CN113625363A

CN113625363A - Mineral exploration method and device for pegmatite-type lithium ore, computer equipment and medium

Info

Publication number: CN113625363A
Application number: CN202110947973.6A
Authority: CN
Inventors: 代晶晶; 姜琪; 王登红; 刘善宝; 王成辉
Original assignee: Institute of Mineral Resources of Chinese Academy of Geological Sciences
Current assignee: Institute of Mineral Resources of Chinese Academy of Geological Sciences
Priority date: 2021-08-18
Filing date: 2021-08-18
Publication date: 2021-11-09
Anticipated expiration: 2041-08-18
Also published as: CN113625363B

Abstract

The embodiment of the invention discloses an ore searching method and device for pegmatite type lithium ore, computer equipment and a medium. One embodiment of the method comprises: determining fault structure information of the target area by using the radar image data and the first remote sensing image data of the target area; determining lithology information of the target area by using the second remote sensing image data of the target area; performing data fusion on the third remote sensing image data and the fourth remote sensing image data of the target area, inputting the data into a trained multi-classification calculation model for recognition to obtain pegmatite lithium ore information of the target area, and determining pegmatite lithium ore distribution density information of the target area according to the pegmatite lithium ore information of the target area; and determining the pegmatite type lithium ore prospecting target area in the target area according to the fault structure information, the lithology information and the pegmatite type lithium ore distribution density information of the target area. The implementation mode can be used for more accurately searching pegmatite type lithium ores and provides a more powerful scientific basis for field ore searching.

Description

Mineral exploration method and device for pegmatite-type lithium ore, computer equipment and medium

Technical Field

The invention relates to the field of mineral resource development. And more particularly, to a pegmatite-type lithium ore prospecting method and apparatus, a computer device, and a medium.

Background

Lithium is an indispensable strategic resource for the development of new industries, can store energy, save energy and produce energy, is highly military and generally civil, and is called as energy metal in the 21 st century. The whole world lithium deposit mainly comprises brine type and pegmatite type, wherein pegmatite type lithium ore is one of important deposit types because of being rich in strategic elements such as Li, and the pegmatite type lithium ore is mainly formed by the enrichment of lithium elements in the pegmatite forming process.

At present, the topography and the topography of some pegmatite type lithium ore areas are complex, so that when the ore finding is carried out based on remote sensing image data, the difficulty coefficient of information extraction and interpretation of pegmatite and other rock blocks is increased, the target area of the ore finding is difficult to accurately define, and great difficulty is brought to the ore finding work.

Disclosure of Invention

The invention aims to provide an ore searching method and device, computer equipment and medium for pegmatite lithium ore, so as to solve at least one of the problems in the prior art.

In order to achieve the purpose, the invention adopts the following technical scheme:

the invention provides an ore searching method for pegmatite type lithium ore, which comprises the following steps:

determining fault structure information of a target area by using radar image data and first remote sensing image data of the target area;

determining lithology information of the target area by using the second remote sensing image data of the target area;

performing data fusion on the third remote sensing image data and the fourth remote sensing image data of the target area, inputting the data into a trained multi-classification calculation model for recognition to obtain pegmatite lithium ore information of the target area, and determining pegmatite lithium ore distribution density information of the target area according to the pegmatite lithium ore information of the target area; and

and determining a pegmatite type lithium ore prospecting target area in the target area according to the fault structure information, the lithology information and the pegmatite type lithium ore distribution density information of the target area.

Optionally, the data fusion of the third remote sensing image data and the fourth remote sensing image data of the target region and then inputting the data into the trained multi-classification calculation model for recognition, and obtaining the pegmatite type lithium mineral information of the target region further includes:

performing data fusion on the third remote sensing image data and the fourth remote sensing image data of the target area to obtain first fusion image data;

calculating three relation values of a reflection peak wave Band and an absorption valley wave Band in the corresponding wave Band of the first fusion image data according to the reflection spectrum curve of the pegmatite lithium ore to obtain three gray image data, wherein the relation values are ((Band)_a+Band_b)/Band_b)，Band_aRepresents a reflection peak Band, Band_bRepresents the absorption valley band;

respectively endowing a red channel gray scale, a green channel gray scale and a blue channel gray scale by using the three gray scale image data to form color synthetic image data; and

and inputting the color synthetic image data into a trained multi-classification calculation model for recognition to obtain pegmatite type lithium ore information of the target area.

performing threshold segmentation: judging whether at least one of the red channel gray scale, the green channel gray scale and the blue channel gray scale of the pixel marked as the pegmatite lithium mine is larger than a preset gray scale threshold value and whether the average value of the red channel gray scale, the green channel gray scale and the blue channel gray scale is larger than the preset gray scale threshold value is true or not in the image data containing the pegmatite lithium mine information output by the multi-classification calculation model, if so, keeping the mark of the pixel, and otherwise, deleting the mark of the pixel.

after the thresholding: and marking the pixel between the two pixel regions marked as the pegmatite lithium ore with the pixel distance smaller than the preset pixel distance threshold value as the pegmatite lithium ore by utilizing the parallel operation, and/or deleting the mark of each pixel in the pixel region marked as the pegmatite lithium ore with the pixel area smaller than the preset pixel area threshold value.

Optionally, the determining the tomographic information of the target region by using the radar image data and the first remote sensing image data of the target region further comprises:

performing data fusion on the radar image data of the target area and the first remote sensing image data to obtain second fused image data;

performing Prewitt edge detection on the fourth remote sensing image data to obtain edge detection image data; and

and determining fault structure information of the target area according to the radar image data, the second fusion image data and the edge detection image data.

Optionally, the determining the pegmatite-type lithium ore exploration target region in the target region according to the fault structure information, the lithological information, and the pegmatite-type lithium ore distribution density information of the target region further includes:

and displaying a laminated graph of a first image containing fault structure information of the target area, a second image containing lithology information of the target area and a third image containing pegmatite lithium ore distribution density information of the target area, and displaying the identification of the pegmatite lithium ore exploration target area in the target area on the laminated graph.

