CN101739396A - Uncertainty spatial data mining-based regional metallogenic prediction method - Google Patents
Uncertainty spatial data mining-based regional metallogenic prediction method Download PDFInfo
- Publication number
- CN101739396A CN101739396A CN200810046559A CN200810046559A CN101739396A CN 101739396 A CN101739396 A CN 101739396A CN 200810046559 A CN200810046559 A CN 200810046559A CN 200810046559 A CN200810046559 A CN 200810046559A CN 101739396 A CN101739396 A CN 101739396A
- Authority
- CN
- China
- Prior art keywords
- data
- spatial data
- metallogenic
- regional
- information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention relates to an uncertainty spatial data mining-based regional metallogenic prediction method, and belongs to the technical field of resource information processing and application. The regional metallogenic prediction method makes full use of multi-source mass geological space data (basic geological mineral data, remote sensing prospecting data, geochemical data and mineral deposit data) and effectively extracts regional metallogenic associated information to perform rapid and effective regional metallogenic prediction based on an uncertainty spatial data mining algorithmic model. The regional metallogenic prediction method comprises three main steps of geometrically registering multi-source and multi-scale geological space data, extracting remote sensing mineralizing information and performing metallogenic prediction based on the uncertainty spatial data mining algorithmic model. The method can evaluate a regional metallogenic prospective area more rapidly and objectively.
Description
Technical field
The present invention relates to a kind of new regional metallogenic prognosis method, belong to the resource information technical field.
Background technology
Metallogenic prognosis method in zone has experienced from " geologic anomaly " in early stage mineral deposit statistical forecast, mid-term and has predicted nearest " tri-coupling type " quantitatively metallogenic prognosis and the non-linear metallogenic prognosis theory of multifractal and method.These regional metallogenic prognosis theories, technology and method reach its maturity, and have brought into play vital role in regional metallogenic prognosis.But, along with zone and the collection day by day of global geological spatial data with obtain, in the face of the geological spatial data of magnanimity multi-source, the very difficult geological spatial data of fast and effeciently handling magnanimity of traditional regional metallogenic prognosis method.The Spatial Data Mining Technique that last century, the nineties was risen provides good theoretical method for the massive spatial data information extraction.The applicant has proposed uncertain Spatial Data Mining algorithm model according to ubiquitous uncertainty of spatial data and spatial autocorrelation.Simultaneously, how to make good use of existing magnanimity multi-source geological spatial data fast, efficiently carry out the zone and become the ore deposit quantitative forecast, seeming has scientific meaning and economic worth.
Summary of the invention
The object of the invention is the deficiency at existing regional metallogenic prognosis method, proposes a kind ofly from one-tenth ore deposit, magnanimity multi-source geological spatial data rapid extraction zone information, reaches the method for carrying out regional metallogenic prognosis efficiently.On the uncertainty of taking geological spatial data into account and the probabilistic basis that becomes the ore deposit information extraction, proposition is based on the regional metallogenic prognosis method of uncertain Spatial Data Mining algorithm model, and sets up relevant evaluating index and come objective evaluation region metallogenic prognosis result.Below be particular content of the present invention:
1, the multiple dimensioned geological spatial data pre-service of multi-source
According to four kinds of different scales (zone becomes feasible location, ore deposit, zone to become ore deposit beneficial zone, zone to become potential location, ore deposit, zone to become distant view location, ore deposit) metallogenic prognosis requirement, carry out the digitizing of geology key element and the coordinate projection conversion (Gauss-Ke Lvge Beijing 1954 plane coordinate systems that adopted 6 degree to be with in 1: 50 ten thousand, 1: 20 ten thousand and 1: 5 ten thousand, Gauss-Ke Lvge Beijing 1954 plane coordinate systems of employing in 1: 13 degree bands) of multi-source geological spatial data by four kinds of different proportion chis (1: 50 ten thousand, 1: 20 ten thousand, 1: 5 ten thousand and 1: 1 ten thousand).Projection coordinate's conversion using of remotely-sensed data is set up corresponding reference mark with the topomap in zone, adopts polynomial fitting method to carry out the coordinate registration.
2, remote sensing mineralising information extraction
Complicacy according to atural object object in the zone, different atural objects are carried out on the different masking method bases, adopt the unusual dividing method of feature principal component analysis (PCA) and threshold values to extract iron ore information and argillization information, (is example with LandSAT ETM multispectral data) specific as follows:
(1) masking method is selected (is example with the northwest high and cold mountain area), for Yun Hexue, adopts the high value district of TM1 (DN value 140~165) as threshold values; For water body, adopt TM7/TM1 (0.3~0.4) as threshold values; For vegetation, adopt TM4/TM3 (1.05~1.2) as threshold values; For desert and water body, directly carry out mask by visual delineation region of interest.
