CN114881892A - Remote sensing image characteristic discretization method and device based on II-type fuzzy rough model - Google Patents

Abstract

The invention provides a method and device for discretizing remote sensing image features based on type II fuzzy rough model. The method includes: acquiring target remote sensing image data, extracting mixed pixels from the target remote sensing image data, and each mixed pixel contains multiple Spectral response characteristics of species types; determine the primary membership degrees of each mixed pixel corresponding to each feature type according to the mixed pixels; calculate the secondary membership degrees of each mixed pixel belonging to each feature type according to the primary membership; and the sub-membership degree to determine the type II fuzzy rough set of each object type; perform feature discretization processing on the target remote sensing image data to obtain the optimal discretization result. In the embodiment of the present invention, the primary membership degree and the secondary membership degree corresponding to the mixed pixels are used to describe the fuzzy components in the discretization process of remote sensing image features, the primary membership degree is used to blur the discretization process, and the primary membership degree is further fuzzified by the secondary membership degree. , to accurately quantify the uncertainty of mixed pixels and obtain accurate discretization results.

Description

Remote sensing image feature discretization method and device based on type II fuzzy rough model

technical field

The invention relates to the field of remote sensing image feature extraction, in particular to a remote sensing image feature discretization method and device based on a type II fuzzy rough model.

Background technique

As an advanced technical means, remote sensing has been widely used in various fields of economic and social development. With the gradual improvement of spatial resolution, temporal resolution, spectral resolution and radiometric resolution, remote sensing data has obvious big data characteristics. Due to the diversity, interpenetration and complexity of the spatial distribution of ground features, the spectral signal of each pixel in the remote sensing image records different types of land cover, and these pixels are called mixed pixels. As one of the most influential data preprocessing techniques, feature discretization plays an important role in knowledge discovery and data mining which are widely used in industrial control. It can convert continuous features into discrete features closer to the representation of the knowledge layer, making the data easier to understand, use and interpret, thereby improving the efficiency of remote sensing data processing and adapting to those learning algorithms that require discrete data as input. In the related art, in order to realize the discretization of remote sensing image features, the feature discretization algorithm of remote sensing image is usually based on the assumption that a sample belongs to only a single category, and cannot describe the uncertainty caused by mixed pixels. Or in order to simplify the operation of type II fuzzy sets, the sub-membership of mixed pixels is defined as a constant. Although the fuzzy rough model quantifies the uncertainty information by introducing the membership degrees of pixels to various categories, the decomposition model of mixed pixels has a large error, which is inconsistent with the distribution information of the data and cannot accurately describe the uncertainty of the data. , resulting in a decrease in data accuracy.

SUMMARY OF THE INVENTION

Therefore, the technical problem to be solved by the present invention is to overcome the defect that the feature discretization algorithm in the prior art cannot accurately quantify and evaluate the uncertainty caused by the mixed pixels, so as to provide a remote sensing image feature based on the type II fuzzy rough model. Discretization method and apparatus.

According to a first aspect, an embodiment of the present invention provides a method for discretizing remote sensing image features based on a type II fuzzy rough model, including the following steps: acquiring target remote sensing image data, extracting mixed pixels from the target remote sensing image data, each mixed The pixels contain the spectral response characteristics of various types of ground objects; the primary membership degrees of each mixed pixel corresponding to each feature type are determined according to the mixed pixels; the secondary membership degrees of each mixed pixel belonging to each feature type are calculated according to the primary membership degrees. ; According to the primary membership degree and the secondary membership degree, determine the type II fuzzy rough set of each object type; perform feature discretization processing on the target remote sensing image data, and obtain the optimal discretization result.

Optionally, determining the primary membership degree of each mixed pixel corresponding to each object type according to the mixed pixel includes: iteratively calculating the fuzzy mean vector and the fuzzy covariance matrix of the preset fuzzy segmentation matrix, and the preset fuzzy segmentation matrix is composed of the mixed pixel. The membership degree corresponding to each feature type is composed; according to the fuzzy segmentation matrix when the iteration termination condition is satisfied, the main membership degree of each mixed pixel corresponding to each feature type is determined.

Optionally, determining the primary membership degree of each mixed pixel corresponding to each feature type according to the iterative calculation of the fuzzy segmentation matrix when the iterative termination condition is satisfied, including: determining the corresponding abundance of each mixed pixel according to the fuzzy segmentation matrix, and calculating the corresponding abundance of each mixed pixel. Abundance is used as the main membership degree of each mixed pixel corresponding to each feature type.

Optionally, calculating the secondary membership degree of each mixed pixel belonging to each feature type according to the primary membership degree includes: determining the hard segmentation matrix according to the iterative calculation of the fuzzy segmentation matrix when the iteration termination condition is satisfied; determining the attribution according to the hard segmentation matrix The set composed of pixels of each object type; calculate the upper approximation, lower approximation, positive domain, negative domain, and boundary domain of the set in the approximate space; determine each The sub-membership of mixed pixels attributable to each object type.

Optionally, performing feature discretization processing on the target remote sensing image data to obtain an optimal discretization result, including: obtaining an initial breakpoint set of mixed pixels from the remote sensing image data; initializing the target based on the number of breakpoints in the initial breakpoint set Remote sensing image data population; iteratively execute the genetic algorithm on the individuals of the target remote sensing image data population to determine the optimal discretization result; wherein, the discretization scheme corresponding to the initialized target remote sensing image data population is the initial discretization scheme, and each Population individuals correspond to a discretization result.

Optionally, the genetic algorithm is iteratively executed on the individuals of the target remote sensing image data population to determine the optimal discretization result, including: determining the fuzzy relationship between the mixed pixels based on the Euclidean distance between the mixed pixels; according to the fuzzy relationship, calculating II The average approximation accuracy of the fuzzy rough set; according to the number of breakpoints in the initial breakpoint set, determine the reduction range of the number of breakpoints corresponding to the target remote sensing image data population; according to the reduction range of the number of breakpoints and the average approximate precision, determine The fitness function of the type II fuzzy rough set; according to the fitness function of the type II fuzzy rough set, the fitness value of the type II fuzzy rough set is determined, and the individual of each target remote sensing image data population corresponding to the optimal fitness value is taken as the Optimal discretization result.

Optionally, the fitness function of the type II fuzzy rough set is expressed by the following formula:

，

,

为平均近似精度。 where α and β are the weight coefficients, | D | is the magnitude of the reduction in the number of breakpoints,

is the average approximate precision.

