International Journal of Computer Science & Network Security

International Journal of Computer Science & Network Security (국제컴퓨터통신보호논문지학회)
권호 표지
DOI QR코드

DOI QR Code

Proficient: Achieving Progressive Object Detection over a Lossless Network using Fragmented DCT Coefficients

  • Received : 2024.04.05
  • Published : 2024.04.30
Download PDF
⟨ Previous Next ⟩

Abstract

In this work, we focused on reducing the amount of image data to be sent by extracting and progressively sending prominent image features to high-performance computing systems taking into consideration the right amount of image data required by object identification application. We demonstrate that with our technique called Progressive Object Detection over a Lossless Network using Fragmented DCT Coefficients (Proficient), object identification applications can detect objects with at least 70% combined confidence level by using less than half of the image data.

Keywords

References

  1. Microsoft COCO Dataset. Online: https://cocodataset.org/ Date Last Visited: March 23, 2024
  2. T. Lin and Pengwei Hao, "Compound image compression for real-time computer screen image transmission," in IEEE Transactions on Image Processing, vol. 14, no. 8, pp. 993-1005, Aug. 2005, doi: 10.1109/TIP.2005.849776.
  3. Zhou, X., Xu, L., Liu, S., Lin, Y., Zhang, L., & Zhuo, C. (2019). An Efficient Compressive Convolutional Network for Unified Object Detection and Image Compression. Proceedings of the AAAI Conference on Artificial Intelligence, 33(01), 5949-5956. https://doi.org/10.1609/aaai.v33i01.33015949
  4. Choi, Hyomin and Ivan V. Bajic. "Deep Feature Compression for Collaborative Object Detection." 2018 25th IEEE International Conference on Image Processing (ICIP) (2018): 3743-3747.
  5. S. Dodge and L. Karam, "Understanding how image quality affects deep neural networks," 2016 Eighth International Conference on Quality of Multimedia Experience (QoMEX), 2016, pp. 1-6, doi: 10.1109/QoMEX.2016.7498955.
  6. Ian J. Goodfellow, Jonathon Shlens, Christian Szegedy: Explaining and Harnessing Adversarial Examples. ICLR (Poster) 2015
  7. Gandor, T.; Nalepa, J. First Gradually, Then Suddenly: Understanding the Impact of Image Compression on Object Detection Using Deep Learning. Sensors 2022, 22, 1104. https://doi.org/10.3390/s22031104
  8. Choi, Hyomin and Ivan V. Bajic. "High Efficiency Compression for Object Detection." 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) (2018): 1792-1796.
  9. J. Redmon and A. Farhadi, "YOLO9000: better, faster, stronger," in IEEE Conf. Computer Vision and Pattern Recognition, Jul. 2017.