(2) the feature principal component analytical method is specific as follows: select TM1, TM3, TM4 and TM5 to obtain iron ore information as principal component analysis, in 4 PC images that obtain, select the image that TM3 and TM1 proper vector load factor absolute value are big and PC image opposite in sign extracts as the iron ore abnormal information; Select TM1, TM4, TM5 and TM7 to obtain argillization information as principal component analysis, in 4 PC images that obtain, the image that TM5 and TM7 proper vector load factor absolute value are big and both opposite in sign PC images extract as the argillization abnormal information.
(3) the unusual dividing method of threshold values is specific as follows: at first the mineralising information of extracting is carried out 3 * 3 or 5 * 5 mean filter, as threshold values, limit abnormal level, to carrying out threshold segmentation unusually with 2.5~3.0 times of standard deviation σ.
3, based on the metallogenic prognosis of uncertain Spatial Data Mining algorithm model
(1), adopt Monte Carlo simulation to carry out the spatial data uncertainty model according to the uncertain type of different geological spatial datas:
1. determine that each treats input space data set D
iThe uncertain type of () (the circular normal model of position data, the one dimension normal model of attribute data);
2. take according to space data sets D
iThe stochastic sampling that () distributes replaces former input space data;
3. to realizing each time, store its Y () as a result;
4. randomly draw 10000 groups of above experimental datas as sample data.
(2) based on EM (Expectation Maximization) algorithm, consider the spatial autocorrelation of geological spatial data, set up on the basis of spatial autocorrelation matrix, adopt neighborhood EM algorithm that geology space attribute data carry out fuzzy discreteizations, specific as follows:
1. the spatial autocorrelation matrix is set up: use Voronoi and Delaunay figure, make up the space weight matrix in conjunction with criterion distance.Consider the uncertainty of spatial data, adopt spatial autocorrelation matrix between three kinds of method computer memory data: center method, minimum method and maximum method.Suppose to have among the regional s n the uncertain point in position, i some P
iError band with a circular Q
iExpression.Specific algorithm is as follows:
Input: the error band Q={Q of one group of point among the regional s
1, Q
2... Q
nAnd neighborhood apart from d
Output: one group of neighborhood of a point figure and spatial autocorrelation matrix among the regional s
Step 1: the Voronoi polygon of structure point set P
Step 2: all adjacent Voronoi polygons are carried out following computing:
Step 2.1: calculate d
Center(C
i, C
j), d
Max(Q
i, Q
j), calculate d
Min(Q
i, Q
j)
Step 2.2: if d<d
n, then in neighborhood figure, connect P
iAnd P
j, w
IjBe 1; Otherwise w
IjBe 0 wherein, d is the neighborhood distance; d
Center(C
i, C
j) represent adjacent error band (Q
i, Q
j) distance between the barycenter; d
Max(Q
i, Q
j) represent adjacent error band (Q
i, Q
j) ultimate range between the interior spatial data; d
Min(Q
i, Q
j) represent adjacent error band (Q
i, Q
j) minor increment between the interior spatial data.
2. carry out fuzzy discreteization based on the geological space attribute data of neighborhood EM algorithm, its algorithm is as follows:
According to the geographic coordinate and the uncertainty models of geological spatial data, according to 1. algorithm can set up the discrete space data spatial neighbor relational matrix V:
In order to consider the spatial autocorrelation of spatial data, improve function:
Introduce new item:
Then, set up new canonical function and be: U (c, θ)=D (c, θ)+β G (c) (β 〉=0)
Wherein, β is the homogeneous parameter in space of control space data sets.
Then, E-step:
Its optimization necessary condition is following form:
Obtain following equation at last:
The M step:
(3) become ore deposit related information excavation and uncertain the evaluation
1. take probabilistic one-tenth ore deposit association rule mining into account: the uncertain MonteCarlo simulation → spatial autocorrelation of geological spatial data matrix structure → continuous type data UNEM method discretize → Apriori space correlation Rule Extraction
2. become the uncertainty evaluation of ore deposit correlation rule, adopt support (support), confidence level (confidence) and three evaluation indexes of interest-degree (interesting) to be carried out to the uncertainty evaluation of ore deposit information of forecasting
Description of drawings
Fig. 1: area, the Kunlun, east, Qinghai becomes ore deposit association rule mining result's uncertain evaluation index (support) figure
Fig. 2: area, the Kunlun, east, Qinghai becomes ore deposit association rule mining result's uncertain evaluation index (confidence level) figure
Fig. 3: area, the Kunlun, east, Qinghai becomes ore deposit association rule mining result's uncertain evaluation index (interest-degree) figure
Embodiment
In order to understand technical scheme of the present invention better, below be example with western part of China important meals metallogenic belt-the Kunlun, east, Qinghai, specific embodiment is provided.
