Remote sensing survey image processing system based on AI prospecting technology

Abstract

The invention belongs to the technical field of prospecting, relates to a data analysis technology, and is used for solving the problem that an existing image processing system does not fully consider the internal relation between pixel spectral data and lacks reasonable grouping, so that the efficiency is low when a large amount of pixel data is processed. The remote sensing survey image processing system based on the AI prospecting technology comprises a remote sensing image processing center, and the remote sensing image processing center is in communication connection with an image acquisition module, a pixel grouping module, a grouping verification module, a spectrum matching module and a standard mineral spectrum library; the image acquisition module is used for acquiring a remote sensing image of an area to be identified, the pixel grouping module is used for grouping pixels in the remote sensing image, the grouping verification module is used for verifying a pixel grouping result, and the spectrum matching module is used for matching the pixels with standard minerals; according to the method, the internal relation of the spectral data between the pixels is considered, the pixel data are integrated, and repeated calculation of each pixel is avoided.

Description

Technical Field

[0001] The present invention belongs to the technical field of prospecting, involves data analysis technology, and specifically is a remote sensing survey image processing system based on AI prospecting technology. Background Art

[0002] In the field of mineral resource exploration, with the development of technology, AI-based prospecting technology has gradually become the mainstream. Among them, remote sensing survey image processing is a key link to obtain underground mineral information. As the basic unit of remote sensing images, the spectral data of each pixel contains rich geological information.

[0003] Hyperspectral remote sensing, as a technology for mineral identification, has numerous bands. Each pixel has corresponding spectral data in multiple bands, resulting in a sharp increase in data volume. Processing hyperspectral massive data not only requires efficient algorithms and powerful computing hardware, but also faces the problems of data redundancy and information extraction.

[0004] Existing image processing systems often do not fully consider the internal relationship between pixel spectral data. Usually, pixels are regarded in isolation and not reasonably grouped, resulting in low efficiency when processing a large amount of pixel data and being unable to quickly extract valuable information.

[0005] In view of the above technical problems, this application proposes a solution. Summary of the Invention

[0006] The purpose of the present invention is to provide a remote sensing survey image processing system based on AI prospecting technology, which is used to solve the problem that existing image processing systems do not fully consider the internal relationship between pixel spectral data and do not reasonably group them, resulting in low efficiency when processing a large amount of pixel data. The technical problem that the present invention needs to solve is: how to provide a remote sensing survey image processing system based on AI prospecting technology that can consider the internal relationship between pixel spectral data and reasonably group them.

[0007] The purpose of the present invention can be achieved through the following technical solutions: A remote sensing survey image processing system based on AI prospecting technology includes a remote sensing image processing center, which is communicatively connected to an image acquisition module, a pixel grouping module, a grouping verification module, a spectral matching module, and a standard mineral spectral library. The image acquisition module is used to acquire remote sensing images of the area to be identified: a sensor is used to acquire remote sensing images of the area to be identified, and preprocessing operations are performed on the remote sensing images to obtain processed images; the basic unit of the processed images is pixels, and each pixel stores the spectral data of the ground area corresponding to the pixel. The pixel grouping module is used to group the pixels in the remote sensing image: obtain the spectral data of the pixels in the processed image, and divide the pixels into several pixel groups according to the spectral data of the pixels; The grouping verification module is used to verify the result of pixel grouping: verify the pixel grouping according to the spectral data of the pixels, remove the pixels that do not belong to the pixel group and mark them as independent pixels; The spectral matching module is used to match the pixels with standard minerals: match the pixels in the processed image with the standard mineral spectral library, and the matching methods include standard matching and fast matching; standard matching is used for independent pixels, and fast matching is used for the pixels in the pixel group.

[0008] Furthermore, the preprocessing operations include radiometric correction, geometric correction, and noise removal; the purpose of radiometric correction is to eliminate the radiometric errors caused by factors such as the characteristics of the sensor itself, atmospheric transmission, and illumination; the purpose of geometric correction is to correct the geometric deformations caused by factors such as terrain undulation and atmospheric refraction; the purpose of noise removal is to reduce interference and improve the image quality.

