当前位置:首页 >> 电力/水利 >>

The Feature Selection and Extraction of Hyperspectral Mineralization Information


Longwave

Thermal

Infrared

Spectral

Variability in Individual Rocks
Balick, L.; Gillespie, A.; French, A.; Danilina, I.; Allard, J.-P.; Mushkin, A.; Geoscience and Remote Sensing Letters, IEEE Volume: 6 , Issue: 1 Digital Object Identifier: 10.1109/LGRS.2008.2006005 Publication Year: 2009 , Page(s): 52 - 56 Cited by: 2 Abstract An hyperspectral imaging spectrometer measuring in the longwave thermal infrared (7.6-11.6 mum), with a spatial resolution less than 5 mm at a range of 10 m, was used in the field to observe the variability of emissivity spectra of individual rock surfaces. The rocks were obtained commercially, were on the order of 20 cm in size, and were selected to have distinct spectral features: they include alabaster (gypsum), soapstone (steatite with talc), obsidian (volcanic glass), norite (plagioclase and orthopyroxene), and ldquojasperrdquo (silica with iron oxides). The advantages of using an imaging spectrometer to characterize these rocks spectrally are apparent. Large spectral variations were observed within individual rocks that may be attributed to roughness, surface geometry, and compositional variation. Nonimaging spectrometers would normally miss these variations as would small samples used in laboratory measurements, spatially averaged spectra can miss the optimum spectra for identification of materials, and spatially localized components of the rock can be obscured.

The Feature Selection and Extraction of Hyperspectral Mineralization Information

Based on Rough Sets Theory
Yunjun Zhan; Guangdao Hu; Yanyan Wu;

Computational Intelligence and Industrial Application, 2008. PACIIA '08. Pacific-Asia Workshop on Volume: 1 Digital Object Identifier: 10.1109/PACIIA.2008.35 Publication Year: 2008 , Page(s): 282 - 286

Abstract Space borne imaging spectrometry provides spectral information which can be used to extract parameters of interest in monitoring regional environmental remediation initiatives. The environment surrounding the City of Greater Sudbury in Ontario Canada has been affected by deposition of sulphur-dioxide and metals from a century of smelting operations. Impacts have been noted up to 100 km downwind of the smelter sites, with areas of barren rock replacing natural forested cover. Regional remediation measures have been undertaken in the area since the late-1970s, involving a reduction of smelter emissions and a re-vegetation initiative. Space borne hyperspectral acquisitions from EO-1 Hyperion are evaluated for providing regional information on land cover change and relative vegetation health in support of remediation monitoring measures.

Spectral mapping capabilities of sedimentary rocks using hyperspectral data in Sicily, Italy
Villa, P.; Pepe, M.; Boschetti, M.; de Paulis, R.; Geoscience and Remote Sensing Symposium (IGARSS), 2011 IEEE International Digital Object Identifier: 10.1109/IGARSS.2011.6049741 Publication Year: 2011 , Page(s): 2625 - 2628 Abstract Geologic applications of remote sensing data often rely on ancillary and support information for effectively mapping geolithologies. This work aims to investigate the capabilities of mapping geologic outcrops of sedimentary rocks using only spectral information coming from aerial hyperspectral data. MI VIS hyperspectral data obtained in the area of Serra di Falco, in southern Italy, have been exploited for testing and comparing geologic maps resulting from 4 different spectral supervised algorithms (SVM, SAM, SID, MAXLIKE) in combination with 4 different methods of selecting the training sample for feeding the classifiers, and making use of various ancillary data (ground surveys, geological map) only as reference information for evaluation the results. Geologic mapping results comparison shows the pros and cons of spectral classification in a complex sedimentary geology context.

Applying

boosting

for

hyperspectral

classification of ore-bearing rocks
Monteiro, S.T.; Murphy, R.J.; Ramos, F.; Nieto, J.; Machine Learning for Signal Processing, 2009. MLSP 2009. IEEE International Workshop on Digital Object Identifier: 10.1109/MLSP.2009.5306219 Publication Year: 2009 , Page(s): 1 - 6 Cited by: 1 Abstract Hyperspectral sensors provide a powerful tool for nondestructive analysis of rocks. While classification of spectrally distinct materials can be performed by traditional methods, identification of different rock types or grades composed of similar materials remains a challenge because spectra are in many cases similar. In this paper, we investigate the application of boosting algorithms to classify hyperspectral data of ore rock samples into multiple discrete categories. Two variants of boosting, GentleBoost and LogitBoost, were implemented and compared with support vector machines as benchmark. Two pre-processing transformations that may improve classification accuracy were investigated: derivative analysis and smoothing, both calculated by the Savitzky-Golay method. To assess the performance of the algorithms over noisy data, white Gaussian noise was added at various levels to the data set. We present experimental results using hyperspectral data collected from rock samples from an iron ore mine.