Optionally, the third remote sensing image data is WorldView-2 remote sensing image data, and the fourth remote sensing image data is WorldView-3 remote sensing image data.

In a second aspect, the present invention provides an ore prospecting device for pegmatite-type lithium ore, comprising:

the first determining module is used for determining fault structure information of the target area by using radar image data and first remote sensing image data of the target area;

the second determination module is used for determining lithology information of the target area by using second remote sensing image data of the target area;

the third determining module is used for performing data fusion on third remote sensing image data and fourth remote sensing image data of the target area, inputting the data fusion into a trained multi-classification calculation model for recognition to obtain pegmatite type lithium ore information of the target area, and determining pegmatite type lithium ore distribution density information of the target area according to the pegmatite type lithium ore information of the target area;

and the fourth determination module is used for determining the pegmatite type lithium ore exploration target area in the target area according to the fault structure information, the lithological information and the pegmatite type lithium ore distribution density information of the target area.

A third aspect of the present invention provides a computer apparatus comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor executes the program to implement the pegmatite-type lithium mine prospecting method provided by the first aspect of the present invention.

A fourth aspect of the present invention provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the mineral exploration method for pegmatite-type lithium minerals provided by the first aspect of the present invention.

The invention has the following beneficial effects:

the technical scheme of the invention can more accurately search the pegmatite type lithium ore, overcomes the problem that pegmatite lithotropic information is difficult to extract and interpret due to complex topography and the like, improves the working efficiency, can more effectively provide scientific basis for field prospecting, and has important significance for breakthrough of pegmatite type lithium ore prospecting.

Drawings

The following describes embodiments of the present invention in further detail with reference to the accompanying drawings.

FIG. 1 illustrates an exemplary system architecture diagram in which an embodiment of the present invention may be applied.

Fig. 2 is a flowchart illustrating an ore prospecting method for pegmatite-type lithium ore according to an embodiment of the present invention.

Fig. 3 shows another flowchart of the pegmatite-type lithium ore prospecting method according to the embodiment of the invention.

Fig. 4 is a schematic display diagram of fused image data of the radar image data and the first remote sensing image data.

Fig. 5 is a schematic diagram illustrating the display of the edge detection image data.

Fig. 6 is a schematic view showing display of tomographic information of a target region.

Fig. 7 shows a schematic of spectral information for the three end members screened.

FIG. 8 shows a schematic representation of the display of end-member abundance synthesis maps.

FIG. 9 illustrates a display diagram of a lithology unit interpretation graph.

Fig. 10 is a schematic diagram showing a feature value map in the principal component analysis result.

Fig. 11 is a schematic view showing the display of pegmatite-type lithium mineral information based on the fused image data.

Fig. 12 shows a schematic representation of a reflection spectrum curve of spodumene-containing pegmatite.

Fig. 13 is a schematic diagram showing a display of color synthesized image data.

Fig. 14 is a schematic view showing the display of pegmatite-type lithium mineral information based on color synthetic image data.

Fig. 15 shows a schematic diagram of a pegmatite-type lithium ore distribution density map.

Fig. 16 shows a schematic view of a stack diagram.

Fig. 17 shows a schematic diagram of the result of the delineated target area of the mineral exploration.

Fig. 18 is a schematic view of an ore searching apparatus for pegmatite-type lithium ore according to an embodiment of the present invention.

Fig. 19 is a schematic structural diagram of a computer system for implementing the pegmatite-type lithium ore prospecting device according to the embodiment of the invention.

Detailed Description

In order to more clearly illustrate the present invention, the present invention will be further described with reference to the following examples and the accompanying drawings. Similar parts in the figures are denoted by the same reference numerals. It is to be understood by persons skilled in the art that the following detailed description is illustrative and not restrictive, and is not to be taken as limiting the scope of the invention.

At present, the topography and the topography of some pegmatite type lithium ore areas are complex, so that when the existing scheme for finding ores based on remote sensing image data is adopted, the difficulty coefficient for extracting and interpreting information of pegmatite and other rock blocks is increased, an ore finding target area is difficult to accurately define, and great difficulty is brought to ore finding work.

In view of this, an embodiment of the present invention provides an ore searching method for pegmatite-type lithium ores, which extracts various information from a plurality of angles by integrating a plurality of image data based on radar image data and multi-source remote sensing image data to determine an ore searching target area, and the ore searching method provided by this embodiment includes the following steps:

The pegmatite type lithium ore prospecting method based on radar image data and multi-source remote sensing image data provided by the embodiment integrates multiple image data from multiple angles to extract multiple information, can more accurately find pegmatite type lithium ore, overcomes the problem that pegmatite travertine information is not easy to extract and interpret due to complex topography and the like, improves the working efficiency, can more powerfully provide scientific basis for field prospecting, and has important significance for breakthrough of pegmatite type lithium ore prospecting.

The pegmatite-type lithium mine exploration method provided in this embodiment may be implemented by a Computer device with data processing capability, specifically, the Computer device may be a Computer with data processing capability, including a Personal Computer (PC), a mini-Computer or a mainframe, or may be a server or a server cluster with data processing capability, and the present embodiment is not limited thereto.

In order to facilitate understanding of the technical solution of the present embodiment, a scene of the method provided by the present embodiment in practice is described below with reference to fig. 1. Referring to fig. 1, the scenario includes a fault construction information interpretation server 110, a lithology information interpretation server 120, a training server 130, a recognition server 140, and an ore-finding target delineation server 150. For example, in the present embodiment, the training server 130 first trains the multi-class calculation model for recognizing pegmatite-type lithium mineral information in the remote sensing image data of the target area by using the image data sample with pegmatite-type lithium mineral labeled pixels (that is, pixels at pegmatite-type lithium mineral positions in the image data sample are labeled or labeled), so as to obtain the trained multi-class calculation model with learned recognition capability. Subsequently, what can be implemented in parallel is: (1) the tomographic information interpretation server 110 determines tomographic information of the target area using the radar image data and the first remote sensing image data of the target area; (2) the lithology information interpretation server 120 determines lithology information of the target area by using the second remote sensing image data of the target area; (3) the recognition server 140 performs data fusion on the third remote sensing image data and the fourth remote sensing image data of the target area, inputs the data into the multi-classification calculation model trained in the training server 130 for recognition, obtains pegmatite-type lithium mineral information of the target area, and determines pegmatite-type lithium mineral distribution density information of the target area according to the pegmatite-type lithium mineral information of the target area; thus, the overall working efficiency can be improved. Finally, the target area delineating server 150 determines the pegmatite type lithium ore target area in the target area according to the fault structure information, the lithology information and the pegmatite type lithium ore distribution density information of the target area. The function of the training server 130 is implemented only before the function of the recognition server 140 is implemented, and the timing relationship between the function of the training server 130 and the function of the fault structure information interpretation server 110 and the lithology information interpretation server 120 is not limited.

