Evolutionary multi-task optimization-based urban land remote sensing image classification method

Abstract

The application discloses a kind of urban land remote sensing image classification methods based on evolution multitask optimization, comprising: 1: obtaining urban land cover remote sensing image sample and pre-processing, to build sample dataset;2: the sample dataset is decomposed into multiple training subsets, and the corresponding sub-population is constructed for each training subset;3: the state of sub-population is evaluated to adjust the probability of executing knowledge transfer operation online;4: different source populations are selected according to the state of sub-population to provide knowledge for it;5: more optimal classifier is obtained by using the model representing global information;6: select a part of poor performance classifier to carry out unbalanced adjustment;7: finally, the best individual is selected from the sub-population to form a multi-classifier, to identify different types of urban land use and cover situation.The application can quickly classify different urban land types by constructing and training the optimal multi-classifier, so as to improve the accuracy and precision of land type classification result.

Description

Technical Field

[0001] The present invention relates to the technical field of urban remote sensing image classification, and in particular to an urban land remote sensing image classification method based on evolutionary multi-task optimization. Background Art

[0002] With the continuous advancement of urbanization, the monitoring and analysis of urban land cover types has become particularly important. Traditional urban land classification methods rely on manual feature extraction and manual adjustment of classification rules, which often suffer from high uncertainty and low efficiency. With the development of remote sensing technology, an increasing number of remote sensing image classification models are being used to monitor and identify urban land use. However, these methods are based on machine learning models for classification, which have black-box characteristics and are difficult to adjust parameters, resulting in low optimization efficiency and poor interpretability. Furthermore, existing urban land remote sensing image samples are often unbalanced. Classification models optimized using unbalanced datasets may overlook a small number of land cover samples, resulting in low classification accuracy and poor generalization ability. Summary of the Invention

[0003] In order to overcome the shortcomings of the existing technology, the present invention proposes an urban land remote sensing image classification method based on evolutionary multi-task optimization, in order to quickly and accurately classify remote sensing images of different urban land cover, thereby improving the accuracy and precision of land type classification results.

[0004] In order to achieve the above-mentioned object, the present invention adopts the following technical solutions:

[0005] The urban land remote sensing image classification method based on evolutionary multi-task optimization of the present invention is characterized in that it is performed according to the following steps:

[0006] Step 1: Collection The remote sensing images of different urban land types are extracted and the attribute characteristics of each land type are normalized to obtain Remote sensing image sample sets of different urban land types ,in, Indicates the Remote sensing image samples of urban land types, Indicates the The attribute feature vectors of remote sensing images of urban land types are: , Indicates the Remote sensing images of urban land types attribute feature values, express Land type label, , d represents the dimension of the attribute feature vector, , is the number of land types;

[0007] Step 2: Divide the original sample set based on one-to-many decomposition strategy , obtaining binary remote sensing image training subsets , ,…, ,…, }, wherein represents the binary remote sensing image training subset, and , wherein represents the attribute feature vector of the remote sensing image of the th urban land type in , and represents the land type label of ; is a remote sensing image sample set in with the land type label ; in which the samples with the th class label are regarded as positive samples, and the remote sensing image sample set with the rest of the land type labels is regarded as negative samples;

[0008] Step 3: Define the current iteration number as , and initialize ;

[0009] Construct sub-classifier groups, wherein the th sub-classifier group contains sub-classifiers, and each sub-classifier is coded as an individual, thereby obtaining the th sub-classifier population of the th generation , ,…, ,…, } and randomly initializing; wherein represents the th individual in the th sub-classifier population of the th generation, and the dimension of each individual is , ;

[0010] Step 4: Initialize ;

[0011] Step 5: Based on in the corresponding th binary remote sensing image training subset Classification on In convergence state or ;

[0012] Step 6: Calculate using formula (1) The migration probability ;

[0013] (1)

[0014] In formula (1), express The migration probability is season is the set initial value; is the scaling factor;

[0015] Step 7: Randomly initialize the Generate the jth random number ,like , then use the differential evolution algorithm to Perform crossover mutation to obtain Generation first child collection , and go to step 10; otherwise, go to step 8;

[0016] Step 8: Multi-state based source population selection strategy Select a classifier population from the m-1 classifier populations outside as source populations;

[0017] Step 9: Based on the global population model Source population Perform knowledge transfer to obtain The first child set of as well as The first child set of ;

[0018] Step 10: Build a bootstrap dataset and generate bootstrap vectors for individuals with poor performance , used to get The second generation set of ;

[0019] Step 11: Assign to ,like , then execute step 12, otherwise, return to step 5 and execute sequentially;

[0020] Step 12: of and Merge into The jth merged population of generation ,calculate Each individual in the On the second-class remote sensing image training subset The area under the curve, thus using the elite selection strategy from Select N outstanding individuals as the first Daidi subclassifier population ;

[0021] Step 13: Assign to ,judge Is it true? If so, go to step 14; otherwise, go back to step 4 and execute sequentially;

[0022] Step 14: From In each generation, select The individual with the largest area under the curve constitutes the optimal multi-classifier for urban land remote sensing image sample classification ={ , ,…, ,…, }, used to realize the classification of different urban land types, where Indicates the Daidi subclassifier population The best individual.

