Hydrothermal alteration mapping is of great significance for porphyry copper deposit (PCD) exploration. Due to different metallogenetic environments, hydrothermal processes are heterogeneous especially in large-scale ore districts, leading to diverse physicochemical properties and spectral variations of wall-rock alteration. Conventional mapping methods mainly rely on shallow spectral features to identify hydrothermal alteration and cannot fully characterize the properties of the materials. By contrast, deep learning (DL) frameworks with multiple nonlinear layers can extract deep spectral and spatial features from hyperspectral imagery which is more robust and discriminative. Deep models were employed to the GaoFen-5 (GF-5) hyperspectral dataset in the Duolong district, to investigate the role of deep features in hydrothermal alteration mapping. For this purpose, we assess stacked autoencoder and several convolution neural networks which are computationally affordable for achieving large-area mineral exploration. More importantly, the mixed convolutions and covariance pooling algorithm achieved the best mapping results with the overall accuracy of 96.77%. The feature fusion method is also recommended, because of its lightweight structures. These two spectral–spatial classifiers produced appealing classification performance while reducing the misclassification among spectrally similar minerals and alleviating noise, demonstrating their powerful learning capabilities. Finally, spatial patterns of the hydrothermal alteration in the Duolong district can be taken as an important indicator of erosion degree of the porphyry-epithermal system. It is proved that the combination of the DL methods and the GF-5 hyperspectral data is effective in large-area mineral exploration and can be applied to other PCDs in the neighboring Bangong Co-Nujiang metallogenic belt.
The fusion of a high-spatial-resolution (HSR) panchromatic band and several multispectral bands with a relative low spatial resolution has become a research focus with the development of HSR remote sensing technology. Previous studies have demonstrated that fused spectra of mixed pixels (MPs) remain mixed, which considerably contributes to spectral distortions observed in fused images produced by most of the current pansharpening methods. Several works have attempted to reduce spectral distortions of fused spectra of MPs to improve the quality of fused products generated by some fusion methods based on component substitution (CS). An image fusion framework for reducing spectral distortions caused by the incorrect fusion of MPs is proposed for both CS and fusion methods based on multiresolution analysis (MRA). Using the proposed framework based on image segmentation, the fused products of two classic MRA-based pansharpening methods were improved by improving the fusion spectra of MPs. The improved fused images were compared with the original fusion products through a fusion experiment using three datasets recorded by WorldView-2, GeoEye-1, and WorldView-3. Experimental results showed that the improved fused products yielded higher Q2n and quality with no reference values and lower relative average spectral error, dimensionless global relative error of synthesis, and spectral angle mapper values than the corresponding original fusion products. This indicates that the proposed image fusion framework is effective for reducing spectral distortions of fused images generated by the two MRA-based fusion methods.
With the development of Hyperspectra and the method of rock-mineral information extraction, several cores were
analyzed based on analytical spectral devices (ASD) and rock-mineral information extraction in Wushan-cooper deposit
area. Aiming at the low accuracy of mineral identification with hyperspectral data, the present study established regional
spectra library on the basis of the study area geological background, section noise filtering and fast Fourier transform
processing methods. Using the rapid quantificational identification model, the rock-mineral alternation information was
extracted to build core profile and 3D model to discuss the deep mineralization evaluation.
Combing with the regional metallogenic background, the alteration information indicated that the ore mineral was related
with multiple alteration assemblages and there may be rock mass in deep space. The Cu element contents and ore
mineral were closely related with the skarnization, silicification and chloritization. It also suggested that the deposit was
skarn type in less than 1000 m depth, which was affected by the sandstone. Meanwhile, in more than 1000 m depth, the
deposit was controlled by composite minerallzation types, which was associated with the previous geology and mineral
deposits studies. In summary,this study supported a two stage mineralization model for the Wushan-copper deposit
area,namely,the first stage of synsedimentary hydrothermal exhalative stage and the second stage of magmatichydrothermal
ore-forming stage.
Image segmentation is the basis of object-based information extraction from remote sensing imagery. Image
segmentation based on multiple features, multi-scale, and spatial context is one current research focus. The scale
parameters selected in the segmentation severely impact on the average size of segments obtained by multi-scale
segmentation method, such as the Fractal Network Evolution Approach (FNEA) employed in the eCognition software. It
is important for the FNEA method to select an appropriate scale parameter that causes no neither over- nor undersegmentation.
A method for scale parameter selection and segments refinement is proposed in this paper by modifying a
method proposed by Johnson. In a test on two images, the segmentation maps obtained using the proposed method
contain less under-segmentation and over-segmentation than that generated by the Johnson’s method. It was
demonstrated that the proposed method is effective in scale parameter selection and segment refinement for multi-scale
segmentation algorithms, such as the FNEA method.
The aim of data conflation is to synergise geospatial information from different sources into a common framework,
which can be realised using multivariate geostatistics. Recently, multiple-point geostatistics (MPG) has been proposed
for data conflation. Instead of the variogram, MPG borrows structures from the training image, so the spatial correlation
is characterised by multiple-point statistics. In pattern-based MPG, two sets of data can be integrated by utilising the
secondary data as a locally varying mean (LVM). The training image provides a spatial correlation model and is
incorporated to facilitate reproduction of similar local patterns in the predicted image. However, the current patternbased
MPG gathers similar patterns based on a prototype class, which extracts spatial structures in an arbitrary way. In
this paper, we proposed an improved pattern-based MPG for conflation of digital elevation models (DEMs). In this
approach, a new strategy for forming prototype class is applied, which is based on the residual surface, vector
ruggedness measure (VRM) and ridge valley class (RVC) of terrain data. The method was tested on the SRTM and
GMTED2010 data. SRTM data at the spatial resolution of 3 arc-second was simulated by conflating sparse elevation
point data and GMTED2010 data at a coarser spatial resolution of 7.5 arc-second. The proposed MPG method was
compared with the traditional pattern-based MPG simulation. Several kriging predictors were applied to provide LVMs
for MPG simulation. The result shows that the new method can achieve more precise prediction and retain more spatial
details than the benchmarks.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.