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

  • 제39권6호
  • /
  • Pages.393-399
  • /
  • 2021
  • /
  • 1598-4850(pISSN)
  • /
  • 2288-260X(eISSN)
한국측량학회 (Korean Society of Surveying, Geodesy, Photogrammetry and Cartography)
권호 표지
DOI QR코드

DOI QR Code

도로 주변 지역의 CCTV영상을 이용한 야간시간대 미세먼지 농도 추정

Estimation of PM concentrations at night time using CCTV images in the area around the road

  • Won, Taeyeon (Dept. of Advanced Technology Fusion, Konkuk University) ;
  • Eo, Yang Dam (Dept. of Civil and Environmental Engineering, Konkuk University) ;
  • Jo, Su Min (Dept. of Advanced Technology Fusion, Konkuk University) ;
  • Song, Junyoung (Dept. of Advanced Technology Fusion, Konkuk University) ;
  • Youn, Junhee (Dept. of Future and Smart Construction Research, Korea Institute of Civil Engineering and Building Technology)
  • 투고 : 2021.11.03
  • 심사 : 2021.12.05
  • 발행 : 2021.12.31
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

본 연구는 다양한 미세먼지 농도 환경의 야간 CCTV 영상을 학습 시켜 영상에 의한 미세먼지 농도 추정에 대한 실험을 진행하였다. 주간 영상의 경우 많은 관련 연구가 존재하였고, 영상의 다양한 텍스쳐와 밝기 정보가 잘 표현되어 학습에 영향을 주는 정보가 뚜렷하다. 하지만 야간 영상의 경우 주간 영상에 비해 담고 있는 정보가 적고 야간 영상만 사용한 연구가 드물다. 따라서 차량과 가로등 같은 광원 때문에 특징이 균일하지 못한 야간 촬영 영상들과 비교적 광원이 일정한 건물 지붕, 건물 벽, 가로등을 ROI로 결합한 실험을 진행하였다. 이후 야간 영상으로 딥러닝 기반 미세먼지 농도 추정이 가능한지 상관도를 주간 실험과 비교 분석하였다. 실험 결과, 지붕 ROI (Region of Interest) 학습한 경우의 결과 값이 가장 높았고, 전체 영상과의 결합 학습모델은 더 향상된 결과를 보여주었다. 전체적으로 R2가 0.9를 상회하여 야간 CCTV 영상으로부터 미세먼지 추정이 가능함을 나타내고 있고, 날씨 자료의 추가 결합 학습은 실험 결과에 크게 영향을 주지 않은 것으로 계산되었다.

In this study, experiments were conducted to estimate the PM concentrations by learning the nighttime CCTV images of various PM concentrations environments. In the case of daytime images, there have been many related studies, and the various texture and brightness information of images is well expressed, so the information affecting learning is clear. However, nighttime images contain less information than daytime images, and studies using only nighttime images are rare. Therefore, we conducted an experiment combining nighttime images with non-uniform characteristics due to light sources such as vehicles and streetlights and building roofs, building walls, and streetlights with relatively constant light sources as an ROI (Region of Interest). After that, the correlation was analyzed compared to the daytime experiment to see if deep learning-based PM concentrations estimation was possible with nighttime images. As a result of the experiment, the result of roof ROI learning was the highest, and the combined learning model with the entire image showed more improved results. Overall, R2 exceeded 0.9, indicating that PM estimation is possible from nighttime CCTV images, and it was calculated that additional combined learning of weather data did not significantly affect the experimental results.

키워드

과제정보

이 성과는 정부(과학기술정보통신부)의 재원으로 한국연구재단의 지원을 받아 수행된 연구임(No. 2019R1A2C1085618).