Synthesis

of

high-spatial

resolution

hyperspectral VNIR/SWIR and TIR image data for mapping weathering and alteration minerals in Virginia City, Nevada
Vaughan, R.G.; Calvin, W.M.; Geoscience and Remote Sensing Symposium, 2004. IGARSS '04. Proceedings. 2004 IEEE International Volume: 2 Digital Object Identifier: 10.1109/IGARSS.2004.1368654 Publication Year: 2004 , Page(s): 1296 - 1299 vol.2 Abstract The Comstock mining district, around Virginia City, Nevada, consists of mostly Miocene volcanic

rocks that have been subjected to multiple episodes of hydrothermal alteration, extensional faulting, and mineralization. The distribution of alteration zones is related to the occurrence of precious metal deposits. Also, as a result of past mining activity, tailings that are now distributed throughout the town site contain-abundant Fe-sulfide minerals that weather to form secondary Fe-minerals often linked to acid mine drainage. To support research into both precious metal exploration, and environmental site characterization high spatial resolution (?2 m) hyperspectral VNIR/SWIR and TIR images were acquired over Virginia City using the newly-developed airborne imaging spectrometers HyperSpecTIR and SEBASS. Spectral reflectance data from HyperSpecTIR and spectral emissivity data from SEBASS were used to generate maps of important weathering and alteration minerals. Acid-sulfate alteration minerals were found to be zoned locally and weakly constrained by structures. Jarosite was found to be relatively abundant on mine tailings along with minor occurrences of hydrous Mg-Ca-Al sulfates, indicative of pH conditions around 3-5. Finally, minerals that could not be identified uniquely with one spectral range could be identified using a combination of both VNIR/SWIR and TIR hyperspectral image data.

Hyperspectral alteration

mapping with

for

rock

and Angle

mineral

Spectral

Mapping and Neural Network classification method: Study case in Warmbad district, south of Namibia
Arvelyna, Y.; Shuichi, M.; Atsushi, M.; Nguno, A.; Mhopjeni, K.; Muyongo, A.; Sibeso, M.; Muvangua, E.; Geoscience and Remote Sensing Symposium (IGARSS), 2011 IEEE International Digital Object Identifier: 10.1109/IGARSS.2011.6049458 Publication Year: 2011 , Page(s): 1752 - 1754 Abstract This study observed rock and alteration mineral mapping using HyMap hyperspectral data over the Warmbad district, south of Namibia. The classification result using proposed methods consist of PPI derived from the Spectral Hourglass method and field spectra with the Neural Network classification method show good result for mapping various sericites in pegmatite veins, propylitic alteration of chlorite and epidote, and amphibole in mafic and ultra mafic rock of gabbro, gabbro norite, which are mostly observed in the Warmbad district.

Spectral feature extraction and analysis based on hyperspectral remote sensing data
Yi Baolin; Xu Chenwei; Li Weiwei; Intelligent Computing and Intelligent Systems (ICIS), 2010 IEEE International Conference on Volume: 3 Digital Object Identifier: 10.1109/ICICISYS.2010.5658558 Publication Year: 2010 , Page(s): 63 - 66 Abstract Hyperspectral data have the characteristics of massive data and high dimensions, which lead lots of difficulties for spectrum analyzing and processing. Especially spectral feature extraction and analysis are the foundation of further processing. In this paper, we discuss a number of spectral data processing algorithms for spectral feature extraction and analysis. The main contribution is that we combine both Douglas-Pcucker (DP) algorithm and spectrum derivative algorithm to extract spectral absorption characteristics from the rock and mineral spectral data. The experiments indicate effectiveness while using the proposed algorithms for spectral feature extraction.


相关文章:
更多相关标签: