Rice cultivated areas and yield information is indispensable for sustainable management and economic policy making for this strategic food crop. Introduction of high spectral and special resolution satellite data has enabled production of such information in a timely and accurate manner. Knowledge of the spectral reflectance of various land covers is a prerequisite for their identification and study. Evaluation of the spectral reflectance of plants using field spectroradiometry provides the possibility to identify and map different rice varieties especially while using hyperspectral remote sensing. This paper reports the results of the first attempt to evaluate spectral signatures of seven north Iranian rice varieties (Fajr, Hybrid, Khazar, Nemat, Neda, Shiroudi and Tarom plots) in the experimental station of the Iranian Rice Research Institute (main station in Amol, Mazanderan Province). Measurements were carried out using a field spectroradiometer in the range of 350-2, 500 nm under natural light and environmental conditions. In order to eliminate erroneous data and also experimental errors in spectral reflectance curves, all curves were individually quality controlled. A set of important vegetation indices sensitive to canopy chlorophyll content, photosynthesis intensity, nitrogen and water content were employed to enhance probable differences in spectral reflectance among various rice varieties. Analysis of variance and Tukey’s paired test were then used to compare rice varieties. Using Datt and PRI1 indices, significant differences (a=0.01) were found among rice varieties reflectances in 19 out of 21 cases. This promises the possibility of accurate mapping of rice varieties cultivated areas based on hyperspectral remotely sensed data.

Homogeneity and uniformity of soil particles is a most potent factor which affects soil spectral signatures. Geometric standard deviation of soil particles (sg) exhibits the homogeneity of soil components. This study presents the influence of sg on spectral studying of soil geometric mean particle diameter as well as lime. The study was conducted in Pol-e-Dokhtar using the data set of LISSIII-P6, September 7th 2007. Subsequent to satellite data preprocessing, some such processes as: Normalized Difference Vegetation Index, Principal Component Analysis, Soil Line Euclidean Distance and Unsupervised Classification were carried out on the acquired data. By stratified randomized sampling method and according to the false color composite as well as photomorphic units of the main image of the study area, 95 sample points were eventually selected and samples collected from 0-5cm of soil surface. Geometric standard deviation (sg), geometric mean particle diameter (dg) and lime (CaCO3) were later determined for each soil sample in the lab. Samples were accordingly divided into two parts on the basis of the computed geometric standard deviation: the first, sg<10 (homogeneous soil samples) and the second, sg³10 (heterogeneous soil samples). Subsequently via correlation operations in SPSS for both groups, it was expressly displayed that for the first group (sg<10), dg and lime were very forcefully and explicitly correlated with the remotely sensed data, in comparison with the second group (sg³10). Hence, the geometric standard deviation of soil particles (sg) in the study region can powerfully impress soil spectral reflectance.

Spectral data which give valuable information, are usually represented with a high dimension according to the accuracy of instrument. During the last decades, applying mathematical methods such as Principal Component Analysis (PCA) has shown that spectral data has limited number of principal vectors. In the present study, the appropriate numbers of PC vectors for representing the spectral reflectance of natural dyes were investigated applying a set of 629 wool samples dyed with natural dyes. GFC, RMSE and color differences under two illuminants were considered to evaluate the magnitude of error. Experimental results showed that it is possible to represent the spectral data of natural dyes with almost 5 principal components with precious accuracy (GFC=0.999). In addition, the best results were obtained when the principal vectors are extracted from the same data set (natural dyes) in comparison to other sets such as dyed polyesters or 1269 Munsell set. J. Color Sci. Tech. 5 (2011), 51-60Ó Institute for Color Science and Technology.

Coating of fabrics with spectrally selective nanopigments, in order to manipulate their optical properties, has been introduced in recent years. This method presents a new field in high performance textiles which covers high technological aspects such as UV protective fabrics, Solar cells and cool fabrics as well. One of the most important approaches in prediction of reflectance factor of coated fabrics is based on Discrete Ordinate Method (DOM), namely, Kubelka-Munk or Multi Flux Model, where the fabric is considered as a continuous medium consists of a dispersion of nanopigments. However, ignored surface scattering on interfaces causes considerable error. Geometric optic, on the other hand, is utilized widely for prediction of spectral behavior of particles with very large dimension compared with incident wavelength. The model lacks limitation of DOM but is not appropriate for nanopigments of small diameters. According to the fact that, coated fabrics are systems consists of nano and micron sized materials, a theoretical model is proposed in which radiative properties of micro particles are described using Geometric Optic, and those for nano and submicron sized particles are described using Mie scattering theory. Subsequently, the experimental realization of proposed model has been conducted on Titanium dioxide nanopigments coated on polyester fabrics, and the successful comparison of the proposed model with experimental results confirms its suitability, and introduces a rather new procedure in the field of radiation transfer through mixed beds of elements with extremely different sizes.

The spectral reflectance of tree crown can be different from spectral reflectance of its leaves because of diverse leaf and branch angles as well as internal space of tree crown. For these reasons it is necessary to study spectral reflectance of both adaxial and abaxial surfaces of the leaves. Such information is necessity for modeling the reflectance of tree crown and forest stands. The main objective of this study was to obtain and study the spectral reflectance of both adaxial and abaxial leaves of beech and hornbeam in natural condition and to investigate their spectral differences using indices sensitive to chlorophyll, chlorophyll to carotenoid ratio and photosynthetic pigments. Field spectroradiometric measurements were performed using a portable spectroradiometer (ASD FieldSpec) in Kheyrud forest. A total of 52 trees were sampled and 312 spectra were recorded and analyzed. Spectral measurements cover the wavelength range between 350 – 2500 nm. The results of the spectral reflectance analysis of these two species showed that the abaxial spectral reflectance from 350 to 2500 nm was higher than the adaxial one for hornbeam species. However, for beech species in the visible region and far infrared region, the abaxial spectral reflectance was higher whereas in the near infrared it was lower than the adaxial one. For more detailed investigation of spectral reflectance difference for these two species, spectral indices sensitive to chlorophyll and carotenoid were calculated and statistically analyzed for both surfaces. The value of adaxial NDI index was found to be higher than abaxial for both species. In contrast, the values of adaxial SIPI and PRI indices were lower than abaxial. The differences significant (α= 0.01, p< 0.0001) for both species.

In this research, to evaluate vegetation cover percentage an to determine optimum vegetation indexes for multivariate regression modeling in the Ab-Mahi subbasin of Doorodzan dam watershed basin with the area of 15, 800 ha, usimg spectral reflection detected by ETM+ sensor of landsat satellite in the year 2002. In the first step, after primitive image processing, including radiometric and geometric correction using optimum index factor (OIF), false color composite map with band combination: RGB=741 were prepared for field studies. Then, by doing field studies, considering false color composite map 63 plots (1m *1 m) were randomly selected at the region area. The percentage of canopy cover were estimated in the field studies. In this regard the coordinates of each point was determined using GPS. These information were used in order to construct the vegetation cover incorporated with ETM+ imagery data. To obtain a reliable vegetation map six main bands of ETM+ and 26 artificial imagery were applied to training data set in ILWIS software. Finally the overall accuracy of prediction was calculated by coefficient of determination (r2) and root means of square error (RMSE). The constructed vegetation canopy map was used in land qualitative evaluation of rangeland for the study area. At last, It was shown that it is possible to model and determine the vegetative land cover percentage with an acceptable accuracy using satellite images (ETM+) and implementing appropriate digital analysis.