[0023] The method for classifying urban land remote sensing image samples based on evolutionary multi-task optimization described in the present invention is characterized in that if the label of the t-th urban land remote sensing image sample , indicating that the t-th urban land remote sensing image sample is classified as artificial building cover, if , indicating that the t-th urban land remote sensing image sample is classified as forest cover, if , indicating that the t-th urban land remote sensing image sample is classified as water cover, if , indicating that the t-th urban land remote sensing image sample is classified as agriculture / grassland cover.

[0024] Furthermore, step 5 is performed as follows:

[0025] Step 5.1: A binary classification remote sensing image training subset All attribute feature vectors in enter Processed in Output of all land type prediction labels , and with the True land type labels in the training subset of the second-class remote sensing images Compare and calculate In the Second-class remote sensing image training subset True positive rate and the false positive rate And form coordinate points and map them to the target space grid, and then The coordinate points composed of the true positive rate and false positive rate of each individual are mapped to the target space grid;

[0026] Step 5.2: According to True positive rate and false positive rate The distribution in the target space is obtained The number of grids occupied by individuals on the Pareto front in the target space grid is taken as Diversity indicators ;

[0027] Step 5.3: Calculation Each individual in the binary classification remote sensing image training subset The classification accuracy on Average precision and as The accuracy index of

[0028] Step 5.4: According to and , discriminate In convergence state or :

[0029] when > and > When In an evolutionary state ;in, Indicates the -1st generation subclassifier population The accuracy index, Indicates the -1st generation subclassifier population Diversity indicators;

[0030] when < or < When In convergence state .

[0031] Furthermore, step 8 is performed as follows:

[0032] Step 8.1: If In convergence state , then calculate The probability of selecting the i-th random task = , thus obtaining Random task selection vector ={ , ,…, ,…, },in,

[0033] like In an evolutionary state , then calculate The probability of selecting the i-th similar task , thus obtaining Similar task selection vector ={ , ,…, ,…, },in, Indicates the Daidi subclassifier populations, represents the cosine similarity,

[0034] Step 8.2: According to or , using the roulette algorithm from the remaining Select a classifier population from the m-1 classifier populations as source population.

[0035] Furthermore, step 9 is performed as follows:

[0036] Step 9.1: Make The element in row a and column b is recorded as ,but Middle dimensional column vector Recorded as , and obeys Gaussian distribution , thus constructing Independent multivariate Gaussian model of ;in, , ;

[0037] make The element in row a and column b is recorded as ,but Middle dimensional column vector Recorded as , and obeys Gaussian distribution , thus constructing Independent multivariate Gaussian model of ;

[0038] Step 9.2: The mean encoding of , ;in, express Middle The mean of the dimensional column vector, and , and The variance of , ;in, express Middle The variance of the dimensional column vector, and ;

[0039] Step 9.3: According to and The KL divergence between is obtained by formula (2) and The crossover probability between , thus obtaining and The crossover probability vector between ={ };

[0040] (2)

[0041] Step 9.4: Initialize b=1;

[0042] Step 9.5: Randomly initialize the Generate the jth random number of the bth dimension ,like , then in and in After the crossover, each of them mutates, and the two b-th dimension means after the crossover mutation are obtained accordingly. and At the same time, in and middle Cross and mutate each other, and get the two b-dimensional variances after cross mutation. and Otherwise, directly and Mutate separately to obtain the two b-th dimension means after mutation and , then and Mutate separately and get two variances after mutation and ;

[0043] Step 9.6: If b <d,则将b+1赋值给b后,返回步骤9.5;否则,表示得到交叉变异后的均值向量 and variance vector , and proceed to step 9.7;

[0044] Step 9.7: Decoding and The mean vector after cross-mutation and variance vector ,get New independent multivariate Gaussian model and with New independent multivariate Gaussian model ;

[0045] Step 9.8: From and Randomly sample out / 2 individuals and form the g-th generation temporary population = ,in, represents the nth temporary individual of the gth generation;

[0046] Step 9.9: If and Euclidean distance Less than and Euclidean distance , then join in The first child set of Otherwise, join in The first child set of Thus, N temporary individuals are added to the first offspring set of the corresponding population.