It should be noted that the fault structure information interpretation server 110, the lithology information interpretation server 120, the training server 130, the recognition server 140 and the target area finding definition server 150 in fig. 1 may be separate servers or integrated servers in practical applications. When stand-alone servers, the servers may communicate over a network that may include various types of connections, such as wired, wireless communication links, or fiber optic cables.

Next, from the perspective of a processing device having data processing capability, the pegmatite-type lithium mine exploration method provided by the present embodiment is described, and as shown in fig. 2, the pegmatite-type lithium mine exploration method provided by the present embodiment includes the following steps:

s210, determining fault structure information of the target area by using the radar image data and the first remote sensing image data of the target area.

In a specific example, the target area may be referred to as a research area, the radar image data is, for example, Sentinel-1 radar image data, the first remote sensing image data is, for example, Landsat-8 remote sensing image data, and the Aster remote sensing image data may also be used as the first remote sensing image data. Wherein: the Sentinel 1 (Sentinel-1) satellite is an earth observation satellite in the European space agency Colbeney program (GMES), consists of two satellites, carries a C-band synthetic aperture radar and can provide continuous images (day, night and various weather). The Landsat-8 satellite carries an OLI and TIRS push-broom imager. The OLI terrestrial imager includes 9 bands, including 8 multispectral bands and 1 panchromatic band. Wherein the multispectral waveband range is 0.43-2.29 μm, and the spatial resolution is 30 m; the panchromatic band ranges from 0.50 to 0.68 μm with a spatial resolution of 15 m.

In one possible implementation, as shown in fig. 3, step S210 further includes:

performing data fusion on the radar image data of the target area and the first remote sensing image data to obtain fused image data of the radar image data and the first remote sensing image data;

performing Prewitt edge detection on the fourth remote sensing image data of the target area to obtain edge detection image data; and

In one possible implementation manner, as shown in fig. 3, the fourth remote sensing image data is WorldView-3 remote sensing image data. In a specific example, the WorldView-3 satellite is a Digital Global high-resolution remote sensing satellite, 8 Short Wave Infrared (SWIR) wave bands are newly added to the WorldView-3 satellite on the basis of 8 visible light-near infrared (VNIR) wave bands, and the extraction capability of the ground feature information is greatly improved. In the embodiment, the WorldView-3 remote sensing image data is short wave infrared remote sensing image data.

In a specific example, as shown in fig. 3, before the data fusion of the radar image data of the target area and the first remote sensing image data, step S210 further includes: and preprocessing the radar image data of the target area and the first remote sensing image data of the target area respectively. In one specific example, the preprocessing of the radar image data of the target area and the first remote sensing image data of the target area includes, for example, radiometric calibration, FLAASH atmospheric correction, coordinate transformation, image cropping, and the like.

In a specific example, in order to embody hue and texture interpretation flag information of the tomographic construction information, the step S210 uses Sentinel-1 radar image data and Landsat-8 remote sensing image data to perform tomographic construction information interpretation. To better fuse the two, the Sentinel-1 radar image data is first resampled to 5 meters by, for example, a bilinear interpolation method, and then the resampled Sentinel-1 radar image data and the Landsat-8 color image data (color image data obtained by color synthesis using the

bands

7, 5, and 2 of the Landsat-8 remote sensing image data corresponding to the red channel, the green channel, and the blue channel, respectively, that is, the

bands

7, 5, and 2 of the Landsat-8 remote sensing image data are given to the red channel gray scale, the green channel gray scale, and the blue channel gray scale, respectively) are subjected to GS (gram schmidt) fusion (GS fusion can maintain the consistency of image spectrum information before and after fusion, and is a high-fidelity image fusion method), and it can be seen that the GS obtained fusion image data well maintains the spatial texture information and the remote sensing image data of the radar image data Spectral characteristic information of data (optical data).

On one hand: based on the fused image data shown in fig. 4, for example, the tomographic structure information of the target region can be interpreted from the hue abnormality and the texture information specific to the radar image data. The abnormal tone refers to the tone difference on two sides of the linear structure or the inconsistency with the target background tone. From the geological perspective, different wave band combinations reflect geologic bodies with different components, and color tone abnormal bands with color mutation at two sides of a linear trace probably represent the fault contact relation between the two geologic bodies; and texture information in the radar image data may be interpreted from tomographic information of the target region in which strong echo bright rays or bright bands formed by cliff of the rock formation are aligned. On the other hand: in order to better embody the texture information, the implementation method performs Prewitt edge detection (or Prewitt spatial filtering) based on the MATLAB platform on the WorldView-3 remote sensing image data with high spatial resolution, and the obtained edge detection image data is shown in fig. 5, for example. Prewitt edge detection has a smoothing effect on noise and better reflects texture information of ground objects. Because the fault structure is usually on one side of the mountain, the relief is large, and the texture density is large, the edge detection image data can be used as one of the basis for interpreting the fault structure information of the target area.

Thus, for example, the tomographic structure information of the target region shown in fig. 6 can be interpreted from the hue flag information represented by the fused image data of the Sentinel-1 radar image data and the Landsat-8 remote sensing image data and the texture flag information represented by the Sentinel-1 radar image data and the edge detection image data, and 8 pieces of tomographic structure information including F1 to F8 are shown in fig. 6.