According to a second aspect, an embodiment of the present invention provides a remote sensing image feature discretization device based on a type II fuzzy rough model, comprising: a mixed pixel extraction unit configured to acquire target remote sensing image data, and obtain target remote sensing image data from the target remote sensing image data. Extracting mixed pixels, each mixed pixel contains spectral response characteristics of multiple types of ground objects; the primary membership determination unit is configured to determine the primary membership of each mixed pixel corresponding to each feature type according to the mixed pixels; the secondary membership The degree determination unit is configured to calculate the secondary membership degree of each mixed pixel belonging to each feature type according to the primary membership degree; the fuzzy rough set determination unit is configured to determine the secondary membership degree of each feature type according to the primary membership degree and the secondary membership degree. The optimal discretization result determination unit is configured to perform feature discretization processing on the target remote sensing image data to obtain the optimal discretization result.

According to a third aspect, an embodiment of the present invention provides a non-transitory computer-readable storage medium, where the non-transitory computer-readable storage medium stores computer instructions, and when the computer instructions are executed by a processor, any one of the first aspect is implemented. The method for discretizing remote sensing image features based on the type II fuzzy rough model described in the embodiment.

According to a fourth aspect, an embodiment of the present invention provides a computer device, comprising at least one processor; and a memory connected in communication with the at least one processor; wherein the memory stores instructions executable by the at least one processor, The instructions are executed by at least one processor, so as to execute the method for discretizing remote sensing image features based on a type II fuzzy rough model according to any embodiment of the first aspect.

The technical scheme of the present invention has the following advantages:

The present invention provides a method and device for discretizing remote sensing image features based on type II fuzzy rough model. The method includes: acquiring target remote sensing image data, extracting mixed pixels from the target remote sensing image data, and each mixed pixel contains multiple Spectral response characteristics of species types; determine the primary membership degrees of each mixed pixel corresponding to each feature type according to the mixed pixels; calculate the secondary membership degrees of each mixed pixel belonging to each feature type according to the primary membership; and the sub-membership degree to determine the type II fuzzy rough set of each object type; perform feature discretization processing on the target remote sensing image data to obtain the optimal discretization result. In the embodiment of the present invention, the rough set and the fuzzy set are combined to describe the fuzzy components in the process of discretization of remote sensing image features by the primary and secondary degrees of membership corresponding to the mixed pixels. The membership degree further fuzzifies the main membership degree, so as to accurately quantify and evaluate the uncertainty of mixed pixels, and obtain more accurate discretization results.

Description of drawings

In order to illustrate the specific embodiments of the present invention or the technical solutions in the prior art more clearly, the following briefly introduces the accompanying drawings that need to be used in the description of the specific embodiments or the prior art. Obviously, the accompanying drawings in the following description The drawings are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings without creative efforts.

1 is a flowchart of a specific example of a method for discretizing remote sensing image features based on a Type II fuzzy rough model in Embodiment 1 of the present invention;

2 is an analysis diagram of a specific example of a method for discretizing remote sensing image features based on a type II fuzzy rough model in Embodiment 1 of the present invention;

3 is a schematic diagram of a structure of a remote sensing image feature discretization device based on a type II fuzzy rough model in Embodiment 2 of the present invention;

FIG. 4 is a structural example diagram of a computer device in Embodiment 4 of the present invention.

Detailed ways

The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are a part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc. The indicated orientation or positional relationship is based on the orientation or positional relationship shown in the accompanying drawings, which is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the indicated device or element must have a specific orientation or a specific orientation. construction and operation, and therefore should not be construed as limiting the invention. Furthermore, the terms "first", "second", and "third" are used for descriptive purposes only and should not be construed to indicate or imply relative importance.

In the description of the present invention, fuzzy rough model is a more powerful uncertainty data analysis model than fuzzy set and rough set. On the basis of rough set, fuzzy set is introduced, and the correlation between samples is described by replacing the equivalence relationship of rough set with similarity relationship. As a generalization of fuzzy rough model, type II fuzzy rough model can provide more accurate uncertainty analysis ability. The type II fuzzy rough model fuzzifies the membership function value of the fuzzy set again, so that the fuzzy phenomenon can be described more profoundly.

In the description of the formula in the present invention, exp refers to an exponential function whose base is a natural constant e in advanced mathematics. inf stands for infimum, which is the largest lower bound of a set. sup stands for supremum, which is the least upper bound of a set.

In addition, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

Example 1

This embodiment provides a method for discretizing remote sensing image features based on a type II fuzzy rough model, as shown in FIG. 1 , including the following steps:

S11: Acquire target remote sensing image data, extract mixed pixels from the target remote sensing image data, and each mixed pixel contains spectral response features of multiple ground object types.

Specifically, the components of the mixed pixel spectral signal are called endmembers, and each endmember corresponds to the spectral response feature of a ground object type.

S12: Determine the master membership degree of each mixed pixel corresponding to each object type according to the mixed pixel.

Specifically, the primary membership degree of each corresponding ground object type is determined according to the mixed pixels by iterative calculation of the fuzzy segmentation matrix; when the iterative calculation satisfies the iterative termination condition, the corresponding abundance of each mixed pixel is determined, and the abundance It is used as the main membership degree of each mixed pixel corresponding to each feature type. Among them, the abundance corresponding to each mixed pixel refers to the abundance of the endmembers of the mixed pixel.

In practical applications, the fuzzy segmentation matrix is composed of the membership degrees of each mixed pixel to the number of categories of the classification scheme. The fuzzy mean vector and fuzzy covariance matrix in the fuzzy partition matrix can be represented by the above membership degrees.

S13: Calculate the secondary membership degree of each mixed pixel belonging to each object type according to the primary membership degree.

Specifically, calculating the sub-membership degree of each mixed pixel belonging to each object type refers to determining the sub-membership degree according to the distribution of the mixed pixel in the rough set boundary area. The distribution of mixed pixels in the rough set boundary area includes determining the set of pixels belonging to each feature type; calculating the distribution area of the set in the approximate space to determine the sub-membership of the mixed pixels belonging to each feature type. Wherein, the distribution area of the set in the approximate space includes: the upper approximation, the lower approximation, the positive domain, the negative domain, and the boundary domain of the set in the approximate space.

In practical applications, the approximate space refers to the rough approximate space ( U , T ), where U represents the set of mixed pixels and T represents the number of bands of remote sensing images.

S14: Determine the type II fuzzy rough set of each feature type according to the primary membership degree and the secondary membership degree.

Specifically, in the process of determining the primary membership degree of each mixed pixel corresponding to each feature type and the secondary membership degree of each mixed pixel belonging to each feature type, the primary membership degree and the secondary membership degree are used to describe the feature dispersion of remote sensing images. Fuzzy components in the process of fuzzification, the discretization process is fuzzified by the primary membership degree, and the primary membership degree is further fuzzified by the secondary membership degree. Through the determined primary membership degree and secondary membership degree, the type II fuzzy rough set of each feature type is determined. .

S15: Perform feature discretization processing on the target remote sensing image data to obtain an optimal discretization result.