[0009] Furthermore, the spectral data includes the band BDn and the spectral reflectance FSn corresponding to the band BDn, where the band BDn is a wavelength range [a, b], and n is the number of bands.

[0010] Furthermore, the specific process of grouping the pixels according to the spectral data includes: Step 1: Calculate the average value of the wavelength range [a, b] of the band BDn to obtain the intermediate wavelength ZBn, and sum the product results of all the intermediate wavelengths ZBn of a single pixel and the corresponding spectral reflectance FSn to obtain the spectral performance value BX of the pixel; Step 2: Arrange all the pixels in the processed image in descending order of the spectral performance value BX to obtain a pixel sequence; calculate the difference in the spectral performance value BX between adjacent two pixels in sequence according to the order of the pixel sequence and mark it as the spectral difference; Step 3: Compare all the spectral differences with a preset spectral difference threshold: if the spectral difference is less than the spectral difference threshold, it is judged that there is no difference between the two pixels for which the spectral difference is calculated, and no processing is required; if the spectral difference is greater than or equal to the spectral difference threshold, it is judged that there is a difference between the two pixels for which the spectral difference is calculated, and the pixel sequence is segmented between these two pixels, and the pixel sequence is divided into several pixel groups through multiple segmentations, and the spectral performance values BX between the pixels within each pixel group are not different.

[0011] Furthermore, the specific process of verifying the pixel grouping according to the spectral data includes: Step 1: Select the first k bands of the pixel as the verification band BDk, where k < n; sum the product results of the middle wavelengths ZBk of all verification bands BDk and the corresponding spectral reflectance FSn to obtain the spectral verification value of the pixel. Step 2: Calculate the average value and variance of the spectral verification values of all pixels in the pixel group to obtain the verification mean and verification variance; calculate the ratio of the difference between the spectral verification value and the verification mean to the verification variance to obtain the verification coefficient YZ of the pixels in the pixel group. Step 3: Compare the verification coefficient YZ of the pixels in the pixel group with the preset verification threshold YZmax: If the verification coefficient YZ is less than the verification threshold YZmax, it is determined that the pixel belongs to the pixel group and no processing is required; if the verification coefficient YZ is greater than or equal to the verification threshold YZmax, it is determined that the pixel does not belong to the pixel group, and the pixel is removed from the pixel group and marked as an independent pixel.

[0012] Further, the specific process of standard matching is as follows: Select a target mineral from the standard mineral spectral library, obtain the characteristic bands and spectral standard vector BZ of the target mineral; extract the spectral reflectance FSn corresponding to the characteristic bands of the target mineral in the pixel and form a spectral characteristic vector TZ; divide the dot product result of the spectral characteristic vector TZ of the pixel and the spectral standard vector BZ of the target mineral by the product of the moduli of the two vectors to obtain the cosine value, and take the inverse cosine of the cosine value to obtain the angle between the two vectors and mark it as the matching angle θ; compare the matching angle θ with the preset matching angle threshold θmax: If the matching angle θ is less than the matching angle threshold θmax, it is determined that the matching is successful, and the target mineral is marked as the matching mineral of the pixel; if the matching angle θ is greater than or equal to the matching angle threshold θmax, it is determined that the matching fails, and a new target mineral is selected from the standard mineral spectral library to perform the matching again until all the matching is completed.

[0013] Further, the specific process of fast matching is as follows: Perform standard matching on the first pixel and the last pixel in the pixel group respectively. If the results of the two standard matchings are the same target mineral, it is determined that the matching is successful, and the target mineral is marked as the matching mineral of all pixels in the pixel group; if the results of the two standard matchings are not the same target mineral, the pixel group is evenly divided into two new pixel groups, and the first pixel and the last pixel in the new pixel groups are respectively subjected to standard matching again, and the above operations are repeated until the results of the two standard matchings in the pixel group are the same target mineral.

[0014] A remote sensing survey image processing method based on AI prospecting technology includes the following steps: Step 1: Use a sensor to obtain a high-resolution remote sensing image of the area to be identified, and perform preprocessing operations on the remote sensing image to obtain a processed image. Step 2: Obtain the spectral data of the pixels in the processed image, and divide the pixels into pixel groups according to the spectral data of the pixels; Step 3: Verify the grouping of the pixels according to the spectral data of the pixels, remove the pixels that do not belong to the pixel group and mark them as independent pixels; Step 4: Select target minerals from the standard mineral spectral library to match with the pixels in the processed image, adopt standard matching for independent pixels, adopt fast matching for the pixels in the pixel group, and mark the successfully matched target minerals as the matching minerals of the pixels.