S220, determining lithology information of the target area by using the second remote sensing image data of the target area.

In a specific example, the second remote sensing image data is, for example, hyperspectral remote sensing image data, and specifically, for example, GF-5 remote sensing image data is selected. The high-resolution five-number satellite (GF-5) is the first satellite in the world to carry out comprehensive observation on land and atmosphere simultaneously. GF-5 star carries 6 loads of an atmospheric trace gas differential absorption spectrometer, an atmospheric main greenhouse gas detector, an atmospheric multi-angle polarization detector, an atmospheric environment infrared very high resolution detector, a visible short wave infrared hyperspectral camera and a full spectrum section spectral imager, and can monitor a plurality of environmental elements such as atmospheric aerosol, sulfur dioxide, nitrogen dioxide, carbon dioxide, methane, water bloom, water quality, nuclear power plant warm drainage, land vegetation, straw burning, urban heat island and the like. The visible short wave infrared hyperspectral camera carried by the high-resolution five-number satellite is the first hyperspectral camera in the world which simultaneously considers wide coverage and wide spectrum range, can acquire 330 spectrum color channels in the spectrum color range from visible light to short wave infrared (400-2500 nm) under the conditions of 60 kilometer width and 30 meters spatial resolution, the color range is nearly 9 times wider than that of a common camera, the number of the color channels is nearly hundred times larger than that of the common camera, the spectrum resolution of the visible spectrum section is 5 nanometers, the visible short wave infrared hyperspectral camera can effectively detect the categories of an inland water body, a terrestrial ecological environment, altered minerals and rock and ore, and high-quality and high-reliability hyperspectral data are provided for industries of environment monitoring, resource exploration, disaster prevention and reduction and the like.

In a specific example, as shown in fig. 3, before determining the lithology information of the target region by using the second remote sensing image data of the target region, step S220 further includes: and preprocessing the second remote sensing image data of the target area. In one specific example, the preprocessing of the second remote sensing image data of the target area includes, for example, radiometric calibration, FLAASH atmospheric correction, coordinate transformation, image cropping, and the like.

In a specific example, as shown in fig. 3, in the process of interpreting the lithological information of the target region in step S220, after preprocessing GF-5 remote sensing image data of the target region, band filtering may be performed, for example: after checking the GF-5 data segment by segment, the segments 192-; meanwhile, because the wavelengths of 145-150(1006.88-1029.18nm) and 151-154(1004.77-1030.05nm) have overlapping parts, the above-mentioned bands are removed, and 287 bands are remained.

Thereafter, a successive maximum angle convex cone analysis (SMACC) algorithm may be employed to extract lithology information of the target region. The SMACC algorithm is an algorithm for automatically acquiring end members in an image and providing an abundance image of the end members based on a convex cone model, the data processing efficiency is greatly improved by an automatic end member extraction mode, and the influence of human factors on results is eliminated. For example, after the GF-5 remote sensing image data is processed by the SAMCC algorithm, 23 end members are extracted in total.

And then, analyzing the end element wave spectrum and screening end elements: end members are subjected to spectral analysis by comparing the remote sensing images and utilizing a USGS (United States Geological survey) spectral library, interference end members such as shadows and clouds and end members with broken distribution are eliminated, and the number of the end members which can assist in interpreting lithology information extraction is 3, as shown in FIG. 7. By analyzing the end-member wave spectrum, the method can be known,the spectrum of the end member 8 has two double reflection peaks between 800-1500nm, and the end member is consistent with quartz and SiO-containing material according to the result of spectrum analysis₂(ii) rock-like spectral features; the spectrum of the end member 10 has absorption characteristics at about 1500nm, 2000nm and 2330nm, and the end member is matched with the spectrum characteristics of the marble according to the spectrum analysis; the spectrum of the end member 11 showed absorption characteristics at about 810nm and 1000nm, and the end member was Fe-containing as seen by spectrum analysis²⁺And Fe³⁺Absorption characteristics typical of rocks.

Then, the 3 selected end members are assigned to a red channel (R), a green channel (G), and a blue channel (B) respectively for color synthesis, to obtain an end member abundance synthesis map shown in fig. 8, for example. In FIG. 8, quartz and SiO-containing₂The rock-like substance is mainly light (representing yellow, orange-red and pink); the marble rocks are darker in main color (representing green and yellow-green); containing Fe²⁺、Fe³⁺The mineral color was the darkest (representing dark blue). Compared with the traditional color synthetic image, the image synthesized by extracting the abundance of the end members through the SMACC algorithm in the embodiment has richer colors and more obvious details, and is very beneficial to lithologic boundary delineation.

Therefore, the distribution position and range of main large lithology in the target area can be identified through the end member color composite map extracted by the SMACC algorithm, the lithology information of the target area can be obtained through screening and interpretation according to the actual geological background by combining with the actual field survey data, and the lithology information of the target area can be displayed through the lithology unit interpretation map shown in the figure 9.

And S230, performing data fusion on the third remote sensing image data and the fourth remote sensing image data of the target area, inputting the data into a trained multi-classification calculation model for recognition to obtain pegmatite type lithium ore information of the target area, and determining pegmatite type lithium ore distribution density information of the target area according to the pegmatite type lithium ore information of the target area.

In one possible implementation manner, as shown in fig. 3, the third remote sensing image data is WorldView-2 remote sensing image data. In one specific example, WorldView-2 is a Digital Global commercial high resolution remote sensing satellite with not only 4 intra-industry standard bands (red, green, blue, near infrared), but also four additional bands (coast, yellow, red edge, and near infrared 2) that provide the user with the ability to perform accurate change detection and mapping. In this embodiment, the WorldView-2 remote sensing image data is visible light-near infrared remote sensing image data. In addition, the fourth remote sensing image data is WorldView-3 remote sensing image data.

