De-Interlace 기법을 이용한 내시경 영상의 화질 개선

Improvement of Endoscopic Image using De-Interlacing Technique

  • 신동익 (울산대학교 의과대학 서울중앙병원 의공학과) ;
  • 조민수 ((주)메디슨) ;
  • 허수진 (울산대학교 의과대학 서울중앙병원 의공학과)
  • 발행 : 1998.10.01

초록

초음파, 내시경 등의 NTSC 영상을 PC를 통해 획득하고, 고해상도의 YGA 모니터에 표시할 경우 주사변환 과정을 거치면서 치명적인 영상의 왜곡(tear-drop)이 나타난다. 본 연구에서는 이러한 왜곡을 해소하는 여러 가찌 방법을 살펴보고 실시간으로 왜곡을 보정할 수 있는 하드웨어를 PC상에서 구현하였다. 하드웨어 시스템은 De-Interlace 전용의 소자와 PCI bridge 등을 이용함으로써 고화질의 영상표현과 실시간의 영상전송이 가능하다 구현된 시스템에서 영상의 질은 눈에 띄게 향상되었으며, PC 기반의 시스템으로 구성함으로써 영상의 저장, 전송 및 텍스트의 기록 등 다양한 기능을 쉽게 구현할 수 있었다.

In the case of acquisition and displaying medical Images such as ultrasonography and endoscopy on VGA monitor of PC system, image degradation of tear-drop appears through scan conversion. In this study, we compare several methods which can solve this degradation and implement the hardware system that resolves this problem in real-time with PC. It is possible to represent high quality image display and real-time processing and acquisition with specific de-interlacing device and PCI bridge on our hardware system. Image quality is improved remarkably on our hardware system. It is implemented as PC-based system, so acquiring, saving images and describing text comment on those images and PACS networking can be easily implemented.metabolism. All images were spatially normalized to MNI standard PET template and smoothed with 16mm FWHM Gaussian kernel using SPM96. Mean count in cerebral region was normalized. The VOls for 34 cerebral regions were previously defined on the standard template and 17 different counts of mirrored regions to hemispheric midline were extracted from spatially normalized images. A three-layer feed-forward error back-propagation neural network classifier with 7 input nodes and 3 output nodes was used. The network was trained to interpret metabolic patterns and produce identical diagnoses with those of expert viewers. The performance of the neural network was optimized by testing with 5~40 nodes in hidden layer. Randomly selected 40 images from each group were used to train the network and the remainders were used to test the learned network. The optimized neural network gave a maximum agreement rate of 80.3% with expert viewers. It used 20 hidden nodes and was trained for 1508 epochs. Also, neural network gave agreement rates of 75~80% with 10 or 30 nodes in hidden layer. We conclude that artificial neural network performed as well as human experts and could be potentially useful as clinical decision support tool for the localization of epileptogenic zones.

키워드

참고문헌

  1. Cameras, Scanners, and Image Acquisition Systems Helen Marz;Robert L. Nielsen
  2. 서울대학교 박사학위논문 혈관내피세포의 형태학적 분석을 위한 의학영상처리 시스템의 개발 신동익
  3. Proc. SPIE v.3205 High-speed image processing algorithms using MMX hardware Miller John W.;Wood Jamse
  4. Proc. SPIE v.1446 Image acquisition unit for the Mayo'IBM project Reardon Frank J.;Salutz James R.
  5. Proc. SPIE v.3300 Effect of user controls on CRT monitor characteristics Deguchi Tatsuya;Katoh
  6. A sumitted for the Ph. D. High Quality De-interlacting of Television Images Nikco van Someren
  7. Proc. SPIE v.2666 Motion-compensated scan conversion for interlaced video sequencies Schultz Richard R.;Stevenson Robert R.
  8. Proc. SPIE v.2419 Motion-compensated in terpolation using trajectories with acceleration Chahine Michel;Konrad Janusz
  9. Proc. SPIE v.2308 Minimum mean square error(MMSE) design of generalized interpolation filters for the motion processing of interlaced image Vandendrope Luc;Delogne Paul;Cuvelier Laurnet;Maison B.
  10. Optical Engineering v.32 no.7 Motion-compensated generic coding of video based on a multiresolution data structure Dafaux Frederic;Moccagatta Iole;Rouchouze Brunco;Ebrahimi Touradj;Kunt Murat
  11. gmVLD8 Databook Genesis Microchip
  12. Hardware Design Guide for Window95 Microsoft
  13. PCI Hardware and Software Edward Solari;George Willse
  14. PCI System Architecture Tom Shanley;Don Anderson