한국정보통신학회논문지 (Journal of the Korea Institute of Information and Communication Engineering)

  • 제25권12호
  • /
  • Pages.1790-1796
  • /
  • 2021
  • /
  • 2234-4772(pISSN)
  • /
  • 2288-4165(eISSN)
한국정보통신학회 (The Korea Institute of Information and Commucation Engineering)
권호 표지
DOI QR코드

DOI QR Code

동영상 내 객체 추적을 위한 영상 데이터셋 구축 방법

Building Method an Image Dataset for Tracking Objects in a Video

  • Kim, Ji-Seong (Department of Computer Engineering, Dong-Eui University) ;
  • Heo, Gyeongyong (Department of Electronic Engineering, Dong-eui University) ;
  • Jang, Si-Woong (Department of Computer Engineering, Dong-Eui University)
  • 투고 : 2021.10.21
  • 심사 : 2021.11.18
  • 발행 : 2021.12.31
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

영상 딥러닝을 위해서는 다량의 영상 데이터셋이 필요한데, 객체의 종류에 따라 영상을 구하고 영상 데이터셋을 구축하는 방법에 많은 차이가 있다. 본 논문에서는 딥러닝을 위한 영상 데이터셋을 구축하는 방법을 제시하고 추적하는 객체에 따라 달라지는 성능을 분석하였다. 제안하는 데이터셋 구축방법을 활용하여 객체를 회전시킨 후 동영상을 촬영, 분할하여 커스텀 데이터셋을 구축하고, 성능을 분석한 결과 95% 이상의 객체 검출률을 보였으며, 이동 시 객체의 형상 변화가 적은 경우에 더 높은 성능이 나타났다. 영상 데이터를 구하기 어렵고, 형태의 변화가 적은 객체를 동영상 내에서 추적하기 위한 상황을 위하여는 본 논문에서 제시한 데이터셋 구축방법을 활용하는 것이 효과적일 것으로 판단된다.

A large amount of image data sets are required for image deep learning, and there are many differences in the method of obtaining images and constructing image data sets depending on the type of object. In this paper, we presented a method of constructing an image data set for deep learning and analyzed the performance that varies depending on the object to be tracked. We took a video by rotating the object, and then created a data set by segmenting the video using the proposed data set construction method. As a result of performance analysis, detection rate was more than 95%, and detection rate of objects with little change in shape was higher performance. It is considered that it is effective to use the data set construction method presented in this paper for a situation in which it is difficult to obtain image data and to track an object with little change in shape within a video.

키워드

과제정보

This research was supported by the MSIT (Ministry of Science and ICT),Korea, under the Grand Information Technology Research Center support program(IITP-2021-2020-0-01791) supervised by the IITP(Institute for Information & communications Technology Planning & Evaluation)

참고문헌

  1. N. H. Thang and J. Kim, "Performance Evaluation of YOLO according to Based Feature Extraction Network in Vespa Detection," Journal of Korea Multimedia Society, vol. 24, no. 2, pp. 305-311, Feb. 2021. https://doi.org/10.9717/KMMS.2020.24.2.305
  2. C. H. Lee and Y. J. Jeong, "Estimation of Moving Direction of Objects for Vehicle Tracking in Underground Parking Lot," Journal of Apiculture, vol. 35, no. 3, pp. 149-159, Sept. 2020. https://doi.org/10.17519/apiculture.2020.09.35.3.149
  3. J. S. Kim, C. H. Lee, and S. C. Kwon, "Real-time Online Study Attitude Action Recognition And Dataset Construction," The Conference of The Institute of Electronics and Information Engineers, Jeju, pp. 1133-1135, 2021.
  4. G. Y. Yi, Y. J. Kim, S. T. Kim, H. E. Kim, and K. G. Kim, "AR based Beverage Information Visualization and Sharing System using Deep Learning," Journal of Korea Multimedia Society, vol. 22, no. 3, pp. 349-356, Mar. 2019. https://doi.org/10.9717/KMMS.2019.22.3.349
  5. J. G. Choe, J. H. Lee, D. G. Kang, and S. H. Seo, "Comparison and Verification of Deep Learning Models for Automatic Recognition of Pills," Journal of Digital Contents Society, vol. 21, no. 3, pp. 445-452, Mar. 2020. https://doi.org/10.9728/dcs.2020.21.3.445
  6. B. J. Kim, I. K. Kim, H. S. Lim, and J. H. Gwak, "Dataset Augmentation Technique for Crack Detection of Wood Building," Proceedings of the Korean Society of Computer Information Conference, Jeju, pp. 645-647, 2021.
  7. D. H. Lee and J. H. Nang, "A Dataset and Keypoint Detection for Club Head Trajectory Estimation in 2D Golf Swing Video," in Proceedings of The Korean Institute of Information Scientists and Engineers Conference, Jeju, pp. 645-647, 2021.
  8. D. W. Park, A. R. Goo, and H. J. Park, "Implementation of Region Specific Image Collection And Dataset Building Application Using Roadview API," in Proceedings of The Korean Institute of Information Scientists and Engineers Conference, online, pp. 1424-1426, 2020.
  9. K. J. Kim, I. S. Jang, and K. T. Lim, "Construction of Wild-fire Smoke Data-set and Comparative Analysis of Detection Method based on Deep Neural Network," in Proceedings of Symposium of the Korean Institute of communications and Information Sciences, YongPyong, pp. 1172-1173, 2021.
  10. S. Y. Lee, Y. J. Kim, Y. B. Shim, H. S. Son, and K. W. Min, "Development of SW technology for recognition, judgement and path control algorithm verification simulation and dataset generation," in Proceedings of Symposium of the Korean Institute of communications and Information Sciences, YongPyong, pp. 353-354, 2020.
  11. T. J. Lee, Y. J. Kim, and H. K. Jung, "Dataset Construction and Model Learning for Manufacturing Worker Safety Management," Journal of the Korea Institute of Information and Communication Engineering, vol. 25, no. 7, pp. 890-895, Jul. 2021. https://doi.org/10.6109/JKIICE.2021.25.7.890
  12. J. W. Lee, D. J. Lee, S. J. Jang, D. G. Choi, and J. W. Jang, "Analysis of Deep Learning Model for the Development of an Optimized Vehicle Occupancy Detection System," Journal of the Korea Institute of Information and Communication Engineering, vol. 25, no. 1, pp. 146-151, Jan. 2021. https://doi.org/10.6109/JKIICE.2021.25.1.146
  13. G. W. Lee, H. L. Lee, and H. W. Cheong, "Object Detection of AGV in Manufacturing Plants using Deep Learning," Journal of the Korea Institute of Information and Communication Engineering, vol. 25, no. 1, pp. 36-43, Jan. 2021. https://doi.org/10.6109/JKIICE.2021.25.1.36
  14. H. K. Lim, "Overview of Image-based Object Recognition AI technology for Autonomous Vehicles," Journal of the Korea Institute of Information and Communication Engineering, vol. 25, no. 8, pp. 1117-1123, Aug. 2021. https://doi.org/10.6109/JKIICE.2021.25.8.1117