Specifically, discretization is to adopt a specific method to divide continuous features into multiple sub-intervals, and associate multiple sub-intervals with candidate breakpoints. Therefore, the feature discretization processing of target remote sensing images can be regarded as the selection of candidate breakpoints. The process of discretizing the target remote sensing image data to obtain the optimal discretization result is to iteratively select the candidate breakpoints through the genetic algorithm; Approximate accuracy, determine the fitness function of the individuals in the population; evaluate the discretization result with the fitness function of the individuals in the determined population, and obtain the optimal discretization result.

In practical applications, the initial discretization scheme is determined by obtaining the initial breakpoint set of mixed pixels in remote sensing image data.

The present invention provides a method for discretizing remote sensing image features based on type II fuzzy rough model. The method includes: acquiring target remote sensing image data, extracting mixed pixels from the target remote sensing image data, and each mixed pixel contains a variety of ground The spectral response characteristics of the object type; determine the primary membership degree of each mixed pixel corresponding to each feature type according to the mixed pixel; calculate the secondary membership degree of each mixed pixel belonging to each feature type according to the primary membership degree; The membership degree is used to determine the type II fuzzy rough set of each object type; the feature discretization process is performed on the target remote sensing image data to obtain the optimal discretization result. In the embodiment of the present invention, the rough set and the fuzzy set are combined to describe the fuzzy components in the process of discretization of remote sensing image features by the primary and secondary degrees of membership corresponding to the mixed pixels. The membership degree further fuzzifies the main membership degree, so as to accurately quantify and evaluate the uncertainty of mixed pixels, and obtain more accurate discretization results.

In an optional embodiment of the present invention, in the above step S12, determining the primary membership degree of each mixed pixel corresponding to each feature type according to the mixed pixel includes:

(1) Iteratively calculate the fuzzy mean vector and the fuzzy covariance matrix of the preset fuzzy segmentation matrix. The preset fuzzy segmentation matrix is composed of the membership degrees of the mixed pixels corresponding to each object type;

Specifically, the preset fuzzy segmentation matrix can be expressed by the following formula:

，

,

Among them, F _s ( X _k ) represents the membership degree of the k -th pixel X _k in U to the category s , s ∈{1,2,…, g }, g denotes the number of categories, k ∈{1,2,…, n }, n represents the number of pixels in U.

，q∈{1, 2,…,g}。 In practical applications, F _s ( X _k ) satisfies: 0≤ F _s ( X _k )≤1,

, q ∈ {1, 2,…, g }.

Specifically, the fuzzy mean vector of the preset fuzzy segmentation matrix can be expressed by the following formula:

，

,

表示模糊均值，i∈{1,2,…,m}，m表示波段个数， in,

represents the fuzzy mean, i ∈{1,2,…, m }, m represents the number of bands,

，w表示权 重，w大于等于1，x _ik表示第k像元在第i波段上的像元值。

, w represents the weight, w is greater than or equal to 1, x _ik represents the pixel value of the k -th pixel on the i -th band.

Specifically, the fuzzy covariance matrix of the preset fuzzy segmentation matrix can be expressed by the following formula:

。 where σ _mms represents the fuzzy covariance, j ∈ {1,2,…, m },

.

Specifically, iteratively calculating the fuzzy mean vector and the fuzzy covariance matrix of the preset fuzzy segmentation matrix includes determining the preset fuzzy segmentation matrix from the fuzzy mean vector and the fuzzy covariance matrix.

In practical applications, the preset fuzzy segmentation matrix is composed of the membership degree of the pixel to the category. The membership degree of the pixel to the category can be determined by the fuzzy mean vector and the fuzzy covariance matrix. The membership degree of the pixel to the category can be determined by the following formula Express:

，

,

Among them, P `( s ) is the prior probability of the occurrence of the s -th category,

。

.

In practical applications, the determination of the prior probability of the occurrence of the s th category belongs to the relatively mature prior art, which will not be repeated in this application.

(2) Determine the master membership degree of each mixed pixel corresponding to each feature type according to the iterative calculation of the fuzzy segmentation matrix when the iterative termination condition is satisfied.

Specifically, the iteration termination condition can be expressed by the following formula:

，

,

Among them, θ is the number of iterative steps of the algorithm, and ε is the error threshold.

In an optional embodiment of the present invention, determining the primary membership degree of each mixed pixel corresponding to each object type according to the iterative calculation of the fuzzy segmentation matrix when the iteration termination condition is satisfied, includes: determining each mixed pixel according to the fuzzy segmentation matrix The corresponding abundance is taken as the main membership degree of each mixed pixel corresponding to each feature type.

，其 中e为自然数。对于各混合像元对应各地物类型的主隶属度可按如下公式表达： Specifically, the determination of the master membership of each mixed pixel corresponding to each feature type is related to the weight and the abundance of each endmember in the mixed pixel.

, where e is a natural number. The master membership degree of each mixed pixel corresponding to each feature type can be expressed by the following formula:

，

,

，

,

表示各混合像元归属于各地物类 型的次隶属度，J _x表示主隶属度的取值范围，u表示一个主隶属度，

表示各混合像元对 应各地物类型的主隶属度的最小值，

表示各混合像元对应各地物类型的主隶属度 的最大值。 Among them, P _s ( x ) represents the abundance corresponding to the mixed pixel x ,

represents the secondary membership degree of each mixed pixel belonging to each object type, J _x represents the value range of the primary membership degree, u represents a primary membership degree,

represents the minimum value of the primary membership of each mixed pixel corresponding to each object type,

Indicates the maximum value of the primary membership of each mixed pixel corresponding to each object type.

。应该理 解的是，关于w与e的取值包括但不限定于实施例中描述方式，w与e的取值只要可用于保证 离散化结果的稳定性并且降低复杂度即可。 In practical applications, e determines the number of weights in w . The larger the value of e , the more elements P _s ( x ) contains, that is, the larger the value range of the primary membership degree. At the same time, it also brings A lot of computation. The weight w not only controls the convexity and concavity of the discretization result, but also controls the sharing degree of the mixed pixels among the various types. The research results show that when the weight is 2, the effect of clustering can reach the best with a larger probability. If the value of e is greater than 1, different weights will be introduced, and the results of fuzzy clustering can be considered more comprehensively, but the clustering quality brought by the introduction of weights cannot be guaranteed, making the discretization results inaccurate and unstable. Therefore, in order to ensure the stability of the discretization result and reduce the complexity, in an optional implementation manner, e is selected to be 1 and w is selected to be 2. at this time,

. It should be understood that the values of w and e include but are not limited to the methods described in the embodiments, and the values of w and e may be used to ensure the stability of the discretization result and reduce the complexity.

，各混合像元对应各地物类型的 主隶属度可按如下公式表达： In practical applications, for

, the master membership degree of each mixed pixel corresponding to each feature type can be expressed by the following formula:

，

,

。

.

