Spatial Information Research

Korea Spatial Information Society (대한공간정보학회)
권호 표지

Wavelet Compression Experiments of the Remotely Sensed Images for Three Kinds of Wavelet Families

  • Jin, Hong-Sung (Department of Applied Mathematics, The Chonnam National University) ;
  • Han, Dong-Yeob (Department of Civil & Environmental Engineering, The Chonnam National University)
  • Published : 2009.12.31
Download PDF
⟨ Previous Next ⟩

Abstract

A method to find the nearly optimal PSNR values for compression was tried to remotely sensed images. There is no rule to find the best wavelet pairs for image processing. The expected wavelet pairs following the suggested algorithm showed the optimal result for various kinds of images. Firstly, the PSNR variations with three wavelet families were analyzed. In many cases the longer wavelet filter shows the higher PSNR value, but the rate is getting less in orthogonal wavelet families. Wavelets with moderate filter length are suggested at the point of computational cost. For biorthogonal families it was hard to predict from the length of filters. Multiresolution wavelet analysis was used up to level 3 with three kinds of wavelet families. Biorthogonal wavelet family showed irregular pattern to get the maximum PSNR values, while orthogonal wavelet families showed regular pattern. In orthogonal wavelet families the nearly optimal wavelet pair can be predicted from the level 1.

원격탐사 영상에서 압축을 위한 근최적의 PSNR 값을 찾는 방법을 연구하였다. 예상 웨이블릿쌍은 다양한 영상에서 최적의 결과로 나타났다. 영상처리를 위한 최고의 웨이블릿쌍을 찾는 규칙은 없다. 제시된 알고리즘에 따라 예상 웨이블릿쌍이 다양한 종류의 영상에서 최적의 결과를 나타냈다. 먼저 세 개의 웨이블릿 패밀리에서 PSNR 값의 변화를 분석하였다. 직교 웨이블릿 패밀리에서는 많은 경우에 웨이블릿 필터의 길이가 길수록 높은 PSNR 값을 나타내지만, 그 증가 비율이 점차로 작아졌다. 연산비용의 측면에서 중간 필터길이의 웨이블릿을 제안한다. 쌍직교 웨이블릿 패밀리에서는 필터의 길이와 PSNR값의 관계를 예측하기는 어려웠다. 다차원 웨이블릿 분석에서는 세 개의 웨이블릿 패밀리가 3단계까지 처리되었다. 쌍직교 웨이블릿 패밀리는 최대 PSNR 값에서 불규칙한 패턴을 보였지만, 직교 웨이블릿 패밀리는 규칙적 패턴을 나타냈다. 직교 웨이블릿 패밀리는 1단계 결과로부터 근최적의 웨이블릿쌍을 예상할 수 있었다.

Keywords

References

  1. 윤공현, 2006, SFR기법을 이용한 영상 융합의 정확도 향상에 관한 연구, 한국GIS학회지, 14(1), pp.85-94.
  2. Anscombe, F. J., 1973, Graphs in statistical analysis, American Statistician, Vol.27, pp.17-21. https://doi.org/10.2307/2682899
  3. Chae, G. J., Park, J. H., Park, J. H., and Kim, K. O., 2003, Compression of Landsat Image using the spectral property and wavelet filter, Proceedings on IGARSS 2003, Vol.6, pp.3583-3585.
  4. Chun, W. J., Joo, Y. J., Moon, K. K., Lee, Y. I., and Park, S. H., 2005, A GIS Vector Data Compression Method Considering Dynamic Updates, The Journal of GIS Association of Korea, 13(4), pp.355-364.
  5. Daubechies, I., 1992, Ten lectures on wavelets, SIAM, Philadelphia.
  6. Gladston, R. S., Revathy, K., and Raju, G., 2008, Study on the choice of wavelet filters for image compression using neural and k-nearest neighbor classifiers, Journal of Wavelet Theory and Applications, Vol.2, No.1, pp.15-30.
  7. Jin, H. S., Yoo, H. Y., Eom, J. Y., Choi, I. S., and Han, D. Y., 2009, Wavelet Pair Noise Removal for Increasing the Classification Accuracy of a Remotely Sensed Image, Korean Journal of Remote Sensing, Vol.25, No.3, pp.1-9.
  8. Mallat, S. G., 1998, A wavelet tour of signal processing, Academic press.
  9. Raghuveer M. R., Ajit S. B., 1998, Wavelet Transforms, Addison Wesley.
  10. Strang, G., Nguyen, T., 1996, Wavelets and Filter Banks, Wellesley and Cambridge Press.
  11. Thomos, N., Boulgouris, N. V., and Strintzis, M. G., 2006, Optimized Transmission of JPEG2000 Streams Over Wireless Channels, IEEE Transactions on Image Processing, Vol.15, No.1, pp.54-67. https://doi.org/10.1109/TIP.2005.860338
  12. Yoo, H. Y., Lee, K. W., Jin, H. S., and Kwon, B. D., 2008, Selecting Optimal Basis Function with Energy Parameter in Image Classification Based on Wavelet Coefficients, Korean Journal of Remote Sensing, Vol.24, No.5, pp.437-444. https://doi.org/10.7780/kjrs.2008.24.5.437