- Geochemistry and Geologic Mapping
- Rough Sets and Fuzzy Logic
- Data Management and Algorithms
- Remote Sensing and Land Use
- Remote-Sensing Image Classification
- Geographic Information Systems Studies
- Advanced Computational Techniques and Applications
- Data Mining Algorithms and Applications
- Image and Signal Denoising Methods
- Customer churn and segmentation
- Sparse and Compressive Sensing Techniques
- earthquake and tectonic studies
- Literature, Language, and Rhetoric Studies
- Geological Modeling and Analysis
- Biopolymer Synthesis and Applications
- Geology and Paleoclimatology Research
- Geological and Geochemical Analysis
- Hydrocarbon exploration and reservoir analysis
- Effects and risks of endocrine disrupting chemicals
- Advanced Image Fusion Techniques
- Image Retrieval and Classification Techniques
- Automated Road and Building Extraction
- Advanced Image Processing Techniques
- Advanced Decision-Making Techniques
China University of Petroleum, Beijing
2008-2021
Abstract Variations of sedimentation rate within a basin over geologic time are series that can be filtered into several cyclic wave curves. Based on back‐stripping and the empirical mode decomposition method, cores from 14 wells in Tarim Basin were selected to do filtering analysis. Four cycles or quasi‐cycles (33 Ma, 64.4 103.6 224 Ma) obtained. Among these, 33 Ma period, which was related internal earth activity, an external force, combination two, most obvious. The period corresponded...
Effective fusion of multiple data, including geographical, geological, geophysical, geochemical and dynamic data for hydrocarbon potential mapping, involves both a algorithm convenient modelling platform. In this study, fuzzy logic geographical information system (GIS) are used to fuse geological geophysical interpretations in mapping the gas Kazakhstan Marsel Territory Carboniferous based on assumed gas-accumulation model. Non-linear membership functions transform input while gamma operator...
Association rule is one of the most important rules in nature. Each type object a remotely sensed image relates to special association rules, thus are features for classification, and mining rational selection effective key issues accurate classification. In this paper, an approach that integrates analysis decision tree presented applied object-oriented high resolution The adopted strong from image, finding optimal A Geoeye-1 used experimental data. Firstly, segmented, then spatial,...
Symbols are considered as the language of a map; hence, accurate understanding meaning symbols is crucial when obtaining geographical information from map: symbolisation spatial data key importance in cartography. A system (GIS) provides convenient mapping platform and powerful functions for symbolisation, while presence various standards impedes maps sharing map information. On other hand, available GIS platforms find it difficult to deal with automatic conversion between different...
Most of the hyperspectral image enhancement methods,which are applied to multispectral processing,have not fully used spectral information data.The unmixing based method is limited by endmember selecting.A fractal signature approach presented in this paper.According modified blanket method,the upper and lower curve corresponding original each pixel can be computed,and formed at scale for enhancement.Three scenes Hyperion experimented work,the images computed.It seen from curves that ground...
Traditional methods effectively used for processing multispectral data is usually limited to deal with hyperspectral images, which are characterized by massive and higher spectral resolution. In this paper, multifractal introduced mining analyzing the holistic feature of curve, blanket method adopted compute fractal dimension (fractal signature) each curve within Hyperion image. The experimental result shows advantage signature over original identify objects, down may be more effective...
The original purpose of association rules mining aims at the analysis customer's purchasing behavior. Practically, customers can be classified into different classes, and each class may show behavior which corresponds to rules. Therefore, corresponding specified considered as features for classification. In this paper, a new idea is proposed that adopted select classification interpret well logging data. Apriori algorithm introduced from preprocessed A frequent 8-term set acquired, two...
Hyperspectral remote sensing provides the possibility of direct detection material information; however, coarse spatial resolution can restrict scope its application. The super-resolution (SR) technique overcome this problem, but separate application SR reconstruction to each spectral band is computationally intensive. We proposed an approach that combines empirical mode decomposition (EMD), single-image using compressed (CS), and principal component analysis (PCA). EMD was used extract...
Road network is the important part of urban map database, and generalization road an issue for city database management. The strategy selection maintenance spatial pattern are two major challenges in networks. An effective process introduced this paper. first step to extract strokes from evaluate every stroke using a grading evaluation method. It helps maintain structure define level importance each stroke. Then, selection, random sampling applied order distribution characteristics network,...
A multifractal based approach to extracting road information is presented in this paper. As a case study, Geoeye-1 image used for the experiment. The result compared with results from Sobel operator, Roberts Prewitt operator. It shown that proposed method can extract different types of more effectively.
Abstract The vertical migration of hydrocarbons from oil and gas reservoirs to the surface and/or subsurface (i.e., microseepage) can induce mineral alteration that results in concentration ferrous iron, clay, carbonate minerals at Earth’s surface. These altered zones may be detected remote sensing imagery based on their specific reflectance spectral characteristics; where three types occur simultaneously reflect possible hydrocarbon microseepage locations with potential for underlying...