[0015] The present invention has the following beneficial effects: 1. The image acquisition module uses satellites, airplanes or other sensors to obtain high-resolution remote sensing images, and performs preprocessing operations such as radiometric correction, geometric correction, and noise removal, providing an accurate and reliable data basis for subsequent pixel analysis, and greatly reducing information misjudgment and omission caused by image quality problems; 2. The pixel grouping module calculates the spectral performance value according to the pixel spectral data and groups the pixels, fully considering the internal relationship of the spectral data between pixels, grouping the pixels with similar spectral performance into one group, effectively integrating a large amount of pixel data, avoiding repeated calculation for each pixel, greatly reducing the system calculation burden, improving the overall operation efficiency, and helping to process more remote sensing image data within a limited time; 3. The grouping verification module selects the first k bands of the pixels as the verification bands, judges whether the pixels belong to the corresponding pixel group by calculating the verification coefficient and comparing it with the preset verification threshold; this multi-band verification makes full use of the multi-band characteristics of the pixel spectral data, can more accurately judge the rationality of the pixel grouping, and effectively reduces misjudgment; 4. The spectral matching module adopts standard matching for independent pixels and fast matching for the pixels in the pixel group, greatly improving the matching efficiency while ensuring a certain accuracy, meeting the requirements of accurate matching for single pixels and also taking into account the requirement of fast processing of a large amount of data in the pixel group. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0017] Figure 1 It is the system block diagram of the first embodiment of the present invention; Figure 2 It is the method flow chart of the second embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0018] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts belong to the scope of protection of the present invention.

[0019] Embodiment 1: As Figure 1 shown, a remote sensing survey image processing system based on AI prospecting technology includes a remote sensing image processing center, which is communicatively connected to an image acquisition module, a pixel grouping module, a grouping verification module, a spectral matching module, and a standard mineral spectral library; The image acquisition module is used to acquire remote sensing images of the area to be recognized: high-resolution remote sensing images of the area to be recognized are obtained by using satellites, airplanes or other sensors, and the acquired remote sensing images are preprocessed to obtain high-quality remote sensing images and marked as processed images; it should be noted that common preprocessing operations include radiation correction, geometric correction, and noise removal; the purpose of radiation correction is to eliminate radiation errors caused by factors such as the characteristics of the sensor itself, atmospheric transmission, and illumination; the purpose of geometric correction is to correct geometric deformations caused by factors such as terrain undulation and atmospheric refraction; the purpose of noise removal is to reduce interference and improve image quality; the basic unit of the processed image is a pixel, and each pixel stores the spectral data of the ground area corresponding to the pixel, and the size of the pixel is determined by the spatial resolution of the sensor; The image acquisition module uses satellites, airplanes or other sensors to obtain high-resolution remote sensing images and performs preprocessing operations such as radiation correction, geometric correction, and noise removal, providing an accurate and reliable data basis for subsequent pixel analysis, and greatly reducing information misjudgment and omission caused by image quality problems.

[0020] The pixel grouping module is used to group the pixels in the remote sensing image: obtain the spectral data of the pixels in the processed image, and the spectral data includes the band BDn and the spectral reflectance FSn corresponding to the band BDn, where the band BDn is a wavelength range [a, b], and n is the number of bands; the specific process of grouping the pixels according to the spectral data of the pixels includes: Step 1: Calculate the average value of the wavelength range [a, b] of the band BDn to obtain the intermediate wavelength ZBn, and sum the product results of all the intermediate wavelengths ZBn of a single pixel and the corresponding spectral reflectance FSn to obtain the spectral performance value BX of the pixel; Step 2: Arrange all the pixels in the processed image in descending order of the spectral performance value BX to obtain a pixel sequence; calculate the difference in the spectral performance value BX between adjacent two pixels in sequence according to the order of the pixel sequence and label it as the spectral difference; Step 3: Compare all the spectral differences with a preset spectral difference threshold: If the spectral difference is less than the spectral difference threshold, it is determined that there is no difference between the two pixels for which the spectral difference is calculated and no processing is required; if the spectral difference is greater than or equal to the spectral difference threshold, it is determined that there is a difference between the two pixels for which the spectral difference is calculated, and the pixel sequence is segmented between these two pixels. Through multiple segmentations, the pixel sequence is divided into several pixel groups, and there is no difference in the spectral performance value BX between the pixels within each pixel group; The pixel grouping module calculates the spectral performance value based on the pixel spectral data and groups the pixels. It fully considers the internal relationship of the spectral data between pixels, groups the pixels with similar spectral performance into one group, effectively integrates a large amount of pixel data, avoids repeated calculations for each pixel, greatly reduces the system calculation burden, improves the overall operation efficiency, and helps to process more remote sensing image data within a limited time.