At first, digitizing multi-source geological spatial data comprises geological and mineral data, geochemistry data and mineral deposit data; Extract remote sensing mineralising information (this explanation is with LandSAT ETM data instance) then; Adopt the uncertain Spatial Data Mining algorithm in this invention to be extracted into ore deposit correlation rule information guiding metallogenic prognosis at last, following result is the one-tenth ore deposit association rule mining result (as following table 1) of metallogenic belt, the Kunlun, east, Qinghai iron ore-deposit.
The one-tenth ore deposit association rule mining result and uncertain evaluation of metallogenic belt, the Kunlun, east, table 1 Qinghai iron ore-deposit
Claims (4)
1. based on the regional metallogenic prognosis method of uncertain Spatial Data Mining, it is characterized in that making full use of multi-source magnanimity geological spatial data (ore deposit data, geochemistry data, mining geology data, mineral deposit data are looked in basic geology mineral products data, remote sensing), take the uncertainty and the uncertainty that becomes the ore deposit information extraction of geological spatial data simultaneously into account, the uncertain Spatial Data Mining algorithm model that adopts the applicant to set up, carry out regional metallogenic prognosis quickly and efficiently, comprise three key steps: the standardization of the multiple dimensioned geological spatial data of multi-source; The information extraction of remote sensing mineralising; Metallogenic prognosis based on uncertain Spatial Data Mining algorithm model.
2. the standardization of the multiple dimensioned geological spatial data of the described multi-source of claim 1, it is characterized in that according to different scale metallogenic prognosis requirement, carry out the pre-service and the geometrical registration of multi-source geological spatial data by the different proportion chi, comprise: 1: 50 ten thousand, 1: 20 ten thousand, 1: 5 ten thousand with 10,004 kinds of different ratio data chi requirements in 1: 1, carry out the conversion of the digitizing of geology key element, projection coordinate.
3. the described remote sensing mineralising of claim 1 information extraction, it is characterized in that complicacy according to atural object object in the zone, different atural objects (cloud, snow, water body, vegetation, desert etc.) are carried out on the different masking method bases, adopt the unusual dividing method of feature principal component analysis (PCA) and threshold values to extract iron ore information and argillization information, (is example with LandSat ETM multispectral data) specific as follows:
(a) masking method is selected (is example with the northwest high and cold mountain area), for Yun Hexue, adopts the high value district of TM1 (DN value 140~165) as threshold values; For water body, adopt TM7/TM1 (0.3~0.4) as threshold values; For vegetation, adopt TM4/TM3 (1.05~1.2) as threshold values; For desert and water body, directly carry out mask by visual delineation region of interest;
(b) the feature principal component analytical method is specific as follows: select TM1, TM3, TM4 and TM5 to obtain iron ore information as principal component analysis; Select TM1, TM4, TM5 and TM7 to obtain argillization information as principal component analysis;
(c) the unusual dividing method of threshold values is specific as follows: at first the mineralising information of extracting is carried out mean filter, as threshold values, limit abnormal level, to carrying out threshold segmentation unusually with 2.5~3.0 times of standard deviation σ.
4. the described metallogenic prognosis of claim 1 based on uncertain Spatial Data Mining algorithm model, specific as follows:
(a), adopt Monte Carlo simulation to carry out the spatial data uncertainty model according to the uncertain type of different geological spatial datas;
(b) take the spatial autocorrelation of geological spatial data into account, set up on the basis of spatial autocorrelation matrix, adopt neighborhood EM algorithm that geology space attribute data carry out fuzzy discreteizations;
(c) multi-source magnanimity geological spatial data is carried out to the ore deposit related information and excavates, and adopt the uncertainty evaluation that support (support), confidence level (confidence) and three evaluation indexes of interest-degree (interesting) are carried out to the ore deposit information of forecasting.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN200810046559A CN101739396A (en) | 2008-11-14 | 2008-11-14 | Uncertainty spatial data mining-based regional metallogenic prediction method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN200810046559A CN101739396A (en) | 2008-11-14 | 2008-11-14 | Uncertainty spatial data mining-based regional metallogenic prediction method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN101739396A true CN101739396A (en) | 2010-06-16 |
Family
ID=42462890
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN200810046559A Pending CN101739396A (en) | 2008-11-14 | 2008-11-14 | Uncertainty spatial data mining-based regional metallogenic prediction method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN101739396A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106126882A (en) * | 2016-06-15 | 2016-11-16 | 中国地质大学(北京) | P-block element p geochemistry data method for optimizing is reconnoitred based on what Kendall's concordance coefficient sorted |
