Page 11 - NGTU_paper_withoutVideo

Modern Geomatics Technologies and Applications

Fig. 6. Scatterplots of obtained LST of MODIS products at LDCM overpass time with LST of superior methods. (a) LST of
the superior method (NBEM) among conventional methods in 1st scene of LDCM, (b) LST of the superior method (VAng)
among KBM proposed methods in 1st scene of LDCM, (c) LST of the superior method (SRSC) among conventional methods
in 2nd scene of LDCM, (d) LST of the superior method (VAng) among KBM proposed methods in 2nd scene of LDCM.

4. Conclusion
In this paper, we proposed a knowledge based approach to overcome the errors and uncertainties in land surface emissivity
(LSE) estimation and consequently land surface temperature (LST) retrieval. The effectiveness of proposed KBMs is empirically
evaluated over two scenes of LDCM datasets and the LSEs achieved by individual conventional and proposed methods were
compared to the LSE product of the ASTER in cases of IBCC. Moreover, an alternative scaling method based on LST products
of MODIS was proposed for LST cross-comparison. In the proposed KBMs, an ensemble of conventional LSE methods can be
made flexibly based on characteristics of the study area and sensor data. Since the proposed methods use a combination of the
results of various LSE estimation methods, the effects of errors and uncertainty is reduced. In other words, the proposed KBMs
take advantage of the unique features of LSE estimation methods in order to overcome their shortcomings. The results obtained
by IBCC showed that in comparison with five conventional individual methods, the achieved results by the VAvg methods
demonstrated better performance in terms of RMSE and MD on both examined scenes. Also, the obtained LST of proposed
scaling method was evaluated by cross-comparison, the results given in Table 1 and Fig. 6a, b, c, d demonstrated that the proposed
2
2
methods provide better estimates in both examined datasets in terms of the three statistical R , the adjusted R , MD (Bias) and
RMSE measures.
In sum, since LSE is an important intrinsic property of the materials its accurate estimation with a greater computational cost
is valuable. In this regard, according to the experimental results, the proposed KBMs yielded a proper estimation for two datasets,
which demonstrated their stability in contrast to the conventional methods for LSE estimation and LST retrieval.
5. Acknowledgments
The authors wish to express their gratitude to any specific product, namely, LDCM imagery held in the USGS archives and
reprocessing datasets (landsat.usgs.gov) and the Jet Propulsion Laboratory (JPL) due to the ASTER spectral library (v2.0) and
the USGS Spectral Libraries. MODIS UCSB emissivity library and the MODIS products data were obtained through the online
Data Pool at the NASA Land Processes Distributed Active Archive Center (LP DAAC), USGS/Earth Resources Observation
and Science (EROS) Center, (lpdaac.usgs.gov).

6. References
[1] J.-x. Jiang, Q.-h. Liu, and H. Li, "A modified NDVI threshold method for estimating LSE from FY3A/VIRR data," in 2nd
International Conference on Remote Sensing, Environment and Transportation Engineering (RSETE), Nanjing, Jiangsu, China, 01-
03 Jun, 2012, pp. 1-4, doi:10.1109/RSETE.2012.6260355.
[2] M. Boonmee, "Land Surface Temperature and Emissivity Retrieval from Thermal Infrared Hyperspectral Imagery," Rochester
Institute of Technology, PHD Thesis, 2007.
[3] Z.-L. Li, B.-H. Tang, H. Wu, H. Ren, G. Yan, Z. Wan, et al., "Satellite-derived land surface temperature: Current status and
perspectives," Remote Sensing of Environment, vol. 131, pp. 14-37, 2013.
[4] J. A. Sobrino, J. C. Jiménez-Muñoz, G. Sòria, M. Romaguera, L. Guanter, J. Moreno, et al., "Land surface emissivity retrieval from
different VNIR and TIR sensors," IEEE Transactions on Geoscience and Remote Sensing, vol. 46, pp. 316-327, 2008.
[5] A. Van de Griend and M. Owe, "On the relationship between thermal emissivity and the normalized difference vegetation index for
natural surfaces," International Journal of remote sensing, vol. 14, pp. 1119-1131, 1993.
8