A comparative study of the denoising methods of Thematic Mapper images for forest areas

Abstract

The noises of remote sensing images, caused by imaging system and ground environment, negatively affect the accuracy and efficiency in extracting forest information from remote sensing images. The denoising is critical for image classifications for forest areas. The objective of this research is to assess the effectiveness of currently used spatial filtering methods for extracting with forest information related from Landsat 5 TM images. Five spatial filtering methods including low-pass filter, median filter, mean filter, sigma filter and enhanced self-adaptive filter were examined. A set of evaluation indices was designed to assess the ability of each denoising method for flatness, edge/boundary retention and enhancement. Based on the designed evaluation indices and visual assessment, it was found that sigma filter (D=1) and enhanced self-adaptive filter were the most effective denoising methods in classifying TM images for forest areas.

Keywords

Denoising / Edge/boundary retention / Enhanced self-adaptive filter / TM image

References

[1] Aikenhead M.J., Mustard M.J., McDonald A.J.S. Using neural networks to predict spatial structure in ecological systems Ecoloigcal Modelling, 2004, 179: 393-403.

[2] Bolstad V.P., Liuesand T.M. Improved classification of forest vegetation in Wisconsin through a rule-based combination of soils, terrain, and Landsat Thematic Mapper data Forest Science, 1992, 38(1): 5-20.

[3] Cibula W.G., Nyquist M.D. Use of topographic and climatlogical models in a geographical database to improve Landsat Mss classification for Olympic National park PE&RS, 1987, 53(1): 67-75.

[4] Dang A., Wang X., Chen X. et al. The processing methods to remote sensing image by ERDAS IMAGINE software, 2003 Beijing: Beijing of China:Tsinghua University Press 577-578.

[5] Delouille V.R., Maarten Jansenb M., Rainer von Sachs R. Second-generation wavelet denoising methods for irregularly spaced data in two dimensions, Signal processing Signal Processing, 2005, 86(7): 1435-1450.

[6] Du P. Research on filtering of RADARSAT image Journal of Chinese Mining Technology University, 2002, 31(2): 132-137.

[7] Gross, C.P., 1999. Contributions of Remote Sensing to Forest Health Monitoring in Europe beyond 2001. In: Conference on of Remote Sensing and Forest Monitoring Proceedings [J/OL], Retrieved from: http://rogow99.sggw.waw.pl.

[8] Lee J.S. Digital image enhancement and noise filtering by use of local statistics. IEEE transactions on pattern analysis and machine intelligence PAMI, 1980, 2(2): 165-168.

[9] Li C., Yan Y., Cheng J. et al. Speckle reduction for high resolution one look spaceborne SAR images Acta Electronica Sinica, 2000, 28(3): 13-16.

[10] Liu Z., Zheng C., Fang J. Relationship between the vegetation type and topography in Mt. Xiaowutai, Hebei province: a remote sensing analaysis Biodiversity Science, 2004, 12(1): 146-154.

[11] Mei A., Peng W., Qing Q. et al. Remote Sensing Generality, 2001 Higher Education Press: Beijing, China 352-355.

[12] Mori M., Yamamoto N., Yokoyama T. Enhancement of JERS-1 SAR imagery The Third Thematic Conference on Remote Sensing for Marine and Coastal Environments, Seattle, WA., 1995, 2: 801-807.

[13] Morillas S., Gregori V., Peris-Fajarne’s G. et al. A fast impulsive noise color image filter using fuzzy metrics, Real-Time Imaging, 2005, 11: 417-428.

[14] Mueller, P.W., 1987. Spatial filtering of Shuttle Imaging dar-B data. LARS-Purdue University, AT&T Bell Laboratories Technical Report 091487, West Lafayette, Indiana, U.S.A.

[15] Price, M.H. 2000. Application of remote sensing to forest resource inventory and habit at modeling. NASA NRA-98-OES-09 [J/OL]. Retrieved from: http://www.sdsmt.edu/research/bhrs

[16] Rigina, O. and Olsson, H. K. 1999. Detection of boreal forest decline by high-resolution satellite imagery. In: Conference on of Remote Sensing and Forest Monitoring Proceedings [J/OL], Retrieved from: http://rogow99.sggw.waw.pl.

[17] Rio J.N.R., Lozano G.D.F. Spatial filtering of radar data (RADARSAT) for wetlands (Brackish Marshes) classification Remote Sensing, 2000, 73: 143-151.

[18] Seyrafi K., Davison G. A. Spatial filter synthesis [J]. Infrared Physics, 1964, 4: 263-274.

[19] Skidmore A.K. An expert system classifies Eucalyptus forest types using Thematic Mapper data and a digital terrain model. Photogramm. Engineering Remote Sensing, 1989, 55(10): 1449-1469.

[20] Smolkaa B., Plataniotisb K.N., Chydzinskia A. et al. Self-adaptive algorithm of impulsive noise reduction in color images Pattern Recognition, 2002, 35: 1771-1784.

[21] Tan Bingxiang. 1996. Application and discussion of satellite remote sensing technology in Chinese forestry. Forest resource managing, (special): 183–187. (In Chinese)

[22] Tang L., Jiang P., Dai C. et al. Evaluation of smoothing filters suppressing speckle noise on SAR images Remote Sensing of environment. China, 1996, 31(3): 206-211.

[23] Vardavoulia M.I., Andreadis I., Tsalides P. A new vector median filter for colour image processing Pattern Recognition Letters, 2001, 22: 675-689.

[24] Weng W.G., Fan W.C., Liao G.X. et al. Wavelet-based image denoising in (digital) particle image velocimetry Signal Processing, 2001, 81(7): 1503-1512.

[25] Xu L. Application of wavelet analysis in compressing images, 2002 Wuhan: Master Thesis of Wuhan University of Science, China 15-19.

[26] Xu X., Liao M., Zhu P. et al. Research on Speckle Filtering of Single-look SAR Image Journal of Wuhai University of Surveying Science, 1999, 24(4): 312-315.

[27] Yang X. Estimating Forest Volume of Pure County by Quantitative Methods Forest Inventory and Planning, 2002, 27(1): 49-52.

[28] Yue C. A Study of monitoring forest vegetation of the long defence forest project in Yun Nian province Journal of Jiangxing Agriculture University, 2004, 26(1): 42-45.

[29] Zhu N. Some research on processing and evaluating for SAR images, 2003 Nanjing: Master Thesis of Nanjing University of Aeronautics and Astronautics, China 18-19.