In a specific example, as shown in fig. 3, before the data fusion of the third remote sensing image data and the fourth remote sensing image data of the target area, step S230 further includes: and respectively preprocessing the third remote sensing image data of the target area and the fourth remote sensing image data of the target area. In a specific example, the preprocessing of the third remote sensing image data of the target area and the fourth remote sensing image data of the target area includes, for example, radiometric calibration, FLAASH atmospheric correction, coordinate conversion, image cropping, and the like. The fourth remote sensing image data of the target area used in step S230 and step S210 may be regarded as the same image data, and may be preprocessed once.

In one specific example, step S230 performs fine pegmatite-type lithium ore extraction using WorldView-2 visible light-near infrared remote sensing image data and WorldView-3 short wave infrared remote sensing image data. Because the spectral characteristics of pegmatite lithium ore are mostly expressed in the short wave infrared band range, HPF fusion can be firstly carried out on the WorldView-2 remote sensing image data (namely, high-frequency information such as details and edges in a high-spatial-resolution panchromatic image is superposed on a low-resolution multispectral image), and then principal component analysis can be carried out on the WorldView-2 remote sensing image data, so that the purposes of data information retention and data compression are achieved, and the information of all bands is not overlapped. From the eigenvalue graph in the principal component analysis result shown in fig. 10, for example, it can be seen intuitively that the information amount mainly in the 8 transformed bands is concentrated in the 1 st band, because the curve exhibits a sharp inflection point at the 1 st band. Then, because the resolution of the WorldView-2 remote sensing image data is 0.46m, in order to improve the resolution of the WorldView-3 remote sensing image data, the step S230 carries out GS fusion capable of keeping the consistency of spectral information of the images before and after fusion on the principal component analysis result of the WorldView-2 remote sensing image data and the WorldView-3 remote sensing image data, and through the GS fusion, the spatial resolution of the WorldView-3 remote sensing image data is improved, the ground feature information content is enriched, and the WorldView-2 remote sensing image data and the WorldView-3 remote sensing image data can be well combined.

In a specific example, the third remote sensing image data and the fourth remote sensing image data of the target area are subjected to data fusion and then input into a trained multi-classification calculation model to be recognized as pegmatite type lithium mine recognition based on color features. It can be understood that the trained multi-class computation model is used for identifying the fused image data of the third remote sensing image data and the fourth remote sensing image data of the target area, so the training sample for training the multi-class computation model should use the fused image data sample with a plurality of pixels labeled with pegmatite-type lithium ore (that is, the pixels at the pegmatite-type lithium ore position in the fused image data sample are labeled or labeled) to train the multi-class computation model for identifying the pegmatite-type lithium ore information in the remote sensing image data of the target area, so as to obtain the trained multi-class computation model with learned identification capability.

In one particular example, recognition using a trained multi-class computational model may be performed based on the Matlab platform. Specifically, the multi-classification calculation model adopts OvO (one-to-one) multi-classification machine learning model, that is, 2 types of samples are picked out each time and combined two by two, and there are (N-1))/2 dichotomy cases in total, and the sample type is predicted by using the (N-1))/2 models, and the sample type with the largest variety and (N-1))/2 prediction results can be regarded as the final prediction type of the sample.

In one possible implementation manner, in order to make the pegmatite-type lithium mineral information of the target region obtained by identifying the fused image data of the third remote sensing image data and the fourth remote sensing image data of the target region by the trained multi-class computation model more accurate, as shown in fig. 3, step S230 further includes:

In one possible implementation manner, in order to remove the interference information in the pegmatite-type lithium mineral information of the target region, which is obtained by identifying the fused image data of the third remote sensing image data and the fourth remote sensing image data of the target region by the trained multi-class computation model, as shown in fig. 3, step S230 further includes:

after the threshold segmentation, performing an interference removal operation: and marking the pixel between the two pixel regions marked as the pegmatite lithium ore with the pixel distance smaller than the preset pixel distance threshold value as the pegmatite lithium ore by utilizing the parallel operation, and/or deleting the mark of each pixel in the pixel region marked as the pegmatite lithium ore with the pixel area smaller than the preset pixel area threshold value.

In one particular example, threshold segmentation and interference rejection operations may be performed based on a Matlab platform. For example, threshold segmentation with a grayscale threshold value of 50 may be adopted, that is, segmentation based on the grayscale threshold value 50 is performed on each of the red channel grayscale, the green channel grayscale, and the blue channel grayscale, if at least one of the red channel grayscale, the green channel grayscale, and the blue channel grayscale of the pixel labeled as pegmatite lithium ore is greater than the grayscale threshold value 50, the flag of the pixel is preliminarily determined to be retained, then, in order to eliminate redundancy of the above-mentioned segmentation results, segmentation based on the grayscale threshold value 50 is performed on an average value of the red channel grayscale, the green channel grayscale, and the blue channel grayscale, and if the average value is greater than the grayscale threshold value 50, the flag of the pixel is determined to be retained.

The result after threshold segmentation still has some interference information, so the interference removing operation is performed on the segmentation detection result:marking the pixel between two pixel regions marked as pegmatite type lithium ores with similar pixel distances (for example, less than 10 pixel distances) as pegmatite type lithium ores by utilizing parallel operation, and simultaneously rejecting the pixels with the area less than 10pixcel²The fragmented labels are pixel areas of the pegmatite lithium ores, and pegmatite lithium ore information extraction results are perfected.

In a specific example, after identifying the fused image data of the third remote sensing image data and the fourth remote sensing image data of the target area by using the trained multi-classification calculation model, respectively performing threshold segmentation and interference elimination operations, and displaying the image data including pegmatite-type lithium ore information output by the multi-classification calculation model based on the fused image data or pegmatite-type lithium ore information based on the fused image data as shown in fig. 11, wherein in fig. 11, the black marked area is a pixel area marked as pegmatite-type lithium ore.