참고문헌

  1. Bo, Q., Yang, W., Rijal, N., Xie, Y., Feng, U., and Zhang, J. (2018), Particle pollution estimation from images using convolutional neural network and weather Features, IEEE International Conference on Image Processing (ICIP), 7-10 October, Athens, Greece, pp. 3433-3437.
  2. Chakma, A., Vizena, B., Cao, T., Lin, J., and Zhang, J. (2017), Image-based air quality analysis using deep convolutional neural network. IEEE International Conference on Image Processing (ICIP), 17-20 September, Beijing, China, pp. 3949-3952.
  3. Cho, C.S. and Park, J.I. (2013), Improved glare region modeling using super pixel, Korean Society of Broadcast Engineers Summer Conference 2013, 26-28 June, JeJu, Korea, pp. 319-321.
  4. Choi, Y.H., Seong, Y.K., Jung, S.Y., and Choi, T.S. (2002), Image mosaicking with automatic scene segmentation for video indexing, Proceedings of Symposium of the Korean Institute of Communications and Information Sciences, pp. 335-338(4 pages).
  5. Choi, T.Y., Kang, D.I., and Cha,J.G. (2019), A study on the effects of green space and wind speed on roadside PM10 concentration, Proceedings of Korean Society of Environment and Ecology Conference. 29(2), 26 April, Wonju, Korea, pp. 67~68.
  6. Jin, J.K. and Jin, J.I. (2021), A study on the effect of traffic congestion on particulate matter concentration in Seoul : Big data approach, Journal of Korea Planning Association, Vol. 56, No. 1, pp. 121-136. (in Korean with English abstract) https://doi.org/10.17208/jkpa.2021.02.56.1.121
  7. Kim, B.K., Chang I.S., Park K.B., Cho J.S., and Lee M.J. (2009), A study of non-ROI real-time CCTV visibility measurements for highway fog warning system, Proceedings of the KAIS(The Korea Academia-Industrial cooperation Society) Spring Conference. unpaginated CD-ROM.(in Korean)
  8. Kwon, T.M., (2004), An Atmospheric Visibility Measurements Using Video Cameras: Relative Visibility, Univ. of Minesota Duluth.
  9. Lee, J.H., Kim, Y.M., and Kim, Y.k. (2017), Spatial panel analysis for PM2.5 concentration in Korea, Journal of The Korean Data and Information Science Society, Vol. 28(3), pp. 473-481(9 pages).(in Korean)
  10. Lee, Y.K., Lee, K.J., Lee, J.S., and Shin, E.S. (2012), Regional characteristics of particle size distribution of PM10, Journal of Korean Society for Atmospheric Environment(J. KOSAE), Vol. 28, No. 6, pp. 666~674.(in Korean with English abstract) https://doi.org/10.5572/KOSAE.2012.28.6.666
  11. Liu, C., Tsow, F., Zou, Y., and Tao, N. (2016), Particle Pollution Estimation Based on Image Analysis. PLoS ONE 11(2): e0145955. doi:10.1371/journal.pone.0145955.
  12. Park, C.S. (2017), Variations of PM10 concentration in Seoul during 2015 and relationships to weather condition, Journal of the Association of Korean Photo-Geographers, Vol. 27, No. 2, pp. 47-64.(in Korean with English abstract) https://doi.org/10.35149/jakpg.2017.27.2.004
  13. Park, S.S., Yu, G.H., Kim, Y.J., Rho, S.G., and Ryu, J.H. (2016), Field evaluation of real-time fine particle monitor using light scattering technique. J. of the Korean Society for Environmental Analysis, Vol. 19, No. 1, pp. 44-53.(in Korean with English abstract)
  14. Simonyan, K. and Zisserman, A. (2014), Very deep convolutional networks for large-scale image recognition, ICLR 2015, 7-9 May, San Diego, CA, pp. 1~14.
  15. Won, T., Eo, Y.D., Sung, H., Chong, K.S., and Youn, J. (2020), Effect of the learning image combinations and weather parameters in the PM estimation from CCTV images, Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography, Vol. 38, No. 6, pp. 573-581.(in Korean with English abstract) https://doi.org/10.7848/KSGPC.2020.38.6.573
  16. Won, T., Eo, Y.D., Sung, H., Chong, K.S., Youn, J., and Lee, G.W. (2021), Particulate Matter Estimation from Public Weather Data and Closed-Circuit Television Images, KSCE Journal of Civil Engineering (in press)
  17. Yun, Y., Kim, T., Oh, J., and Han, Y. (2021), Analysis of co-registration performance accordingto geometric processing level of KOMPSAT-3/3A reference image, Korean Journal of Remote Sensing, Vol. 37, No. 2, pp. 221~232.(in Korean with English abstract) https://doi.org/10.7780/KJRS.2021.37.2.4