KIPS Transactions on Software and Data Engineering (정보처리학회논문지:소프트웨어 및 데이터공학)

Korea Information Processing Society (한국정보처리학회)
권호 표지
DOI QR코드

DOI QR Code

A Dynamically Segmented DCT Technique for Grid Artifact Suppression in X-ray Images

X-ray 영상에서 그리드 아티팩트 개선을 위한 동적 분할 기반 DCT 기법

  • 김형규 (제이피아이헬스케어 사업개발본부 소프트웨어팀) ;
  • 정중은 (제이피아이헬스케어 사업개발본부 소프트웨어팀) ;
  • 이지현 (한동대학교 전산전자공학부) ;
  • 박준혁 (한동대학교 전산전자공학부) ;
  • 서지수 (한동대학교 전산전자공학부) ;
  • 김호준 (한동대학교 전산전자공학부)
  • Received : 2018.12.18
  • Accepted : 2019.02.22
  • Published : 2019.04.30
Download PDF
⟨ Previous Next ⟩

Abstract

The use of anti-scatter grids in radiographic imaging has the advantage of preventing the image distortion caused by scattered radiation. However, it carries the side effect of leaving artifacts in the X-ray image. In this paper, we propose a grid line suppression technique using discrete cosine transform(DCT). In X-ray images, the grid lines have different characteristics depending on the shape of the object and the area of the image. To solve this problem, we adopt the DCT transform based on a dynamic segmentation, and propose a filter transfer function for each individual segment. An algorithm for detecting the band of grid lines in frequency domain and a band stop filter(BSF) with a filter transfer function of a combination of Kaiser window and Butterworth filter have been proposed. To solve the blocking effects, we present a method to determine the pixel values using multiple structured images. The validity of the proposed theory has been evaluated from the experimental results using 140 X-ray images.

방사선 진단에서 산란선 보정 그리드의 사용은 굴절되는 신호에 의한 영상의 왜곡을 방지할 수 있는 장점이 있는 반면, X-ray 영상에서 그리드 아티팩트를 발생시키는 부작용을 수반한다. 본 논문에서는 이산코사인변환(DCT: discrete cosine transform)을 사용하여, 그리드 라인을 개선하는 기법을 제안한다. X-ray 영상에서 그리드 라인은 피사체의 형태와 영상의 영역에 따라 서로 다른 특성을 보인다. 이러한 점을 해결하기 위하여 동적 분할 구조를 기반으로 DCT 변환을 적용하고, 개별 분할별로 적합한 필터전달함수를 설계하였다. 세부적으로 주파수 영역 데이터에 대하여 그리드 라인의 대역을 검출하는 알고리즘을 제안하였으며, 필터전달함수로 Kaiser 윈도우와 Butterworth 필터를 조합한 형태의 밴드스톱필터(BSF: band stop filter)를 구현하였다. 또한 블로킹 현상을 개선하기 위하여 다중구조의 영상으로부터 픽셀값을 결정하는 방법론을 제시하였다. 총 140개의 실제 X-ray 영상을 사용한 실험결과로부터 제안된 이론의 타당성을 실험적으로 평가하였다.

Keywords

JBCRJM_2019_v8n4_171_f0001.png 이미지

Fig. 1. Examples of Grid Artifacts in X-ray Images:(A) Grid Lines, (B) Moire Effect

JBCRJM_2019_v8n4_171_f0002.png 이미지

Fig. 2. The Structure of the Grid Line Suppression Technique based on DCT

JBCRJM_2019_v8n4_171_f0003.png 이미지

Fig. 3. Examples of the Typical Form of the Grid Lines: (A) Input Image, (B) Frequency Analysis Graph, (C) Frequency Data Distribution

JBCRJM_2019_v8n4_171_f0004.png 이미지

Fig. 4. A Comparison of Image Characteristics Depending on Different Grid Device Models

JBCRJM_2019_v8n4_171_f0005.png 이미지

Fig. 5. An Example of Image Segmentation

JBCRJM_2019_v8n4_171_f0006.png 이미지

Fig. 6. An Example of the Band Stop Filtering

JBCRJM_2019_v8n4_171_f0007.png 이미지

Fig. 7. Examples of Erroneous Frequency Range Detection

JBCRJM_2019_v8n4_171_f0008.png 이미지

Fig. 8. Examples of the Blurring Operation Results

JBCRJM_2019_v8n4_171_f0009.png 이미지

Fig. 9. Remnants of Grid Lines Appearing at the Boundary of the Object

JBCRJM_2019_v8n4_171_f0010.png 이미지

Fig. 10. An Example of the Blocking Effect

JBCRJM_2019_v8n4_171_f0011.png 이미지

Fig. 11. A Comparison Between Images Before and After Applying the Proposed Dynamically Segmented DCT Technique: (A), (C) Input Images, (B), (D), Result Images

