VPLIV STISKANJA JPEG2000 Z IZGUBAMI NA KLASIFIKACIJO POSNETKOV WORLDVIEW-2
EFFECTS OF LOSSY JPEG2000 COMPRESSION METHOD ON WORLDVIEW-2 IMAGE CLASSIFICATION

Aleš Marsetič, Žiga Kokalj, Krištof Oštir

DOI: 10.15292/geodetski-vestnik.2012.02.275-289

 

Izvleček:

Stiskanje z izgubami se vse pogosteje uporablja v daljinskem zaznavanju, čeprav njegov vpliv na rezultate obdelav posnetkov še ni v celoti raziskan. V članku so predstavljeni učinki stiskanja JPEG2000 z izgubami na klasifikacijo posnetkov zelo visoke ločljivosti WorldView-2. Za klasifikacijo stisnjenih posnetkov smo prvič uporabili objektni metodi k-najbližji sosed (k-nearest neighbor) in metodo podpornih vektorjev (support vector machine), ki smo ju tudi primerjali. Rezultati razkrivajo vpliv stiskanja na same posnetke, segmentacijo in končno klasifikacijo. Študija dokazuje, da v splošnem stiskanje z izgubo ne vpliva negativno na klasifikacijo posnetkov, še več, v nekaterih primerih ima klasifikacija stisnjenih posnetkov boljše rezultate kot klasifikacija izvirnih posnetkov. Natančnost klasifikacij z metodo podpornih vektorjev kaže na možnost stiskanja podob do razmerja 30 : 1 brez izgube natančnosti klasifikacije. Najboljši rezultat z metodo k-najbližji sosed smo pridobili z najvišjim kompresijskim razmerjem (100 : 1). V raziskavi smo
ugotovili, da metoda podpornih vektorjev daje boljše rezultate klasifikacije kot metoda k-najbližji sosed ter se priporoča tudi za nadaljnje raziskave. Poleg metode klasifikacije ima pri natančnosti rezultatov pomembno vlogo tudi segmentacija posnetkov.

Ključne besede: slikovno stiskanje z izgubami, objektna klasifikacija, JPEG2000, WorldView-2, metoda podpornih vektorjev, k-najbližji sosed

 

Abstract:

Lossy compression is becoming increasingly used in remote sensing, although its effect on the processing results has yet not been fully investigated. This paper presents the effects of JPEG2000 lossy compression on the classification of very high-resolution WorldView-2 imagery. For the first time, the k-nearest neighbor and support vector machine methods of the object based classification were used. The results explore the impact of compression on the images, segmentation and resulting classification. The study proves that in general lossy compression does not adversely affect the classification of images; moreover, in some cases classification of compressed images yields better results than classification of the original image. Classification accuracy of the support vector machine method indicates that compression ratios of up to 30:1 can be used without any loss of classification accuracy. The best result of the k-nearest neighbor method was  obtained with the highest compression ratio (100:1). The support vector machine is recommended for further research. In addition to the classification method, image segmentation also plays an important role in the accuracy of the results.

Keywords: image lossy compression, object classification, JPEG2000, WorldView-2, support vector machine, k-nearest neighbour

 

Literatura / References:

Aiazzi, B., Baronti, S., Selva, M., Alparone, L. (2006). Enhanced Gram-Schmidt Spectral Sharpening Based on Multivariate Regression of MS and Pan Data. Denver, ZDA, Proceedings of IGARSS 2006, 3789–3792.

Blaschke, T. (2010). Object based image analysis for remote sensing. ISPRS Journal of Photogrammetry and Remote Sensing, 65(1), 2–16.
http://dx.doi.org/10.1016/j.isprsjprs.2009.06.004

Boser, B. E., Guyon, I. M., Vapnik, V. N. (1992). A training algorithm for optimal margin classifiers. V D. Haussler (ur.), 5th Annual ACM Workshop on COLT. Pittsburgh, ZDA: ACM Press, 144–152.

Fix, E., Hodges, J. L. (1989). Discriminatory Analysis. Nonparametric Discrimination: Consistency Properties. International Statistical Review / Revue Internationale de Statistique, 57(3), 238–247.
http://dx.doi.org/10.2307/1403797

García-Vílchez, F., Mu~n oz-Marí, J., Zortea, M., Blanes, I., González-Ruiz, V., Camps-Valls, G., Plaza, A., Serra-Sagrist`a. (2011). On the Impact of Lossy Compression on Hyperspectral Image Classification and Unmixing. IEEE Geoscience and Remote Sensing Letters, 8(2), 253–257.
http://dx.doi.org/10.1109/LGRS.2010.2062484

Goodenough, D. G., Dyk, A., Han, T., Jazayeri, A., Li, J. (2004). Impacts of Lossy Compression on Hyperspectral Products for Forestry. Anchorage, Proceedings of IGARSS 2004, 465–468.

ITT Visual Information Solutions (2008). ENVI User's guide. Boulder: ITT VIS.

Janža, M. (2005). Določitev rabe tal s klasifikacijo satelitske podobe za namene hidrološkega modeliranja na območju zaledja izvira Rižane. Geologija, 48(1), 153–159. URN:NBN:SI:DOC-X4J1PAM2

Kiema, J. B. K. (2000). Effect of wavelet compression on the automatic classification of urban environments using high resolution multispectral imagery and laser scanning data. Amsterdam, International Archives of Photogrammetry and Remote Sensing, 33 (Part B3), 488–495.

Laben, A., Brower, B. V. (2000). Process for enhancing the spatial resolution of multispectral imagery using pan-sharpening. U.S. Patent 6011875.

Lam, K. W., Lau, W., Li, Z. (2000). The effects on image classification using image compression technique. Amsterdam, International Archives of Photogrammetry and Remote Sensing, 33 (Part B7), 744–751.

Lang, S. (2008). Object-based image analysis for remote sensing applications: modeling reality – dealing with complexity. V Thomas Blaschke, Stefan Lang, in G. J. Hay (ur.), Object-Based Image Analysis, Lecture Notes in Geoinformation and Cartography. Berlin Heidelberg: Springer-Verlag, 3–27.

Liang, Z., Tang, X., Zhang, G., Wu, X. (2008). Effects of JPEG2000 and SPIHT compression on image classification. Peking, The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 37 (Part B7), 541–544.

Mercier, G., Mouchot, M.-C., Cazuguel, G. (1999). Joint classification and compression of hyperspectral images. Hamburg, Proceedings of IGARSS 1999, 28(4), 2035–2037.

Serra-Sagrist`a , J., Aulí-Llin`a s, F. (2008). Remote Sensing Data Compression. V M. Gra~n a in R. J. Duro (ur.), Computational Intelligence for Remote Sensing. Berlin, Nemčija: Springer-Verlag, 27–61.
http://dx.doi.org/10.1007/978-3-540-79353-3_2

Standard ISO/IEC 15444-1 (2004).

Taubman, D. S., Marcellin, M. W. (2002). JPEG2000: Image compression fundamentals, standards and practice. Boston, ZDA: Kluwer academic publishers.

Zabala, A., Pons, X. (2011). Effects of lossy compression on remote sensing image classification of forest areas. International Journal of Applied Earth Observation and Geoinformation, 3(1), 43–51.
http://dx.doi.org/10.1016/j.jag.2010.06.005