In an optional embodiment of the present invention, the fuzzy partition matrix is determined by iteratively calculating the fuzzy mean vector and the fuzzy partition matrix. The abundance corresponding to the mixed pixels is determined by the fuzzy segmentation matrix, and the abundance is used as the main membership degree of each feature type corresponding to the mixed pixels. In this way, the process of fuzzification and discretization based on the master membership degree is realized.

In an optional embodiment of the present invention, in the above step S13, calculating the secondary membership degree of each mixed pixel belonging to each feature type according to the primary membership degree, including:

(1) Determine the hard partition matrix according to the fuzzy partition matrix when the iteration termination condition is satisfied.

Specifically, the determination of the hard partition matrix is to modify the maximum value of each column in the fuzzy partition matrix when the iteration termination condition is satisfied to 1, and other values in the corresponding column to be modified to 0, thereby completing the determination of the hard partition matrix.

(2) According to the hard segmentation matrix, determine the set of pixels that belong to each object type.

Specifically, the set of pixels belonging to various object types can be expressed as follows:

，

,

Among them, Xs represents the set of pixels belonging to the sth type of ground object in the mixed pixel set, and Cs(x) is the membership degree of the pixel x in the hard segmentation matrix C to the sth category.

In practical applications, the hard segmentation matrix has determined the corresponding position of the membership value of 1 in each column, and determines whether the mixed pixel in the mixed pixel set U belongs to the corresponding position of the determined membership value of 1 in each column. According to each object type, the set of pixels belonging to each object type can be determined according to the hard segmentation matrix.

(3) Calculate the upper approximation, lower approximation, positive domain, negative domain, and boundary domain of the set in the approximate space.

Specifically, calculating the upper approximation, lower approximation, positive domain, negative domain, and boundary domain of the set X _s composed of pixels belonging to various object types in the approximate space includes: determining the equivalence class set of the mixed pixel set; The determined equivalence class set calculates the upper approximation, lower approximation, positive domain, negative domain, and boundary domain of the set composed of pixels belonging to each object type in the approximate space.

Specifically, the equivalence class set of the mixed pixel set can be expressed by the following formula:

，

,

表示任意像元x及任意像元y属于混合像元X，

表示任意波段t 存在与像元x、像元y分别对应的波段等价关系。 Among them, U | IND ( T ) represents the equivalence class set of the mixed pixel set, T ={ t ₁ ,…,t _m },

Indicates that any pixel x and any pixel y belong to the mixed pixel X ,

Indicates that any band t has band equivalence relation corresponding to pixel x and pixel y respectively.

Specifically, the upper approximation of X _s in the approximation space can be calculated as follows:

，

,

where T ^* ( X _s ) represents the upper approximation of X _s in the approximation space.

Specifically, the lower approximation of X _s in the approximation space can be calculated as follows:

，

,

where T _* ( X _s ) represents the lower approximation of the set in the approximation space.

Specifically, the positive field of X _s in the approximate space can be calculated as follows:

，

,

Among them, POST ( _{X s} ) represents the positive field of the set in the approximate space.

Specifically, the negative field of X _s in the approximate space can be calculated as follows:

，

,

where NGT _T ( X _s ) represents the negative field of the set in the approximate space.

Specifically, the boundary domain of X _s in the approximate space can be calculated as follows:

，

,

Among them, BNT ( _{X s} ) represents the boundary domain of the set in the approximate space.

(4) According to the upper approximation, the lower approximation, the positive domain, the negative domain, and the boundary domain, determine the sub-membership degree of each mixed pixel belonging to each object type.

Specifically, the sub-membership degrees of each mixed pixel belonging to each object type are determined according to the upper approximation, the lower approximation, the positive domain, the negative domain, and the boundary domain, including: according to the upper approximation, the lower approximation, the positive domain, the negative domain, and the boundary domain , determine the probability that the pixel belongs to the set composed of the pixels of each object type; according to the determined probability that the pixel belongs to the set composed of the pixels of each object type, determine the sub-membership of each mixed pixel belonging to each object type Spend.

Specifically, according to the upper approximation, the lower approximation, the positive domain, the negative domain, and the boundary domain, determining the probability that a pixel belongs to the set of pixels of each object type refers to the upper approximation, the lower approximation, the positive domain, the negative domain, the The boundary domain determines the distribution of pixels; the determined distribution, according to Bayes' theorem, determines the probability that the pixels belong to the set of pixels of various object types.

Exemplarily , POST _{(X s ) is a set of pixels belonging to X s in the mixed pixel set, and NGT T} ( X _s ) _is a _set of _pixels that do not belong to X _s in the mixed pixel set , BNT ( _{X s} ) is a set of pixels that cannot be determined to belong to X s _in the mixed pixel set. Therefore, BNT ( _{X s} ) is the uncertainty domain of mixed cells . As shown in FIG. 2 , a specific example analysis diagram of the remote sensing image feature discretization method based on the type II fuzzy rough model in the embodiment of the present invention, each rectangle in the figure represents an equivalence class in U | IND ( T ). The circular area corresponding to reference numeral 22 represents X _s , the octagonal area corresponding to reference numeral 24 represents T ^* ( X _s ), and the rectangular area corresponding to reference numeral 23 represents T _* ( X _s ) or POST ( X _{s )} . The area outside the octagon corresponding to the reference numeral 24 represents NGT _T ( X _s ), the area within the octagon corresponding to the reference numeral 24, and the area outside the rectangle corresponding to the reference numeral 23 represents B N _T ( X _s ). When the pixel x appears in the positive or negative domain, the relationship between x and X _s is deterministic, and when x appears in the boundary domain, the relationship between x and X _s is uncertain. That is, the distribution of pixels is determined according to the upper approximation, the lower approximation, the positive domain, the negative domain, and the boundary domain.

Specifically, according to the determined distribution, according to Bayes' theorem, determine the probability that the pixel belongs to the set composed of the pixels of each object type, including: the probability that the pixel belongs to the set composed of the pixels of each object type can be It is expressed by the following formula:

，

,

Among them, E [ i ] represents the ith equivalence class on the pixel set, G [ s ] represents the s th class, and P ( G [ s ]| E [ i ]) represents the image of class s in the equivalence class i The proportion of the element, P ( E [ i ] | G [ s ]) represents the proportion of the pixels belonging to the equivalence class i among all the pixels of the category s , and P ( E [ i ]) represents etc. The probability that a pixel in valence class i appears in the mixed pixel set U , P ( G [ s ]) represents the probability that a pixel of class s appears in U.

Specifically, according to the determined probability that the pixel belongs to the set composed of the pixels of each object type, the sub-membership degree of each mixed pixel belonging to each object type is determined, including: the sub-membership of each mixed pixel belonging to each object type The degree of membership can be expressed by the following formula:

。

.

