Detection of Alteration Minerals Using Hyperion Data Analysis in Lahroud
- PDF / 3,773,043 Bytes
- 9 Pages / 595.276 x 790.866 pts Page_size
- 10 Downloads / 172 Views
RESEARCH ARTICLE
Detection of Alteration Minerals Using Hyperion Data Analysis in Lahroud Majid M. Oskouei 1 & Solmaz Babakan 1
Received: 24 January 2015 / Accepted: 5 January 2016 # Indian Society of Remote Sensing 2016
Abstract The study aims to detect alteration indicative minerals on a part of Hyperion scene in the Lahroud region using the image processing methods. However, it is rarely possible to find actually pure pixels in the mineralogical scale inside the study scenes. This implies the necessity of the identification of sub-pixel materials before classification and mineral mapping using the spectral unmixing algorithms. The Linear Mixture Model (LMM) based standardized hyperspectral processing methodology was employed for this purpose. The necessary pre-processing tasks including the atmospheric and topographic corrections and data quality assessment were also utilized to increase the classification accuracy. The mineralogical and alteration map of the study area was then extracted and evaluated quantitatively with respected to the geological setting of the study area. Despite of the presence of complex facies in the region, the possibility of the applied methodology in the alteration mapping by linear unmixing was proved on Hyperion datasets. The low signal to noise ratio of the Hyperion sensor caused some difficulties but, considering the high cost and consumed time of the field sampling and geochemical studies, the applied method is an advantageous tool for primary steps of the exploration.
Keywords Hyperspectral images . Mineral detection . Spectral feature fitting . Mineral mapping
* Majid M. Oskouei [email protected]
1
Mining Engineering Faculty, Sahand University of Technology, Tabriz, Iran
Introduction Development of the hyperspectral technology reveals a completely new perspective in many remote sensing applications especially in the mineral exploration programs. Hyperspectral sensors acquire image data simultaneously in hundreds of narrow adjacent spectral bands which makes it possible to obtain a continuous spectrum for each image pixel (Oskouie and Busch 2008; Babakan and Oskouei 2015). The solar spectral range of 0.4–2.5 μm provides valuable information about the earth surface. In particular, the shortwave infrared (SWIR) spectral range (2.0–2.5 μm) covers spectral features of hydroxyl-bearing minerals, sulphates, and carbonates common to the many geologic units and hydrothermal alteration assemblages. The detection of the spectrally distinct alteration minerals such as kaolinite, alunite, muscovite, and pyrophyllite are important in mineral exploration and characterization. Therefore, considering their high spectral resolution, the hyperspectral sensors could be efficiently exploited in mineral identification and distribution mapping even in sub pixel abundances (Kruse and Boardman 2000; Kruse et al. 2003). The mixed pixels are a challenging problem in the hyperspectral remote sensing which arises from their low spatial resolution. Accordingly, it is impossible to find pixels purel
Data Loading...