JBCRJM_2019_v8n4_171_f0012.png 이미지

Fig. 12. Grid Line Suppression Results

JBCRJM_2019_v8n4_171_f0013.png 이미지

Fig. 13. A Result of the Blocking Effect Reduction Method:(A) Input Image, (B) Result Image

JBCRJM_2019_v8n4_171_f0014.png 이미지

Fig. 14. Experimental Result on a Phantom Image:(A) Non-Grid Image, (B) Grid Image, (C) Grid Line Suppressed Image

References

  1. R. Rana, A. Jain, A. Shankar, D. R. Bednarek, and S. Rudin, "Scatter estimation and removal of anti-scatter grid-line artifacts from anthropomorphic head phantoms images taken with a high resolution image detector," Proceedings SPIE, Medical Imaging 2016.
  2. R. Rana, D. R. Bednarek, and S. Rudin, "Grid-line artifact minimization for high resolution detectors using iterative residual scatter corrections," Medical Physics., Vol.42, No.6, pp.3695, 2015.
  3. Igor Belykh, "Grid artifacts suppression in computed radiographic images," International Journal of Biomedical and Biological Engineering, Vol.8, No.8, pp.1402-1405, 2014.
  4. V. Singh, A, Jain, D. R. Bednarek, and S. Rudin, "Limitations of anti-scatter grids when used with high resolution image detectors," Proceeding of SPIE, Medical Imaging. 2014.
  5. D. M. Gauntt, G. T. Barnes, "Grid line artifact formation: A comprehensive theory," Medical physics., Vol.33, No.6, pp.1668-1677, 2006. https://doi.org/10.1118/1.2164069
  6. Chih-Yang Lin, Wen-Jeng Lee, Shyh-jye Chen, Ching-Hwa Tsai, Jei-Han Lee, Chia-Hung Chang, and Yu-Tai Ching, "A Study of Grid Artifacts Formation and Elimination in Computed Radiographic Images," Journal of Digital Imaging, Vol.19, No.4, pp.351-361, 2006. https://doi.org/10.1007/s10278-006-0630-8
  7. R. Rana, V. Singh, A. Jain, D.R. Bednarek and S. Rudin, "Anti-scatter grid Artifact elimination for high resolution x-ray imaging CMOS detetors," Proc. of SPIE, Medical Imaging 2015.
  8. Yulun Zhang, Yapeng Tian, Yu Kong, Bineng Zhong, and Yun Fu. "Residual Dense Network for Image Restoration." arXiv preprint arXiv:1812.10477, 2018.
  9. Y. Chen and T. Pock, "Trainable nonlinear reaction diffusion: A flexible framework for fast and effective image restoration," IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol.39, No.6. pp.1256-1272, 2017. https://doi.org/10.1109/TPAMI.2016.2596743
  10. K. Zhang, W. Zuo, Y. Chen, D. Meng, and L. Zhang, "Beyond a gaussian denoiser: Residual learning of deep CNN for image denoising," IEEE Transaction of Image Processing, Vol.26, No.7, pp.3142-3155, 2017. https://doi.org/10.1109/TIP.2017.2662206
  11. Pengju Liu, Hongzhi Zhang, Kai Zhang, Liang Lin, and Wangmeng Zuo., "Multi-level Wavelet-CNN for Image Restoration." arXiv preprint arXiv:1805.07071v2 [cs.CV], 2018.
  12. Kazi Nazmul Haque, Mohammad Abu Yousuf, and Rajib Rana. "Image denoising and restoration with CNN-LSTM Encoder Decoder with Direct Attention." arXiv preprint arXiv:1801.05141v1 [stat.ML], 2018.
  13. David M. Gauntt and Gary T. Barnes, "A novel technique to suppress grid line artifacts," Medical Physics, Vol.33, No.6, pp.1654-1667, 2006. https://doi.org/10.1118/1.2184444
  14. Frederic Sur and Michel Grediac, "automated removal of quasiperiodic moise using frequency domain statistics," Journal of Electronic Imaging, Vol.24, No.1, pp.013003-1-19, 2015. https://doi.org/10.1117/1.JEI.24.1.013003