In an optional embodiment of the present invention, the upper approximation, the lower approximation, the positive domain, the negative domain, and the boundary domain of the set in the approximate space are calculated by determining the set composed of the pixels of each feature type; Approximation, positive domain, negative domain, boundary domain, determine the probability that the pixel belongs to the set composed of the pixels of each object type; The sub-membership of the pixel attributable to each object type. In this process, since the discretization scheme is equivalent to dividing the mixed pixel set into multiple equivalence classes, elements in the same equivalence class have the same attribute value. Then, the probability that mixed pixels in the same equivalence class belong to each class can be considered to be the same, while for mixed pixels in different equivalence classes, their probability of belonging to each class will be different. The uncertainty brought about by the difference is a further fuzzification of the master membership. Therefore, the process of determining the secondary membership degree in the technical solution of the present invention is equivalent to further fuzzifying the primary membership degree through the secondary membership degree, so as to accurately quantify and evaluate the uncertainty of mixed pixels.

，各地物类型的II型模糊粗糙集通过如下公式表达：In an optional embodiment of the present invention, in the foregoing step S14, for

, the type II fuzzy rough set of each object type is expressed by the following formula:

。

.

In an optional embodiment of the present invention, in the above step S15, feature discretization processing is performed on the target remote sensing image data to obtain an optimal discretization result, including:

(1) Obtain the initial breakpoint set of mixed pixels from remote sensing image data;

Specifically, obtaining the initial breakpoint set of mixed pixels from remote sensing image data refers to obtaining the initial breakpoint set from the input remote sensing image features. The acquisition of the initial breakpoint set belongs to a relatively mature prior art, which will not be repeated in this application.

(2) Initialize the target remote sensing image data population based on the number of breakpoints in the initial breakpoint set;

Specifically, discretization is to adopt a specific method to divide continuous features into multiple sub-intervals, and associate multiple sub-intervals with candidate breakpoints. Therefore, the feature discretization processing of target remote sensing images can be regarded as the selection of candidate breakpoints, and each discretization scheme corresponds to a division result on the mixed pixel set. The initial target remote sensing image data population refers to the number of breakpoints in the initial breakpoint set as the individual length of the initial population, so as to complete the initialization of the target remote sensing image data population.

Exemplarily, it is assumed that the number of initial breakpoints is 10. Since each population individual is coded in binary, the length of the individual is the number of initial breakpoints, that is, the length of each population individual is 10 bits, corresponding to the initial breakpoint set. 10 breakpoints.

(3) Iteratively execute the genetic algorithm on the individuals of the target remote sensing image data population to determine the optimal discretization result; among them, the discretization scheme corresponding to the initialized target remote sensing image data population is the initial discretization scheme, and each population individual corresponds to a discretization result.

Specifically, the discretization scheme corresponding to the individuals in the initial population is the initial discretization scheme, and the discretization result corresponding to the initial discretization scheme is the initial discretization result.

Exemplarily, it is assumed that the number of individuals in the target remote sensing image data population is 50, and there are 50 discretization schemes corresponding to the population size of the target remote sensing image data. In each iteration, these 50 individuals in the population will undergo selection, mutation , Crossover, that is, the evolution function of the genetic algorithm to update the population, so that the 50 individuals in the next generation of the population are different from the 50 individuals in the previous generation.

In an optional embodiment of the present invention, the genetic algorithm is iteratively executed on the individuals of the target remote sensing image data population to determine the optimal discretization result, including:

(1) Determine the fuzzy relationship between mixed pixels based on the Euclidean distance between mixed pixels;

Specifically, the fuzzy relationship between mixed pixels can be expressed by the following formula:

，

,

，

,

where d ( x , y ) represents the Euclidean distance between x and y , and x _h and y _h represent the pixel values of x and y on the h -band, respectively.

(2) Calculate the average approximation accuracy of Type II fuzzy rough sets according to the fuzzy relationship;

In an optional embodiment of the present invention, calculating the average approximation accuracy of the type II fuzzy rough set according to the fuzzy relationship includes: calculating the upper approximation and the lower approximation of the type II fuzzy rough set according to the fuzzy relationship; according to the determined upper approximation With the following approximation, the average approximation accuracy of the Type II fuzzy rough set is calculated.

Specifically, the upper approximation of the type II fuzzy rough set can be expressed by the following formula:

，

,

， in,

,

，u2为主隶属度中的一个，

, u 2 is one of the master membership degrees,

，J _y2表示对应隶属度的取值 范围，a(y2)表示混合像元y2的次隶属度，

和

分别是

的 主隶属度最小值和最大值。

, J _y2 represents the value range of the corresponding membership degree, a ( y 2 ) represents the secondary membership degree of the mixed pixel y 2 ,

and

respectively

The minimum and maximum values of primary membership.

Specifically, the lower approximation of the type II fuzzy rough set can be expressed by the following formula:

，

,

， in,

,

，u1为主隶属度中的一个，

, u 1 is one of the master membership degrees,

，J _y1表示对应隶 属度的取值范围，a(y1)表示混合像元y1的次隶属度。

, J _y1 represents the value range of the corresponding membership degree, a ( y 1 ) represents the secondary membership degree of the mixed pixel y 1 .

Specifically, the average approximation accuracy of the type II fuzzy rough set can be expressed by the following formula:

，

,

为平均近似精度，

表示II型模糊粗糙集的 近似精度。 in,

is the average approximation accuracy,

Represents the approximate accuracy of a type II fuzzy rough set.

Specifically, the approximate accuracy of the type II fuzzy rough set can be expressed by the following formula:

Among them, x , y 1, y 2 ∈ U .

，则II 型模糊粗糙集的近似精度可按如下公式表达： Exemplarily, for

, then the approximate accuracy of the type II fuzzy rough set can be expressed by the following formula:

。

.

(3) According to the number of breakpoints in the initial breakpoint set, determine the magnitude of the reduction in the number of breakpoints corresponding to the target remote sensing image data population individuals;

Specifically, according to the number of breakpoints in the initial breakpoint set, determine the extent of reduction in the number of breakpoints corresponding to the target remote sensing image data population individuals, including: determining the length of the individual as the number of initial breakpoints; determining the length of the individual selected The number is the number of breakpoints corresponding to the individual; the difference between the length of the individual and the number of individual breakpoints is determined as the reduction in the number of breakpoints corresponding to the target remote sensing image data population.