[0021] The grouping verification module is used to verify the result of pixel grouping: The specific process of verifying the pixel grouping according to the spectral data of the pixels includes: Step 1: Select the first k bands of the pixel as the verification band BDk, where k < n; calculate the sum of the product results of the middle wavelength ZBk of all the verification bands BDk and the corresponding spectral reflectance FSn to obtain the spectral verification value of the pixel; Step 2: Calculate the average value and variance of the spectral verification values of all the pixels in the pixel group to obtain the verification mean and verification variance; calculate the ratio of the difference result between the spectral verification value and the verification mean to the verification variance to obtain the verification coefficient YZ of the pixels in the pixel group; Step 3: Compare the verification coefficient YZ of the pixels in the pixel group with a preset verification threshold YZmax: If the verification coefficient YZ is less than the verification threshold YZmax, it is determined that the pixel belongs to the pixel group and no processing is required; if the verification coefficient YZ is greater than or equal to the verification threshold YZmax, it is determined that the pixel does not belong to the pixel group, and the pixel is removed from the pixel group and labeled as an independent pixel; The grouping verification module selects the first k bands of the pixel as the verification band, calculates the verification coefficient and compares it with the preset verification threshold to determine whether the pixel belongs to the corresponding pixel group; this multi-band verification makes full use of the multi-band characteristics of the pixel spectral data, can more accurately judge the rationality of pixel grouping, and effectively reduces misjudgment.

[0022] The spectral matching module is used to match pixels with standard minerals: match the pixels in the processed image with the standard mineral spectral library, and the matching methods include standard matching and fast matching; among them, standard matching is used for independent pixels, and fast matching is used for pixels in the pixel group; The specific process of standard matching is as follows: select a target mineral from the standard mineral spectral library, obtain the characteristic band BDm of the target mineral, where m is the number of characteristic bands of the target mineral; obtain the spectral standard vector BZ = [FS1, FS2,..., FSm] of the target mineral, and FSm is the spectral reflectance of the target mineral at the corresponding characteristic band BDm; extract the spectral reflectance FSn corresponding to the characteristic band of the target mineral in the pixel and form the spectral feature vector TZ = [FS1, FS2,..., FSm]; divide the dot product result of the spectral feature vector TZ of the pixel and the spectral standard vector BZ of the target mineral by the product of the moduli of the two vectors to obtain the cosine value, and take the arccosine of the cosine value to obtain the angle between the two vectors and mark it as the matching angle θ; compare the matching angle θ with the preset matching angle threshold θmax: if the matching angle θ is less than the matching angle threshold θmax, it is judged that the matching is successful, and the target mineral is marked as the matching mineral of the pixel; if the matching angle θ is greater than or equal to the matching angle threshold θmax, it is judged that the matching fails, and a new target mineral is selected from the standard mineral spectral library to perform the matching again until all the matching is completed; The specific process of fast matching is as follows: perform standard matching on the first pixel and the last pixel in the pixel group respectively. If the results of the two standard matchings are the same target mineral, it is judged that the matching is successful, and the target mineral is marked as the matching mineral of all the pixels in the pixel group; if the results of the two standard matchings are not the same target mineral, the pixel group is evenly divided into two new pixel groups, and standard matching is performed on the first pixel and the last pixel in the new pixel groups respectively, and the above operations are repeated until the results of the two standard matchings in the pixel group are the same target mineral; The spectral matching module adopts standard matching for independent pixels and fast matching for pixels in the pixel group, which greatly improves the matching efficiency while ensuring a certain accuracy, meeting both the requirement of precise matching for individual pixels and the requirement of rapid processing of a large amount of data in the pixel group.

