Theoretical Background and State of the Art - Historical Land Use/Land Cover Classification Using Remote Sensing - page 30

geographical coverage at continental and global scales, but need detailed spatial
information. The fact that not each pixel in an image represents only single crop
type can introduce uncertainty into area estimates because of the mixture
(Ozdogan and Woodcock 2006 ). Where cultivated areas are smaller than the
spatial resolution of the image, here, both cultivated and uncultivated areas (e.g.,
roads, houses, irrigation channels) are integrated in a pixel classified as agriculture
or cropland. In agricultural situations, the amount of uncultivated area has been
reported to vary from 10 to 40 % (Crapper 1980 ; Frolking et al. 1999 ). To rela-
tively solve this mixed pixel problem which occurs especially in high temporal
resolution data at low spatial resolution, some contributors have developed tech-
niques that use the concept of temporal un-mixing (Adams et al. 1986 ). It is similar
to the traditional spectral un-mixing technique, where pure end-members are
distinguished by their spectral response. Temporal un-mixing uses end-members
defined by their single temporal response to improve the fractional area of each
end-member based on its part to the mixed temporal reaction observed by the
sensor (Ozdogan 2010 ).
There are two generally used area estimation methods with remote sensing
(Ozdogan and Woodcock 2006 ). The first method calculates portions/fractions of a
thematic category of interest for each pixel (Hansen et al. 2002 ). The essential
drawback here is the accuracy assessment of fractions of the thematic field.
However, area estimation by this method is becoming more common (Liu and Wu
2005 ). A second method is based on generating the thematic map through image
classification and then multiplying the area of the pixels with their number in a
specific class. The drawback here is the classification accuracy of the thematic map
(Ozdogan and Woodcock 2006 ).
References
Abuzar, M., McAllister, A., & Morris, M. (2001). Classification of seasonal images for
monitoring irrigated crops in a salinity-affected area of Australia. International Journal of
Remote Sensing, 22(5), 717-726.
Adams, J. B., Smith, M. O., & Johnson, P. E. (1986). Spectral mixture modeling: A new analysis
of rock and soil types at Viking Lander I site. Journal of Geophysical Research, 91 (B8),
8098-8112.
Adam, R., Adams, M., & Willens, A. (1978). Dry lands: Man and plants (1st ed.). London: The
Architectural Press Ltd.
Alberga, V. (2009). Similarity measures of remotely sensed multi-sensor images for change
detection applications. Remote Sensing, 1(3), 122-143.
Bagan, H., Jianwen, M., Qiqing, L., Xiuzhen, H., & Zhili, L. (2004). Land-use classification from
ASTER data using wavelet fusion and self-organized neural networks. Science in China
Series D-Earth Sciences, 47(7), 651-658.
Bastin, G. N., Pickup, G., & Pearce, G. (1995). Utility of AVHRR data for land degradation
assessment: A case study. International Journal of Remote Sensing, 16(4), 651-672.
Beaumont, P. (1996). Agricultural and environmental changes in the upper Euphrates catchment
of Turkey and Syria and their political and economic implications. Applied Geography, 16(2),
137-157.