Exemplarily, it is assumed that the number of initial breakpoints is 10. Since each population individual is coded in binary, the length of the individual is the number of initial breakpoints, that is, the length of each population individual is 10 bits, corresponding to the initial breakpoint set. 10 breakpoints. Each bit in the binary code corresponds to a candidate breakpoint, and the values '1' and '0' represent that the breakpoint is selected and not selected, respectively. For each population individual, the number of selected binary codes is determined as the number of breakpoints of the corresponding individual, that is, the number of binary bits with a value of '1' can be determined to obtain the number of breakpoints of the individual. The difference between the length of the individual and the number of individual breakpoints is determined as the magnitude of the reduction in the number of breakpoints of the corresponding individual, that is, subtracting the number of individual breakpoints from 10 is equal to the magnitude of the reduction in the number of individual breakpoints. For example, the binary code of an individual is 1110000111, the number of breakpoints corresponding to the individual is 6, and the magnitude of the reduction in the number of individual breakpoints is 4.

Specifically, the magnitude of the reduction in the number of breakpoints refers to the amount of reduction in the number of individual breakpoints of the population in each iteration process. The discrete effect is measured by comparing the number of population breakpoints reduced in the iterative process. The greater the reduction in the number of breakpoints, the better the discrete effect.

(4) Determine the fitness function of type II fuzzy rough set according to the reduction of the number of breakpoints and the average approximation accuracy;

Specifically, the fitness function of the type II fuzzy rough set is used to calculate the fitness value of each target remote sensing image data population. and composition.

In an optional embodiment of the present invention, the fitness function of the type II fuzzy rough set is expressed by the following formula:

，

,

为平均近似 精度。 where α and β are the weight coefficients, | D | is the magnitude of the reduction in the number of breakpoints,

is the average approximate precision.

Specifically, the selection of the weight coefficient is selected according to the actual working conditions, and the rationality of the weight setting is generally judged according to the characteristics of the data set and experimental observation, which is not specifically limited in this application.

(5) Determine the fitness value of the type II fuzzy rough set according to the fitness function of the type II fuzzy rough set, and take the individual of each target remote sensing image data population corresponding to the optimal fitness value as the optimal discretization result .

Specifically, multiple discretization schemes are used as the population individuals in the genetic algorithm, and the individual with the largest fitness value is iteratively calculated through the evolution function of the genetic algorithm, and the individual with the largest fitness value is used as the optimal fitness value. individual. The discretization scheme corresponding to the optimal fitness value is the optimal discretization scheme.

Exemplarily, assuming that there are 50 individuals in the target remote sensing image data population, the fitness function corresponding to the population individuals has 50 values. In each iteration, these 50 individuals in the population undergo evolution to update the population. So that the 50 individuals in the next generation are different from the 50 individuals in the previous generation. In each iteration, the global variable records the individual with the highest fitness value among the 50 individuals. When the fitness value of an individual in the next generation is higher than that of the individual recorded in the global variable, the global variable is updated with an individual with a higher fitness value. After all iterations are completed, the global variable records the optimal individual, and the discretization scheme corresponding to the optimal individual is the optimal discretization scheme.

In an optional embodiment of the present invention, the fitness function of the type II fuzzy rough set is constructed by calculating the average approximation accuracy of the type II fuzzy rough set and determining the magnitude of the reduction in the number of breakpoints. And the individual with the optimal fitness value is determined by genetic algorithm, the fuzzy components in the process of discretization of remote sensing image features are described by the primary and secondary degrees of membership corresponding to the mixed pixels, and the discretization process is fuzzified by the primary membership degree. The primary membership degree is further fuzzified by the secondary membership degree, and the more accurate discretization results are obtained through the fitness function of the constructed II fuzzy rough set.

Example 2

This embodiment provides a remote sensing image feature discretization device based on a type II fuzzy rough model. As shown in FIG. 3 , FIG. 3 is a remote sensing image feature based on a type II fuzzy rough model provided by an optional embodiment of the present invention. The connection diagram of the discretization device includes: a mixed pixel extraction unit 31 , a primary membership determination unit 32 , a secondary membership determination unit 33 , a fuzzy rough set determination unit 34 , and an optimal discretization result determination unit 35 .

The mixed pixel extraction unit 31 is configured to acquire target remote sensing image data, and extract mixed pixels from the target remote sensing image data, and each mixed pixel contains spectral response features of multiple ground object types. For details, reference may be made to the relevant description of step S11 in any of the above method embodiments, which will not be repeated here.

The primary membership determination unit 32 is configured to determine the primary membership of each mixed pixel corresponding to each feature type according to the mixed pixel. For details, reference may be made to the relevant description of step S12 in any of the above method embodiments, which will not be repeated here.

The secondary membership degree determination unit 33 is configured to calculate the secondary membership degree of each mixed pixel belonging to each feature type according to the primary membership degree. For details, reference may be made to the relevant description of step S13 in any of the above method embodiments, which will not be repeated here.

The fuzzy rough set determining unit 34 is configured to determine the type II fuzzy rough set of each feature type according to the primary membership degree and the secondary membership degree. For details, reference may be made to the relevant description of step S14 in any of the above method embodiments, which will not be repeated here.

The optimal discretization result determination unit 35 is configured to perform feature discretization processing on the target remote sensing image data to obtain the optimal discretization result. For details, reference may be made to the relevant description of step S15 in any of the above method embodiments, which will not be repeated here.

The present invention provides a remote sensing image feature discretization device based on type II fuzzy rough model. The device includes: a mixed pixel extraction unit 31, which is configured to obtain target remote sensing image data, and extract mixed pixels from the target remote sensing image data. , and each mixed pixel contains the spectral response characteristics of multiple ground object types. The primary membership determination unit 32 is configured to determine the primary membership of each mixed pixel corresponding to each feature type according to the mixed pixel. The secondary membership degree determination unit 33 is configured to calculate the secondary membership degree of each mixed pixel belonging to each feature type according to the primary membership degree. The fuzzy rough set determining unit 34 is configured to determine the type II fuzzy rough set of each feature type according to the primary membership degree and the secondary membership degree. The optimal discretization result determination unit 35 is configured to perform feature discretization processing on the target remote sensing image data to obtain the optimal discretization result. In the embodiment of the present invention, the rough set and the fuzzy set are combined to describe the fuzzy components in the process of discretization of remote sensing image features by the primary and secondary degrees of membership corresponding to the mixed pixels. The membership degree further fuzzifies the main membership degree, so as to accurately quantify and evaluate the uncertainty of mixed pixels, and obtain more accurate discretization results.

In an optional embodiment of the present invention, the primary membership degree determination unit 32 includes: an iterative calculation subunit and a primary membership degree determination subunit. For details, please refer to the relevant description about determining the primary membership degree of each mixed pixel corresponding to each feature type according to the mixed pixel in any of the above method embodiments.

The iterative calculation subunit is configured to iteratively calculate the fuzzy mean vector and the fuzzy covariance matrix of the preset fuzzy segmentation matrix, wherein the preset fuzzy segmentation matrix is composed of the membership degrees of the mixed pixel corresponding to each object type.