[0023] Embodiment 2: As Figure 2 shown, a remote sensing survey image processing method based on AI prospecting technology includes the following steps: Step 1: Use a sensor to obtain a high-resolution remote sensing image of the area to be recognized, and perform preprocessing operations on the remote sensing image to obtain a processed image; Step 2: Obtain the spectral data of the pixels in the processed image, and divide the pixels into pixel groups according to the spectral data of the pixels; Step 3: Verify the grouping of pixels based on the spectral data of the pixels, remove the pixels that do not belong to the pixel group and mark them as independent pixels; Step 4: Select target minerals from the standard mineral spectral library to match with the pixels in the processed image. Use standard matching for independent pixels and fast matching for pixels in the pixel group, and mark the successfully matched target minerals as the matching minerals of the pixels.

[0024] A remote sensing survey image processing system based on AI prospecting technology. During operation, it uses a sensor to obtain a high-resolution remote sensing image of the area to be identified, and performs preprocessing operations on the remote sensing image to obtain a processed image; Obtain the spectral data of the pixels in the processed image, divide the pixels into pixel groups according to the spectral data of the pixels; verify the grouping of the pixels according to the spectral data of the pixels, remove the pixels that do not belong to the pixel group and mark them as independent pixels; select target minerals from the standard mineral spectral library to match with the pixels in the processed image, use standard matching for independent pixels and fast matching for pixels in the pixel group, and mark the successfully matched target minerals as the matching minerals of the pixels.

[0025] The above content is only an example and explanation of the structure of the present invention. Those skilled in the art of this technology can make various modifications or supplements to the described specific embodiments or use similar methods to replace them. As long as they do not deviate from the structure of the invention or exceed the scope defined by this claim book, they should fall within the protection scope of the present invention.

[0026] In the description of this specification, the descriptions referring to terms such as "one embodiment", "example", "specific example", etc. mean that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in a suitable manner in any one or more embodiments or examples.

[0027] The above-disclosed preferred embodiments of the present invention are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the present invention to only the specific implementation manners. Obviously, according to the content of this specification, many modifications and changes can be made. This specification selects and specifically describes these embodiments to better explain the principle and practical application of the present invention, so that those skilled in the art can understand and utilize the present invention well. The present invention is only limited by the claim book and its full scope and equivalents.

Claims

1. A remote sensing survey image processing system based on AI prospecting technology, characterized in that: It includes a remote sensing image processing center, which is communicatively connected with an image acquisition module, a pixel grouping module, a grouping verification module, a spectrum matching module and a standard mineral spectrum library; The image acquisition module is used to acquire remote sensing images of the area to be identified: a sensor is used to acquire remote sensing images of the area to be identified, and a preprocessing operation is performed on the remote sensing images to obtain processed images; the basic unit of the processed image is a pixel, and each pixel stores the spectral data of the ground area corresponding to the pixel; The pixel grouping module is used to group pixels in the remote sensing image: obtain spectral data of pixels in the processed image, and divide the pixels into a plurality of pixel groups according to the spectral data of the pixels; The grouping verification module is used to verify the result of pixel grouping: verify the pixel grouping according to the spectral data of the pixel, remove the pixels that do not belong to the pixel group and mark them as independent pixels; The spectral matching module is used to match pixels with standard minerals: matching pixels in the processed image with the standard mineral spectral library, the matching methods include standard matching and fast matching; standard matching is used for independent pixels, and fast matching is used for pixels in a pixel group.

2. According to claim 1, a remote sensing survey image processing system based on AI prospecting technology is characterized in that: Preprocessing operations include radiation correction, geometric correction and noise removal; the purpose of radiation correction is to eliminate radiation errors caused by the characteristics of the sensor itself, atmospheric transmission and lighting factors; the purpose of geometric correction is to correct geometric deformation caused by terrain undulations and atmospheric refraction factors; the purpose of noise removal is to reduce interference and improve image quality.