In one possible implementation manner, in order to more accurately acquire pegmatite-type lithium ore information of the target region, as shown in fig. 3, the step S230 further includes:

performing data fusion on the third remote sensing image data and the fourth remote sensing image data of the target area to obtain fused image data of the third remote sensing image data and the fourth remote sensing image data;

and (3) carrying out wave band ratio operation: calculating three relation values of a reflection peak wave Band and an absorption valley wave Band in the corresponding wave Band of the first fusion image data according to the reflection spectrum curve of the pegmatite lithium ore to obtain three gray image data, wherein the relation values are ((Band)_a+Band_b)/Band_b)，Band_aRepresents a reflection peak Band, Band_bRepresents the absorption valley band;

Compared with the fused image data of the third remote sensing image data and the fourth remote sensing image data which are used as original images, the color synthetic image data obtained based on the wave band ratio operation has more obvious contrast or color difference, and is beneficial to more accurately identifying and obtaining the pegmatite type lithium ore information of the target area by a trained multi-classification calculation model.

In one specific example, before performing the band ratio operation, the characteristics of the reflection spectrum curve of the spodumene-containing pegmatite, that is, the valley point and the peak point of the reflection curve, need to be obtained and analyzed. In this example, a reflection spectrum curve of pegmatite containing lithium was extracted from the usgs (united States pharmaceutical survey) spectrum library, and the reflection spectrum curve was plotted based on coreldaw software, and the obtained reflection spectrum curve was shown in fig. 12. As can be seen from fig. 12, the pegmatite pulse spectrum containing spodumene has a moderate reflectance, and among them, has a strong absorption characteristic at a wavelength of about 1913nm, secondary absorption characteristics at a wavelength of about 2205nm and a wavelength of about 1413nm, and reflection characteristics at a wavelength of about 1385nm, a wavelength of about 2175nm, and a wavelength of about 2270 nm. Meanwhile, the characteristics can find corresponding wave bands in fused image data of the WorldView-2 remote sensing image data and the WorldView-3 remote sensing image data, for example, absorption characteristics with the wavelength of about 2205nm correspond to Band6 (wave Band6) of the fused image data; reflection characteristics at about 2175nm and 2270nm correspond to Band5 (Band 5) and Band7 (Band 7) of the fused image data, respectively; in addition, since three relation values need to be obtained, the Band4 (Band 4) of the fused image data corresponding to the relatively gentle reflection feature can be selected. Then, a ratio operation can be performed to calculate the ratio (as shown in formula 1) of the sum of the reflection peak band and the absorption valley band of the fused image data, thereby enhancing the color of the pegmatite-type lithium ore information.

(Band_a+Band_b)/Band_bEquation 1

In formula 1, Band_aRepresenting a characteristic Band of reflection, Band_bIndicating the absorption characteristic band. Respectively connected with the Band6 by using absorption characteristic wave BandThe ratio calculation is carried out on Band5, Band7 and Band4, and three gray scale image data corresponding to the three ratio calculation results of (Band5+ Band6)/Band6, (Band7+ Band6)/Band6, (Band4+ Band6)/Band6 can be obtained. It should be noted that, for the case that the third remote sensing image data and the fourth remote sensing image data are remote sensing image data other than WorldView-2 remote sensing image data and WorldView-3 remote sensing image data, the selection of the reflection characteristic band and the absorption characteristic band may be determined according to the band of the fused image data, and is not limited to selecting one absorption valley band and three reflection peak bands, for example, three absorption valley bands and one reflection peak band may be selected to obtain three ratio operation results, or two absorption valley bands and two reflection peak bands may be selected to obtain three ratio operation results.

Then, the three gray scale image data corresponding to the obtained three Band ratio calculation results are respectively assigned to a red channel, a green channel and a blue channel for color synthesis to form color synthesized image data, for example, the gray scale image data corresponding to (Band5+ Band6)/Band6 is assigned to the red channel, the gray scale image data corresponding to (Band7+ Band6)/Band6 is assigned to the green channel, and the gray scale image data corresponding to (Band4+ Band6)/Band6 is assigned to the blue channel, and the gray scale value of each pixel in the gray scale image data corresponding to (Band5+ Band6)/Band6 is used as the red channel gray scale of each pixel corresponding to the color synthesized image data. For example, fig. 13 shows color synthesized image data. As can be seen from fig. 13, the pegmatite-type lithium mineral information is in a highly reflective state and appears as a white stripe.

In one specific example, in the present implementation, color synthetic image data is input into a trained multi-class computational model for recognition as pegmatite-type lithium mineral recognition based on color features. It is to be understood that, in the present implementation, the trained multi-class calculation model is used for recognizing the color synthetic image data, so the training sample for training the multi-class calculation model should use the color synthetic image data sample with a plurality of pixels labeled with pegmatite-type lithium mineral to train the multi-class calculation model for recognizing pegmatite-type lithium mineral information in the remote sensing image data of the target region, so as to obtain the trained multi-class calculation model with learned recognition capability.

In one specific example, the multi-classification calculation model in the present embodiment may be a multi-classification machine learning model such as OvO (one-to-one), and the threshold segmentation and interference elimination operation based on a preset gray scale threshold of 50, for example, may be performed on image data including pegmatite-type lithium mineral information that is output after color synthetic image data is input to a trained multi-classification calculation model and recognized. After the color synthetic image data is identified by using the trained multi-class calculation model, threshold segmentation and interference removal operations are respectively performed, and finally, the image data based on the color synthetic image data and including the pegmatite-type lithium mineral information output by the multi-class calculation model or the pegmatite-type lithium mineral information based on the color synthetic image data is displayed as shown in fig. 14, wherein in fig. 14, a black marked area is a pixel area marked as pegmatite-type lithium mineral.