The main membership degree determination subunit is configured to determine the main membership degree of each mixed pixel corresponding to each feature type according to the iterative calculation of the fuzzy segmentation matrix when the iteration termination condition is satisfied.

In an optional embodiment of the present invention, the sub-membership determination subunit, including the abundance determination subunit, is configured to determine the abundance corresponding to each mixed pixel according to the fuzzy segmentation matrix, and use the abundance as each mixed image The element corresponds to the master membership degree of each object type. For details, please refer to the relevant description of determining the primary membership degree of each mixed pixel corresponding to each feature type according to the iterative calculation of the fuzzy segmentation matrix when the iteration termination condition is satisfied in any of the above method embodiments.

In an optional embodiment of the present invention, the secondary membership determination unit 33 includes: a hard partition matrix determination subunit, a pixel set determination subunit, an approximate spatial boundary determination subunit, and a secondary membership determination subunit. For details, please refer to the relevant description about calculating the secondary membership degree of each mixed pixel belonging to each feature type according to the primary membership degree in any of the above method embodiments.

The hard partition matrix determination subunit is configured to determine the hard partition matrix according to the fuzzy partition matrix when the iteration termination condition is satisfied.

The pixel set determination subunit is configured to determine a set composed of pixels belonging to each object type according to the hard segmentation matrix.

The approximate space boundary determination subunit is configured to calculate the upper approximation, the lower approximation, the positive field, the negative field, and the boundary field of the set in the approximate space.

The sub-membership determination subunit is configured to determine the sub-membership of each mixed pixel belonging to each object type according to the upper approximation, the lower approximation, the positive domain, the negative domain, and the boundary domain.

In an optional embodiment of the present invention, the optimal discretization result determination unit 35 includes an initial breakpoint set acquisition subunit, a population initialization subunit, and an optimal discretization result determination subunit. For details, please refer to the relevant description about discretizing the target remote sensing image data to obtain the optimal discretization result in any of the above method embodiments.

The initial breakpoint set obtaining subunit is configured to obtain the initial breakpoint set of mixed pixels from the remote sensing image data.

The population initialization subunit is configured to initialize the target remote sensing image data population based on the number of breakpoints in the initial breakpoint set. Among them, the discretization scheme corresponding to the initialized target remote sensing image data population is the initial discretization scheme, and each population individual corresponds to a discretization result.

The optimal discretization result determination subunit is configured to iteratively execute the genetic algorithm on the individuals of the target remote sensing image data population to determine the optimal discretization result.

In an optional embodiment of the present invention, the subunit for determining the optimal discretization result includes: a subunit for determining the fuzzy relationship between pixels, a subunit for calculating, a subunit for determining the magnitude of the reduction in the number of breakpoints, and a subunit for determining a fitness function , the fitness value determines the subunit. For details, please refer to the relevant description about iteratively executing the genetic algorithm on the individuals of the target remote sensing image data population to determine the optimal discretization result in any of the above method embodiments.

The subunit for determining the fuzzy relationship between the pixels is configured to determine the fuzzy relationship between the mixed pixels based on the Euclidean distance between the mixed pixels.

A calculation subunit is configured to calculate the average approximation accuracy of the type II fuzzy rough set according to the fuzzy relationship.

The subunit for determining the magnitude of the reduction in the number of breakpoints is configured to determine the magnitude of the reduction in the number of breakpoints corresponding to the target remote sensing image data population individuals according to the number of breakpoints in the initial breakpoint set.

The fitness function determination subunit is configured to determine the fitness function of the type II fuzzy rough set according to the magnitude of the reduction in the number of breakpoints and the average approximation accuracy.

The fitness value determination subunit is configured to determine the fitness value of the type II fuzzy rough set according to the fitness function of the type II fuzzy rough set, and take the individual of each target remote sensing image data population corresponding to the optimal fitness value as the Optimal discretization result.

The present invention provides a remote sensing image feature discretization device based on type II fuzzy rough model. The device includes: a mixed pixel extraction unit 31, which is configured to obtain target remote sensing image data, and extract mixed pixels from the target remote sensing image data. , and each mixed pixel contains the spectral response characteristics of multiple ground object types. The primary membership determination unit 32 is configured to determine the primary membership of each mixed pixel corresponding to each feature type according to the mixed pixel. The secondary membership degree determination unit 33 is configured to calculate the secondary membership degree of each mixed pixel belonging to each feature type according to the primary membership degree. The fuzzy rough set determining unit 34 is configured to determine the type II fuzzy rough set of each feature type according to the primary membership degree and the secondary membership degree. The optimal discretization result determination unit 35 is configured to perform feature discretization processing on the target remote sensing image data to obtain the optimal discretization result. In the embodiment of the present invention, the rough set and the fuzzy set are combined to describe the fuzzy components in the process of discretization of remote sensing image features by the primary and secondary degrees of membership corresponding to the mixed pixels. The membership degree further fuzzifies the main membership degree, so as to accurately quantify and evaluate the uncertainty of mixed pixels, and obtain more accurate discretization results.

Example 3

An embodiment of the present invention further provides a non-transitory computer storage medium, where the computer storage medium stores computer-executable instructions, where the computer-executable instructions can execute the method described in any of the above method embodiments. Wherein, the storage medium may be a magnetic disk, an optical disk, a read-only memory (Read-Only Memory, ROM), a random access memory (Random Access Memory, RAM), a flash memory (Flash Memory), a hard disk (Hard) DiskDrive, abbreviation: HDD) or solid-state drive (Solid-State Drive, SSD), etc.; the storage medium may also include a combination of the above-mentioned types of memories.

Example 4

An embodiment of the present invention further provides a computer device. As shown in FIG. 4 , FIG. 4 is a schematic structural diagram of a computer device provided by an optional embodiment of the present invention. The computer device may include at least one processor 41 , at least one One communication interface 42, at least one communication bus 43 and at least one memory 44, wherein the communication interface 42 may include a display screen (Display) and a keyboard (Keyboard), and the optional communication interface 42 may also include a standard wired interface and a wireless interface. The memory 44 may be a high-speed RAM memory (Random Access Memory, volatile random access memory), or may be a non-volatile memory (non-volatile memory), such as at least one disk memory. The memory 44 can optionally also be at least one storage device located away from the aforementioned processor 41 . The processor 41 can be combined with the device described in FIG. 3 , the memory 44 stores application programs, and the processor 41 calls the program codes stored in the memory 44 to execute the steps of the methods described in any of the above method embodiments.

The communication bus 43 may be a peripheral component interconnect (PCI for short) bus or an extended industry standard architecture (EISA for short) bus or the like. The communication bus 43 can be divided into an address bus, a data bus, a control bus, and the like. For ease of presentation, only one thick line is used in FIG. 4, but it does not mean that there is only one bus or one type of bus.

