Using Principal techniques on ETM+ 2002 for arid and semi-arid environment Central Iran PC5=-0.07 (B1) -0.014 (B2) +0.014 (B3) +0.13(B4)-0.71(B5)+0.68(B7)…………………………eq.3  Fig. 7. Explicit the clay minerals using band ratio ( 5/7) image in PC5 PC3 has four negative values (B1 to B4) and two positive values in band 5 and 7. The higher loading is in band 7 (0.49) and loading value in band 1 (-0.48) but in negative sign. It shows that amphibole minerals are more enhanced and can be seen in bright appearance on the image. By applying formula (Eq.2), it reveals that band ratio 7/1 and 7/3 are the best for enhancing amphibole minerals. In PC3 the granitic minerals appear dark color having low pixel values. The analyses indicate that separation of the minerals can be possible using above techniques. One of the best is called principal components analysis. It is still difficult to identify the rocks without on-site information "ground-truth". Usually, color anomalies are difficult to explain from the imagery alone and warrant examination on the ground. However, different spectral behavior on objects is given in Fig 8. The classification on different FCC images beside OIF method for understanding the best image combination using Crosta method is becoming more realizable for discrimination of the minerals. Classification processing on the best FCC images (Fig 9) derived from OIF method shows that applying PCA Crosta method in the same images gives more accurate information. Hence, it has become much easier to identify contacts and even variations within a particular rock unit. The above analyses indicated that the minerals variability in the area on account of the environmental impact and chemical weathering on the surface of the landscape which are responsible changing the geochemical characteristics and its reflective properties.  Fig. 8. Showing spectral trends of objects  Fig. 9. Showing classification of image 5. Conclusions Environment condition is one of the factors to hinder right information at right time by conventional techniques. This method provides real information through digital analysis of satellite data. Geologic maps (assuming that they are even available) may be inaccurate and, at best, are usually generalizations. The geologist doing the mapping may have painstakingly mapped the boundaries of a particular formation and missed an obvious mineralization zone within it. Similarly, geologic structures and mineral information can be quite vivid on satellite imagery. Subtle color variations that would go unnoticed on the ground can be made quite bold in false-color renditions, begging to be assayed, as it were. Known mining areas can sometimes be identified by using satellite imagery. However, this study shows that Crosta method with applying PCA is useful to identify a PC in which give more information about minerals and rock types. It is seen that PC3 enhance the amphibole minerals and amphibolite rocks. PC5 also enhances the clay minerals on the image. OIF is useful to identify the specific spectral bands which give more information about minerals. Thus combining of two methods is highly increasing the accuracy to identify the minerals and rock type in arid and semi arid environment having scanty vegetation. Field observations have suggested the combination of OIF and Crosta methods is more accurate than use in separate form. While capable of producing images similar to the ETM+ scanners, their real advantage is providing a spectral signature for each pixel in the image. By matching laboratory-derived reflection spectra of minerals against the individual pixel spectra, the dominant mineral and rock type (Igneous body). Under ideal conditions, even mixtures of two or three minerals can be identified. However, there is the potential for prospecting directly for key minerals associated with economically interesting mineralization. This study reveals that I various processes of rock discrimination and mineral identification in arid and semi-arid environment, satellite data across field observation is preferable. It also shows that applying PC analysis could be used for specific minerals. Thus, comparative study shows that arid and semi-arid environment condition is responsible for fluctuation in the spectral variability in the area due to fluctuating of environmental condition. This area required high frequency analysis and multi angle high resolution data to manage and monitor the impact of the environment in the area. Acknowledgements The authors are thankful to staffs of the Islamic Azad University of Dezful Iran for their help and continue encouragement during this work. The authors are thankful to Leila Farzinpor for her kind support. Reference:
Page 3 of 3
| Previous | |