Wykorzystanie transformacji falkowej do oceny spadku jakości radiometrycznej w procesie mozaikowania ortofotomapy

• Autorzy: Pyka K.

Warianty tytułu

EN

The use of wavelets for evaluation of loss in radiometric quality in the orthophoto-mosaicking process

• Języki publikacji: PL

Abstrakty

PL

Przedmiotem pracy jest wykorzystanie transformacji falkowej do ilościowej oceny degradacji kontrastu, jaki ma miejsce podczas kompilowania ortofotomapy z ortoobrazów. Jest to problem rzadko podejmowany, a tymczasem w sytuacji masowego wykorzystywania ortofotomapy coraz większej wagi nabiera ocena jakości, także radiometrycznej. W pracy wyjaśniono przyczyny spadku kontrastu przy mozaikowaniu. Następnie pokrótce scharakteryzowano transformację falkową, przedstawiając istotę zachowania wariancji podczas dekompozycji. Dla kontrastu krytyczne znaczenie mają wariancje komponentów detalicznych. Badania polegały na obserwacji wariancji podczas dekompozycji ortofotomap i źródłowych ortoobrazów. Przebadano 26 przypadków. We wszystkich widoczny był spadek wariancji, ale tylko w czterech skutki rozmycia krawędzi były wykrywane przez obserwatora. Wyniki potwierdziły możliwość wnioskowania o jakości radiometrycznej na podstawie badania wariancji komponentów. W obecnym stadium badań wskaźniki falkowe mają głównie charakter porównawczy.

EN

In the paper, the use of wavelet transformation for evaluation of loss in radiometric quality during orthophoto mosaic process is proposed. The automation of the production and relatively low costs of development imply that the orthophotomap is one of the more important and probably the most popular geo-information technology product. Due to the global application of the orthophotomap, as it is particularly evident in case of web geoportals, the quality assessment becomes more and more important. During the end stage of orthophoto production, the individual rectified images are assembled to form seamless mosaic. The process is executed with the aim to achieve mosaics with good radiometric quality. Due to the different interpretation of the term: "radiometric quality", the question which image is good or better than other is difficult to rate. The human perception of the picture is very complex. At the first look on image, the slope of illumination and then the colour changes are detected. That is why during the mosaicking, feathering technique and radiometric balancing are applied. However, such techniques could result in loss of local contrast and small object could not be recognized from the back ground. This is why there is a need to look for the indicators of radiometric quality. For some years, the discrete wavelet transformation has been used in the image processing. The wavelet transformation is regarded as the most effective method of lossy compression of multitonal images [1, 6] and it is used in fusion of images [7], etc. The wavelet representation of the image can be also used for evaluation of the image radiometric quality, first of all the noise contents [8, 10, 11, 13]. In the paper, based on the studies, it has been proven that the analysis of the equation of preservation of image relative variance is a good indication of the local contrast preserving. In the described research, the variance of wavelets components in 3D scale, for a set of orthophotomaps and orthoimages, was compared. The test material was taken from typical photogrammetric project in which the orthophotomaps from DMC images with GSD 10 cm was produced. In all examined cases, loss of variance of orthophotomaps details' components in comparison to orthoimages has been noted. This result was expected: due to tonal balancing the edge contrast is reduced. The question is how significant the degradation for photointerpretation is? In 4 out of 26 cases, the degradation was significant for interpretation, for example some technical devices on the streets and some small plants were not recognizable. The potential of wavelet transform applied for valuation of the part of radiometric quality has been proved in this study. In the current phase of researches, the results of wavelet indicators are mainly comparative - it is possible to estimate in which image there are better contrasts. Such researches should be carried out in the future, because unsatisfactory quality of the radiometric images reduces the interpretation value of the images, including orthophotomaps, so popular nowadays.

Słowa kluczowe

PL

geodezja i kartografia; ortofotomapa; jakość radiometryczna; transformacja falkowa

EN

geodesy and cartography; orthophotomap; radiometric quality; wavelets transform

• Wydawca: Wojskowa Akademia Techniczna im. Jarosława Dąbrowskiego
• Czasopismo: Biuletyn Wojskowej Akademii Technicznej
• Rocznik: 2011
• Tom: Vol. 60, nr 3
• Strony: 353--364
• Opis fizyczny: Bibliogr. 15 poz., tab., wykr.

Twórcy

autor

Pyka K.

• Akademia Górniczo-Hutnicza im. Stanisława Staszica, Katedra Geoinformacji, Fotogrametrii i Teledetekcji Środowiska, 30-059 Kraków, Al. Mickiewicza 30,

Bibliografia

• [1] R. W. Buccigrossi, E. P. Simoncelli, Image compression via joint statistical characterization in the wavelet domain, IEEE Trans. on Image Processing, 8, 1999, 1688-1701.
• [2] P. Jun, W. Mi, L. Deren, Auto-detection of heterogeneous areas in radiometric normalization, The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 37, Part B4, Beijing, 2008, 783-786.
• [3] M. Madani, Today's Orthophoto Production - The Business Model, Photogrammetric Week, 2007.
• [4] S. Mallat, A Theory for Multiresolution Signal Decomposition: The Wavelet Representation, EEE Trans. on Pattern Analysis and Machine Intelligence, 11, 1989, 674-693.
• [5] S. Mallat, A Wavelet Tour of Signal Processing. Academic Press, 1998.
• [6] C. Mulcahy, Image Compression Using The Haar Wavelet Transform, Spelman College Science & Mathematics Journal, 1, 1, April 1997, 22-31.
• [7] G. Pajares, J. M. de la Cruz, A wavelet-based image fusion tutorial, Pattern Recognition 37, 2004, 1855-1872.
• [8] J. Portilla, V. Strela, M. J. Wainwright, E. P. Simoncelli, Image Denoising using Scale Mixtures of Gaussians in the Wavelet Domain, IEEE Transactions on Image Processing, 12, 111, November 2003, 1338-135.
• [9] K. Pyka, Falkowe wskaźniki zmian radiometrycznych zachodzących w procesie opracowania ortofotomapy, UWND AGH, Kraków, 2005.
• [10] K. Pyka, Zastosowanie transformacji falkowej do detekcji i usuwania szumów z danych rastrowych i pseudo-rastrowych, Archiwum Fotogrametrii, Kartografii i Teledetekcji, 17, 2007.
• [11] K. Pyka, J. Siedlik, The Use of Wavelets for Noise Detection in the Images taken by the Analog and Digital Photogrammetric Camera, International Archives of Photogrammetry and Remote Sensing. 37, part B1, 2008.
• [12] K. Pyka, Jak ocenić jakość fotometryczną ortofotomapy?, Archiwum Fotogrametrii, Kartografii i Teledetekcji, 19, 2009.
• [13] E. P. Simoncelli, E. H. Adelson, Noise removal via Bayesian Wavelet Coring, Proceedings of 3rd IEEE International Conference on Image Processing, IEEE Signal Processing Society, Lausanne, Switzerland, 16-19 September 1996, 1, 1996, 379-382.
• [14] M. W. Sun, J. Q. Zhanga, Dodging research for digital aerial images, The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 37, Part B4, Beijing, 2008.
• [15] G. S. Yang, H. L Zhang, Optimal Image Mosaic Wavelet Method Based on Fuzzy Integral, Proceedings of the Fifth International Conference on Machine Learning and Cybernetics, Dalian, 2006.