks for large-scale image recognition," arXiv preprint arXiv:1409.1556, 2014.
  18. J. Long, E. Shelhamer, and T. Darrell, "Fully convolutional networks for semantic segmentation," in CVPR, pp.3431-3440, 2015.
  19. A. Krizhevsky, I. Sutskever, and G. E. Hinton, "Imagenet classification with deep convolutional neural networks," in Ad-vances in Neural Information Processing Systems, 25, pp.1097-1105, 2012.
  20. C. Szegedy, W. Liu, Y. Jia, P. Sermanet, S. Reed, D. Anguelov, D. Erhan, V. Vanhoucke, and A. Rabinovich, "Going deeper with convolutions," in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp.1-9, 2015.
  21. H. Noh, S. Hong, and B. Han, "Learning deconvolution network for semantic segmentation," in Proceedings of the IEEE International Conference on Computer Vision, pp.1520-1528, 2015.
  22. A. Chaurasia and E. Culurciello, "LinkNet: Exploiting encoder representations for efficient semantic segmentation," IEEE Visual Communications and Image Processing (VCIP), 2017.
  23. W. Liu, A. Rabinovich, and A. C. Berg, "ParseNet: Looking wider to see better, computer vision and pattern recognition," 2016. https://doi.org/10.48550/arXiv.1506.04579
  24. T.-Y. Lin, P. Dollar, R. Girshick, K. He, B. Hariharan, and S. Belongie, "Feature pyramid networks for object detection," Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp.2117-2125, 2017.
  25. M. L. L. de Oliveira and M. J. G. Bekooij, "Deep convolutional autoencoder applied for noise reduction in range-doppler maps of FMCW radars," 2020 IEEE International Radar Conference (RADAR).
  26. S. Park, S. Yu, M. Kim, K. Park, and J. Paik, "Dual autoencoder network for retinex-based low-light image enhancement," IEEE Access (Volume: 6), 2018.
  27. P. Schuch, S. Schulz, and C. Busch, "De-convolutional auto-encoder for enhancement of fingerprint samples," Sixth International Conference on Image Processing Theory, Tools and Applications (IPTA), 2016.
  28. Satellite images of Dubai Dataset [Internet], https://www.kaggle.com/datasets/humansintheloop/semantic-segmentation-of-aerial-imagery
  29. Abien Fred Agarap, Deep Learning using Rectified Linear Units (ReLU), 2019. https://doi.org/10.48550/arXiv.1803.08375
  30. K. He, X. Zhang, S. Ren, and J. Sun, "Deep residual learning for image recognition," Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp.770-778, 2016.
  31. D. P. Kingma and J. Ba, "Adam: A method for stochastic optimization," 3rd International Conference for Learning Representations, San Diego, 2015.