The memory 44 may include volatile memory (English: volatile memory), such as random-access memory (English: random-access memory, abbreviation: RAM); the memory may also include non-volatile memory (English: non-volatile memory) memory), such as flash memory (English: flash memory), hard disk (English: hard diskdrive, abbreviation: HDD) or solid-state drive (English: solid-state drive, abbreviation: SSD); the memory 44 may also include the above-mentioned types of memory The combination.

The processor 41 may be a central processing unit (English: central processing unit, abbreviation: CPU), a network processor (English: network processor, abbreviation: NP), or a combination of CPU and NP.

The processor 41 may further include a hardware chip. The above hardware chip may be an application-specific integrated circuit (English: application-specific integrated circuit, abbreviation: ASIC), a programmable logic device (English: programmable logic device, abbreviation: PLD) or a combination thereof. The above PLD may be a complex programmable logic device (English: complex programmable logic device, abbreviation: CPLD), a field programmable gate array (English: field-programmable gate array, abbreviation: FPGA), a general-purpose array logic (English: generic arraylogic , abbreviation: GAL) or any combination thereof.

Optionally, memory 44 is also used to store program instructions. The processor 41 may invoke program instructions to implement the method described in any of the embodiments of the present invention.

Obviously, the above-mentioned embodiments are only examples for clear description, and are not intended to limit the implementation manner. For those of ordinary skill in the art, changes or modifications in other different forms can also be made on the basis of the above description. There is no need and cannot be exhaustive of all implementations here. And the obvious changes or changes derived from this are still within the protection scope of the present invention.

Claims (10)

1. a remote sensing image feature discretization method based on type II fuzzy rough model, is characterized in that, comprising: Obtaining target remote sensing image data, extracting mixed pixel from described target remote sensing image data, each described mixed pixel is respectively Including the spectral response characteristics of multiple ground object types; Determine the main membership degrees of each mixed pixel corresponding to each feature type according to the mixed pixels; Calculate each mixed pixel attributable to each feature according to the main membership degree the secondary membership degree of the type; according to the primary membership degree and secondary membership degree, determine the type II fuzzy rough set of each of the ground object types; perform feature discretization processing on the target remote sensing image data to obtain the optimal discretization result . 2 . The method for discretizing remote sensing image features based on a type II fuzzy rough model according to claim 1 , wherein the primary membership degree of each mixed pixel corresponding to each feature type is determined according to the mixed pixel. 3 . , including: iteratively calculating a fuzzy mean vector and a fuzzy covariance matrix of a preset fuzzy segmentation matrix, where the preset fuzzy segmentation matrix is composed of the membership degrees of mixed pixels corresponding to each object type; The matrix is divided to determine the master membership degree of each mixed pixel corresponding to each feature type. 3. The method for discretizing remote sensing image features based on type II fuzzy rough model according to claim 2, characterized in that, according to the iterative calculation of the fuzzy segmentation matrix when the iterative termination condition is satisfied, it is determined that each of the mixed pixels corresponds to The main membership degree of each feature type includes: determining the abundance corresponding to each mixed pixel according to the fuzzy segmentation matrix, and taking the abundance as the main membership degree corresponding to each feature type of each mixed pixel. 4 . The method for discretizing remote sensing image features based on type II fuzzy rough model according to claim 1 , wherein, according to the primary membership degree, the secondary membership of each mixed pixel belonging to each object type is calculated according to the primary membership degree. 5 . degree, including: determining a hard partition matrix according to the iterative calculation of the fuzzy partition matrix when the iteration termination condition is satisfied; determining a set of pixels belonging to each object type according to the hard partition matrix; calculating the set in the approximate space The upper approximation, the lower approximation, the positive domain, the negative domain, and the boundary domain are determined according to the upper approximation, the lower approximation, the positive domain, the negative domain, and the boundary domain. 5. The remote sensing image feature discretization method based on type II fuzzy rough model according to claim 1, wherein the feature discretization process is performed on the target remote sensing image data to obtain an optimal discretization result, comprising: : obtaining the initial breakpoint set of the mixed pixels from the remote sensing image data; initializing the target remote sensing image data population based on the number of breakpoints in the initial breakpoint set; Iteratively execute the genetic algorithm to determine the optimal discretization result; wherein, the discretization scheme corresponding to the initialized target remote sensing image data population is the initial discretization scheme, and each population individual corresponds to a discretization result. 6. The remote sensing image feature discretization method based on type II fuzzy rough model according to claim 5, wherein the genetic algorithm is iteratively executed to the individual of the target remote sensing image data population to determine the optimal discretization result , comprising: determining the fuzzy relationship between mixed pixels based on the Euclidean distance between the mixed pixels; calculating the average approximate accuracy of the type II fuzzy rough set according to the fuzzy relationship; The number of breakpoints, to determine the magnitude of the reduction in the number of breakpoints corresponding to the target remote sensing image data population; according to the magnitude of the reduction in the number of breakpoints and the average approximation accuracy, to determine the adaptation of the type II fuzzy rough set degree function; according to the fitness function of the type II fuzzy rough set, determine the fitness value of the type II fuzzy rough set, and use the individual of each target remote sensing image data population corresponding to the optimal fitness value as the The optimal discretization result is described. 7. The remote sensing image feature discretization method based on type II fuzzy rough model according to claim 6, is characterized in that, the fitness function of described type II fuzzy rough set is expressed by the following formula:

, where α and β are the weight coefficients, | D | is the magnitude of the reduction in the number of breakpoints,

is the average approximate precision. 8. A remote sensing image feature discretization device based on type II fuzzy rough model, characterized in that, comprising: a mixed pixel extraction unit, configured to obtain target remote sensing image data, and extract a mixed image from the target remote sensing image data. Each of the mixed pixels contains spectral response characteristics of multiple ground object types; the master membership degree determination unit is configured to determine the master membership degree of each mixed pixel corresponding to each feature type according to the mixed pixels The secondary membership degree determination unit is configured to calculate the secondary membership degree that each of the mixed pixels belongs to each feature type according to the primary membership degree; the fuzzy rough set determination unit is configured to calculate the secondary membership degree according to the primary membership degree and the secondary membership degree. The membership degree is used to determine the type II fuzzy rough set of each of the ground object types; the optimal discretization result determination unit is configured to perform feature discretization processing on the target remote sensing image data to obtain the optimal discretization result. 9 . A non-transitory computer-readable storage medium, characterized in that, the non-transitory computer-readable storage medium stores computer instructions, and when the computer instructions are executed by a processor, any one of claims 1 to 7 is implemented. 10 . The discretization method of remote sensing image features based on type II fuzzy rough model described in Item. 10. A computer device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, The instructions are executed by the at least one processor, so as to execute the method for discretizing remote sensing image features based on a type II fuzzy rough model according to any one of claims 1-7.

Images