Next, the difference in accuracy between the pegmatite-type lithium mineral information based on the color synthetic image data and the pegmatite-type lithium mineral information based on the fused image data is evaluated by an evaluation factor:

firstly, manufacturing a verification point: and converting pegmatite type lithium ore and other ground object information in the field mapping data into point SHP files, and using the point SHP files as standards for checking the accuracy of the extraction results of the two-time machine learning. And finally, carrying out vectorization on the pegmatite lithium mine information based on the color synthetic image data and the pegmatite lithium mine information based on the fusion image data, respectively superposing the verification points, the lithology information vectors, the pegmatite lithium mine information based on the color synthetic image data and the pegmatite lithium mine information vectorized file based on the fusion image data into ArcGIS software, and respectively assigning the attributes of the two machine learning extraction results into the verification points by using a space coupling tool (Spatial Join). And then, performing precision evaluation on the two extraction results by using a confusion matrix: firstly, directly using EXCEL to open dbf files in SHP according to the results of the previous space hooking, making a data pivot table, and using columns as verification data and rows as extraction results to generate a confusion matrix. Then, an evaluation factor program in the confusion matrix is compiled by using an MATLAB platform, wherein the evaluation factors mainly comprise: overall classification accuracy (OA), Kappa coefficient (Kappa), missing score error (OE), drawing accuracy (PA), User Accuracy (UA). Wherein, the OA and Kappa are the integral precision evaluation, and the higher the numerical value is, the higher the precision is; OE, PA, and UA are accuracy evaluations of pegmatite-type lithium ore extracted automatically, and smaller OE results in higher accuracy, whereas smaller OE results in opposite results in higher accuracy in PA and UA. The implementation of this program requires the separate entry of the pivot tables. The run results were as follows:

pegmatite-type lithium ore information based on color synthetic image data: OA1 ═ 87.5934, Kappa1 ═ 0.8396, OE1 ═ 0.3600, PA1 ═ 89.6400, UA1 ═ 88.5800;

pegmatite type lithium ore information based on fused image data: OA2 ═ 73.2947, Kappa2 ═ 0.7179, OE2 ═ 0.4800, PA2 ═ 70.2900, UA2 ═ 69.9700;

it can be seen that the pegmatite lithium mineral information based on the color synthetic image data and the pegmatite lithium mineral information based on the fused image data have better overall accuracy, and particularly, the pegmatite lithium mineral information based on the color synthetic image data has higher accuracy, which verifies the effectiveness of the identification based on the color synthetic image data after the ratio operation on the improvement of the identification accuracy.

In one specific example, in order to more objectively represent the Density of the pegmatite-type lithium mineral spatial distribution, pegmatite-type lithium mineral information extracted based on the chromatic synthetic image data or the band ratio calculation may be converted into a point vector, and pegmatite-type lithium mineral distribution Density information of a target region may be acquired by a Kernel Density analysis tool in ArcGIS, and a pegmatite-type lithium mineral distribution Density map may be plotted, where the pegmatite-type lithium mineral distribution Density map is shown in fig. 15, and a darker color region in fig. 15 indicates a higher Density of pegmatite-type lithium mineral.

Note that, for pegmatite-type lithium mineral information based on the fused image data, the pegmatite-type lithium mineral information may be converted into a point vector, and pegmatite-type lithium mineral distribution Density information of the target region may be acquired by a Kernel Density analysis tool in ArcGIS, and a pegmatite-type lithium mineral distribution Density map may be plotted.

S240, determining a pegmatite type lithium ore exploration target area in the target area according to the fault structure information, the lithological information and the pegmatite type lithium ore distribution density information of the target area.

The fault structure information of the target region and the lithological property information of the target region are used as indirect basis for determining the pegmatite type lithium mineral exploration target region in the target region, and the pegmatite type lithium mineral distribution density information of the target region is used as direct basis for determining the pegmatite type lithium mineral exploration target region in the target region.

In a specific example, referring to fig. 6 and 9, according to the fault structure information and lithology information of the target area decoded in steps S210 and S220, it can be seen that the stratum of the target area is mainly a set of metamorphic watchcase rocks of low-angle amphibole phase of ancient meta-ancient kingdom rock group, and can be generally divided into a gneiss section and a schist section, the surrounding rocks are biotite plagioclasite, and the invasion rocks are mainly otactic quartzite, quartzite and senegary dilongite. The biotite granite in the area can be differentiated into light-colored granite and lithiumbite granite-pegmatite, and the formation of pegmatite type lithium ores is related to dimotite granite; the fourth rock group marble rock clamp metamorphic rock is mainly exposed in the zone, the bottom is calcareous mylonite which is in flexible shear zone (fault) contact with the Arkinsonia rock group and is one of the main mineral-bearing strata in the zone.

And (4) integrating the ore-forming analysis to define the target area of the ore-finding. The fault structure can be subjected to buffer analysis, and a buffer zone of 2km for example is established for the fault structure, so that the position relation between the pegmatite type lithium ore and the fault structure can be better represented.

In one possible implementation manner, step S240 further includes:

and displaying a laminated graph of a first image containing fault structure information of the target area and buffer area information thereof, a second image containing lithologic information of the target area and a third image containing pegmatite type lithium ore distribution density information of the target area so as to be convenient for manually delineating an ore-searching target area, and displaying the identification of the pegmatite type lithium ore-searching target area in the target area by using the laminated graph.

In a specific example, a pegmatite-type lithium mineral distribution density map such as shown in fig. 15, a lithology unit interpretation map such as shown in fig. 9 (which may be processed with a transparency of 50%), tomographic structure information of a target region such as shown in fig. 6, and a buffer of the tomographic structure may be displayed in an Arcgis in an overlaid manner, and a superimposed overlay map such as shown in fig. 16 may be displayed in an overlaid manner. The following analytical results were obtained from FIG. 16: (1) the extracted pegmatite type lithium ore information is densely distributed near a fourth rock group and an Argin a rock group fault of the New Taigu kingdom, and an ore-forming target area is trapped; (2) the formation of the pegmatite type lithium ore is related to diascovite granite, and the extracted pegmatite type lithium ore information is more concentrated in distribution of the aotao biotite diaxovite granite in the area, so that the pegmatite type lithium ore has high ore forming possibility; (3) meanwhile, the main geologic body of the rare metal lithium ore is granite pegmatite vein or muscovite microcline feldspar pegmatite vein, the vein is mainly formed in the stratums of Milangbei rock group and Argin rock group, and the extracted pegmatite type lithium ore is densely distributed in the Argin rock group according to the information distribution. Based on the analysis results, the target area for mineral exploration or the target area for mineral exploration of pegmatite-type lithium mineral can be automatically or manually identified, and the result of the identified target area for mineral exploration is shown in fig. 17, for example.