3. The remote sensing survey image processing system based on AI prospecting technology according to claim 2 is characterized in that: The spectral data includes a band BDn and a spectral reflectance FSn corresponding to the band BDn, wherein the band BDn is a wavelength range [a, b], and n is the number of bands.

4. The remote sensing survey image processing system based on AI prospecting technology according to claim 3 is characterized in that: The specific process of grouping pixels according to spectral data includes: Step 1: Take the average value of the wavelength range [a, b] of the band BDn to obtain the intermediate wavelength ZBn, and sum the product results of all intermediate wavelengths ZBn of a single pixel and the corresponding spectral reflectance FSn to obtain the spectral performance value BX of the pixel; Step 2: Arrange all pixels in the processed image in descending order of spectral performance value BX to obtain a pixel sequence; calculate the difference between the spectral performance value BX of two adjacent pixels in the order of the pixel sequence and mark it as the spectral difference; Step 3: Compare all spectral differences with the preset spectral difference threshold: if the spectral difference is less than the spectral difference threshold, it is judged that there is no difference between the two pixels for calculating the spectral difference, and no processing is required; if the spectral difference is greater than or equal to the spectral difference threshold, it is judged that there is a difference between the two pixels for calculating the spectral difference, and the pixel sequence is divided between the two pixels. The pixel sequence is divided into several pixel groups through multiple divisions, and the spectral performance values ​​BX between the pixels in each pixel group are not different.

5. The remote sensing survey image processing system based on AI prospecting technology according to claim 4 is characterized in that: The specific process of verifying the grouping of pixels based on spectral data includes: Step 1: Select the first k bands of the pixel as the verification bands BDk, where k < n; sum the product results of the middle wavelengths ZBk of all verification bands BDk and the corresponding spectral reflectances FSn to obtain the spectral verification value of the pixel. Step 2: Calculate the average value and variance of the spectral verification values of all pixels in the pixel group to obtain the verification mean and verification variance; calculate the ratio of the difference between the spectral verification value and the verification mean to the verification variance to obtain the verification coefficient YZ of the pixels in the pixel group. Step 3: Compare the verification coefficient YZ of the pixels in the pixel group with the preset verification threshold YZmax: If the verification coefficient YZ is less than the verification threshold YZmax, it is determined that the pixel belongs to the pixel group and no processing is required; if the verification coefficient YZ is greater than or equal to the verification threshold YZmax, it is determined that the pixel does not belong to the pixel group, and the corresponding pixel is removed from the pixel group and marked as an independent pixel.

6. The remote sensing survey image processing system based on AI prospecting technology according to claim 5 is characterized in that: The specific process of standard matching is as follows: Select a target mineral from the standard mineral spectral library, and obtain the characteristic bands and spectral standard vector BZ of the target mineral; extract the spectral reflectances FSn corresponding to the characteristic bands of the target mineral in the pixel and form a spectral feature vector TZ; divide the dot product result of the spectral feature vector TZ of the pixel and the spectral standard vector BZ of the target mineral by the product of the norms of the two vectors to obtain the cosine value, and take the inverse cosine of the cosine value to obtain the angle between the two vectors and mark it as the matching angle θ; compare the matching angle θ with the preset matching angle threshold θmax: If the matching angle θ is less than the matching angle threshold θmax, it is determined that the matching is successful, and the corresponding target mineral is marked as the matching mineral of the pixel; if the matching angle θ is greater than or equal to the matching angle threshold θmax, it is determined that the matching fails, and a new target mineral is selected from the standard mineral spectral library to perform the matching again until all matching is completed.

7. The remote sensing survey image processing system based on AI prospecting technology according to claim 6 is characterized in that: The specific process of fast matching is as follows: Perform standard matching on the first pixel and the last pixel in the pixel group respectively. If the results of the two standard matchings are the same target mineral, it is determined that the matching is successful, and the target mineral is marked as the matching mineral of all pixels in the pixel group. If the results of the two standard matchings are not the same target mineral, the pixel group is evenly divided into two new pixel groups, and standard matching is performed on the first pixel and the last pixel in the new pixel groups respectively, and the above operations are repeated until the results of the two standard matchings in the pixel group are the same target mineral.

Images