The sub-pixel level mineral information with reasonable accuracy could be a valuable guide to geological and exploration community for expensive ground and/or lab experiments to discover economic deposits. Thus, several studies demonstrate the feasibility of hyperspectral images for sub-pixel mineral mapping.
Target detection is one of the most important applications in hyperspectral remote sensing image analysis. Relative to multispectral sensing, hyperspectral sensing can increase the detectability of pixel and subpixel size targets by exploiting finer detail in the spectral signatures of targets and natural backgrounds. Over the past several years, different algorithms for the detection subpixel targets with known spectral signature have been developed.
Using derivative spectra in analytical chemistry is an established technique which is renowned as derivative spectroscopy. On the other hand, hyperspectral data nature allows the implementation of derivative spectroscopy, in hyperspectral images. Application of this technique in hyperspectral remote sensing is due to its ability for resolving complex spectra of several target species within individual pixels. Several researches have shown that the application of derivative spectrum in target detection, prepare better detection results in each individual pixel.
However, one of the points that we have to pay special attention is that spectrum differentiation eliminates low frequency components of the spectrum. Hence, having a spectra and its best derivative order in a unified approach cause an increase in the information content of a spectral curve. Thus, an ensemble approach was proposed in this research.
To justify the above-mentioned framework, this study was carried out with an airborne hyperspectral data in comparison to the previous works whereby the standard laboratory images were applied. Both proposed detection algorithms were used for identification of four mineral targets including alunite, kaolinite, epidote and hematite which were located in a hydrothermally altered mineral region (Gonabad county) in Iran east (Lat. and Lon. ). The Hymap images were acquired on September 11th 2006 with a spatial resolution of .
Gonabad is structurally a part of Iran’s central desert (Lūt Desert). The regional rock units show the presence of altered mineralized rocks consisting of volcanic and subvolcanic tracit, agglomera, tuff, riolite, riodacite, dacite, etc. These mass rocks were altered due to the effect of hydrothermal solutions. The importance of geological researches in this region comes from above-mentioned hydrothermally altered zones.
Ground sample data were collected for four dominant targets in the study area, including alunite, kaolinite, hematite and epidote. Stratified random sampling strategy was used for collecting the minerals in the field based on geological experts’ opinion. The precise target’s coordinates were determined by using GPS sensors.
After the completion of field sampling procedure, samples were transformed to laboratory. Samples were grained in lab for spectral measurements. Then, reflectance measurements were done using a SVC HR 1024 lab spectrometer covering a full spectral range of VIS/SWIR (350 to 2500 nm). After spectral measurements were recorded, all obtained spectra were resampled to Hymap spectral response function. Finally, the resampled spectra was applied in the proposed ensemble approach for subpixel mapping of above-mentioned mineral targets.
Experimental results show that the proposed method had clearly better detection results in entire understudy samples. The best performance upgrade was about 28, 24, 26 and 16 percent respectively for Alunite, Kaolinite, Hematite and Epidote targets.