CN110096622A (en) * | 2019-05-28 | 2019-08-06 | 成都理工大学 | A kind of multiple dimensioned data Unified Expression method and system |
CN110490061A (en) * | 2019-07-11 | 2019-11-22 | 武汉大学 | A kind of uncertainties model and measure of characteristics of remote sensing image |
CN112632467A (en) * | 2020-12-14 | 2021-04-09 | 电子科技大学 | Forest combustible load calculation method based on cooperative optical and microwave data |
CN112818603A (en) * | 2021-02-06 | 2021-05-18 | 中国科学院新疆生态与地理研究所 | Method, terminal and storage medium for adaptively selecting optimal mineralization prediction elements |
-
2008
- 2008-11-14 CN CN200810046559A patent/CN101739396A/en active Pending
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106126882A (en) * | 2016-06-15 | 2016-11-16 | 中国地质大学(北京) | P-block element p geochemistry data method for optimizing is reconnoitred based on what Kendall's concordance coefficient sorted |
CN106126882B (en) * | 2016-06-15 | 2019-01-29 | 中国地质大学(北京) | Element geochemistry data preferred method based on Kendall coefficient of concordance sequence |
CN110096622A (en) * | 2019-05-28 | 2019-08-06 | 成都理工大学 | A kind of multiple dimensioned data Unified Expression method and system |
CN110096622B (en) * | 2019-05-28 | 2021-07-06 | 成都理工大学 | Multi-scale data unified expression method and system |
CN110490061A (en) * | 2019-07-11 | 2019-11-22 | 武汉大学 | A kind of uncertainties model and measure of characteristics of remote sensing image |
CN110490061B (en) * | 2019-07-11 | 2021-10-22 | 武汉大学 | Uncertainty modeling and measuring method for remote sensing image characteristics |
CN112632467A (en) * | 2020-12-14 | 2021-04-09 | 电子科技大学 | Forest combustible load calculation method based on cooperative optical and microwave data |
CN112632467B (en) * | 2020-12-14 | 2023-03-21 | 电子科技大学 | Forest combustible carrying capacity calculation method based on cooperative optical and microwave data |
CN112818603A (en) * | 2021-02-06 | 2021-05-18 | 中国科学院新疆生态与地理研究所 | Method, terminal and storage medium for adaptively selecting optimal mineralization prediction elements |
CN112818603B (en) * | 2021-02-06 | 2023-11-24 | 中国科学院新疆生态与地理研究所 | Method, terminal and storage medium for adaptively selecting optimal mineral formation prediction element |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Yin et al. | Integrating remote sensing and geospatial big data for urban land use mapping: A review | |
Chen et al. | Landslide susceptibility assessment using evidential belief function, certainty factor and frequency ratio model at Baxie River basin, NW China | |
Jochem et al. | Identifying residential neighbourhood types from settlement points in a machine learning approach | |
Bishop et al. | Validation of digital soil maps at different spatial supports | |
Zheng et al. | Understanding the tourist mobility using GPS: How similar are the tourists? | |
Liu et al. | Simulating urban dynamics in China using a gradient cellular automata model based on S-shaped curve evolution characteristics | |
Li et al. | Spatial data quality and beyond | |
Chen et al. | Modeling growth curve of fractal dimension of urban form of Beijing | |
Fathizad et al. | Evaluating desertification using remote sensing technique and object-oriented classification algorithm in the Iranian central desert | |
Etienne et al. | Trajectory Box Plot: a new pattern to summarize movements | |
CN101739396A (en) | Uncertainty spatial data mining-based regional metallogenic prediction method | |
Xia et al. | Exploring the effects of partitioned transition rules upon urban growth simulation in a megacity region: A comparative study of cellular automata-based models in the Greater Wuhan Area | |
CN102243628A (en) | Mineralizing case reasoning model and method | |
CN102930275A (en) | Remote sensing image feature selection method based on Cramer's V index | |
Wen et al. | Assessment of land ecological security and analysis of influencing factors in Chaohu Lake Basin, China from 1998–2018 | |
CN102184423A (en) | Full-automatic method for precisely extracting regional impervious surface remote sensing information | |
Gan et al. | A new spatial modeling method for 3D formation drillability field using fuzzy c-means clustering and random forest | |
Marzban et al. | Cluster analysis for object-oriented verification of fields: A variation | |
Wu et al. | Object-oriented and deep-learning-based high-resolution mapping from large remote sensing imagery | |
Zhang | Exploration on coal mining-induced rockburst prediction using Internet of things and deep neural network | |
CN105243503A (en) | Coastal zone ecological safety assessment method based on space variables and logistic regression | |
Ji et al. | Multicascaded feature fusion-based deep learning network for local climate zone classification based on the So2Sat LCZ42 benchmark dataset | |
CN102609721B (en) | Remote sensing image clustering method | |
Zheng et al. | Extraction of impervious surface with Landsat based on machine learning in Chengdu urban, China | |
CN104794335A (en) | General multistage space sampling method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
WD01 | Invention patent application deemed withdrawn after publication |
Open date: 20100616 |