It should be understood by those skilled in the art that although the steps are described in the order of S210-S240, the steps are not necessarily performed in such an order, for example, S220 may be performed first and then S210 or S230 may be performed first and then S220 may be performed, and then for example, S210, S220 and S230 may be performed in parallel, as long as the logic is not violated.

As shown in fig. 18, another embodiment of the present invention provides an ore searching apparatus for pegmatite-type lithium ore, including:

It should be noted that the mineral exploration apparatus for pegmatite-type lithium mine provided in this embodiment may be understood as a server that integrates the fault structure information interpretation server 110, the lithology information interpretation server 120, the training server 130, the recognition server 140, and the target area delineation server 150 in fig. 1. The principle and the working flow of the pegmatite type lithium ore prospecting device provided by the embodiment are similar to those of the pegmatite type lithium ore prospecting method, and the above description can be referred to for relevant parts, and the details are not repeated herein.

As shown in fig. 19, a computer system suitable for implementing the mineral exploration device for pegmatite-type lithium mines provided in the above-described embodiment includes a Central Processing Unit (CPU) that can perform various appropriate actions and processes according to a program stored in a Read Only Memory (ROM) or a program loaded from a storage portion into a Random Access Memory (RAM). In the RAM, various programs and data necessary for the operation of the computer system are also stored. The CPU, ROM, and RAM are connected thereto via a bus. An input/output (I/O) interface is also connected to the bus.

An input section including a keyboard, a mouse, and the like; an output section including a speaker and the like such as a Liquid Crystal Display (LCD); a storage section including a hard disk and the like; and a communication section including a network interface card such as a LAN card, a modem, or the like. The communication section performs communication processing via a network such as the internet. The drive is also connected to the I/O interface as needed. A removable medium such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like is mounted on the drive as necessary, so that a computer program read out therefrom is mounted into the storage section as necessary.

In particular, the processes described in the above flowcharts may be implemented as computer software programs according to the present embodiment. For example, the present embodiments include a computer program product comprising a computer program tangibly embodied on a computer-readable medium, the computer program comprising program code for performing the method illustrated in the flow chart. In such an embodiment, the computer program may be downloaded and installed from a network via the communication section, and/or installed from a removable medium.

The flowchart and schematic diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to the present embodiments. In this regard, each block in the flowchart or schematic diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the schematic and/or flowchart illustration, and combinations of blocks in the schematic and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

The modules described in the present embodiment may be implemented by software or hardware. The described modules may also be provided in a processor, which may be described as: a processor includes a first determination module, a second determination module, a third determination module, and a fourth determination module. Wherein the names of the modules do not in some cases constitute a limitation of the module itself. For example, the first determination module may also be described as a "first interpretation module".

On the other hand, the present embodiment also provides a nonvolatile computer storage medium, which may be the nonvolatile computer storage medium included in the apparatus in the foregoing embodiment, or may be a nonvolatile computer storage medium that exists separately and is not assembled into a terminal. The non-volatile computer storage medium stores one or more programs that, when executed by a device, cause the device to: determining fault structure information of a target area by using radar image data and first remote sensing image data of the target area; determining lithology information of the target area by using the second remote sensing image data of the target area; performing data fusion on the third remote sensing image data and the fourth remote sensing image data of the target area, inputting the data into a trained multi-classification calculation model for recognition to obtain pegmatite lithium ore information of the target area, and determining pegmatite lithium ore distribution density information of the target area according to the pegmatite lithium ore information of the target area; and determining a pegmatite type lithium ore prospecting target area in the target area according to the fault structure information, the lithology information and the pegmatite type lithium ore distribution density information of the target area.

In the description of the present invention, it should be noted that the terms "upper", "lower", and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the drawings, which are merely for convenience in describing the present invention and simplifying the description, and do not indicate or imply that the referred device or element must have a specific orientation, be constructed in a specific orientation, and operate, and thus, should not be construed as limiting the present invention. Unless expressly stated or limited otherwise, the terms "mounted," "connected," and "connected" are intended to be inclusive and mean, for example, that they may be fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood by those skilled in the art according to specific situations.

It is further noted that, in the description of the present invention, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.

It should be understood that the above-mentioned embodiments of the present invention are only examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention, and it will be obvious to those skilled in the art that other variations and modifications can be made on the basis of the above description, and all embodiments cannot be exhaustive, and all obvious variations and modifications belonging to the technical scheme of the present invention are within the protection scope of the present invention.

Claims

1. An ore searching method for pegmatite type lithium ore is characterized by comprising the following steps:

2. The method of claim 1, wherein the step of performing data fusion on the third remote sensing image data and the fourth remote sensing image data of the target region and inputting the data into a trained multi-classification calculation model for recognition to obtain the pegmatite-type lithium ore information of the target region further comprises the steps of:

3. The method according to claim 1, wherein the step of performing data fusion on the third remote sensing image data and the fourth remote sensing image data of the target region, inputting the data into a trained multi-classification calculation model for recognition, and obtaining the pegmatite-type lithium ore information of the target region further comprises the steps of:

4. The method according to claim 3, wherein the step of performing data fusion on the third remote sensing image data and the fourth remote sensing image data of the target region and inputting the data into a trained multi-classification calculation model for recognition to obtain the pegmatite type lithium ore information of the target region further comprises the steps of:

5. The method of claim 1, wherein determining tomographic information of the target region using the radar image data and the first remotely sensed image data of the target region further comprises:

6. The method according to claim 1, wherein the determining the pegmatite-type lithium mineral exploration target area in the target region according to the fault structure information, the lithological information and the pegmatite-type lithium mineral distribution density information of the target region further comprises:

7. The method according to any one of claims 1-6, wherein the third remote sensing image data is WorldView-2 remote sensing image data and the fourth remote sensing image data is WorldView-3 remote sensing image data.

8. An ore searching device for pegmatite-type lithium ore, comprising:

9. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the method according to any of claims 1-7 when executing the program.

10. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the method according to any one of claims 1-7.