[0047] Furthermore, the step 10 is performed as follows:

[0048] Step 10.1: Calculation Each individual in A binary classification remote sensing image training subset on The area under the curve, and select The area under the curve is the smallest Individuals form the jth population to be guided in the gth generation ={ , ,…, ,…, },in, express The zth individual to be guided;

[0049] Step 10.2: Find In the Second-class remote sensing image training subset The remote sensing image samples with incorrect classification and the remote sensing image samples with minority class labels are combined to form the individuals to be guided The bootstrap dataset ;

[0050] Step 10.3: Calculate using formula (3) The guiding vector on ;

[0051] (3)

[0052] In formula (3), express The distance matrix between remote sensing image samples of different land categories, express The distance matrix of remote sensing image samples of the same land category;

[0053] Step 10.4: Using the crossover operator pair of differential evolution and Interact and get the zth new guided individual and join The second generation set of middle.

[0054] The electronic device of the present invention includes a memory and a processor, wherein the memory is used to store a program that supports the processor to execute the urban land remote sensing image classification method, and the processor is configured to execute the program stored in the memory.

[0055] The present invention provides a computer-readable storage medium, wherein a computer program is stored on the computer-readable storage medium. The characteristic of the computer program is that when the computer program is run by a processor, the steps of the urban land remote sensing image classification method are executed.

[0056] Compared with the prior art, the present invention has the following beneficial effects:

[0057] 1. Different from the traditional urban land remote sensing image classification method, the present invention proposes an urban land remote sensing image classification method based on the combination of evolutionary multi-task optimization and decomposition multi-classification. By extracting attribute features that can effectively distinguish different types of urban land from urban land remote sensing image samples, and using evolutionary multi-task optimization to train the classification model, the land cover of the samples can be quickly, efficiently and accurately classified from a large number of urban land remote sensing image features. This method decomposes the multi-classification problem of urban land remote sensing images into multiple binary classification land remote sensing image problems, so that each binary classification model only needs to distinguish the characteristic differences between one type of land cover and other land covers; at the same time, when a new land cover category appears, only a new binary classification model needs to be trained to identify the difference between the new land cover and the original land cover, and the remaining binary classification models can be fine-tuned, avoiding the complexity of redesigning the entire land remote sensing image multi-classifier; in addition, the method of the present invention can independently optimize the binary classification model for specific land cover types that are more difficult to optimize, and can train special models for certain land cover categories (such as artificial building cover) based on the importance priority of identification, thereby improving the recognition rate of specific land cover categories; if a binary classification model performs poorly, it can also be clearly mapped to a specific binary classification model, which helps to locate the source of the problem and optimize the training process, thereby saving manpower and time on a large scale, and in most cases, it can also save money, and can achieve the same or higher accuracy as manual interpretation and the combination of traditional statistical methods, thereby improving the recognition efficiency of land cover categories.

[0058] 2. In the process of training urban land remote sensing image prediction parameters, the present invention adopts an evolutionary multi-task method. Each binary land remote sensing image classification model uses Area under the curve ( ) as the optimization goal, it can well measure the overall performance and classification accuracy of unbalanced data processing, thereby overcoming the problems that traditional land remote sensing image classification algorithms cannot effectively solve the imbalance of existing land remote sensing image sample categories and the difficulty of black-box optimization and parameter adjustment of land remote sensing image classification models. The trained classification model is superior to the traditional land remote sensing image classification model, thereby improving the classification performance and accuracy of urban land remote sensing images.

[0059] 3、The application in the process of training urban land remote sensing image sample, through the online judgment each two classification remote sensing image sub-problem corresponding land remote sensing image classifier population state, and timely adjust the population when land cover feature knowledge migration and suitable two classifier population migration, at the same time, by establishing the distribution model for each population which can represent the global urban land image feature information, according to the similar land cover feature dimension interaction, the dissimilar land cover feature dimension independent optimization principle optimization, make each sub-population evolution obtain the most beneficial optimization operation, thereby improve the classification effect of each two classification sub-problem, combination obtains multi-classifier, finally improve the urban land remote sensing image classification performance. BRIEF DESCRIPTION OF DRAWINGS

[0060] Figure 1 For the application based on one-to-many decomposition multi-classification solution diagram

[0061] Figure 2 For the population of the application The diagram of independent multivariate Gaussian model ;

[0062] Figure 3 For the flow chart of the method of the application. DETAILED DESCRIPTION

[0063] In this embodiment, a city land remote sensing image classification method based on evolutionary multi-task optimization, including city land remote sensing image classification model training step and city land remote sensing image detection step, specifically, as shown in Figure 3 , is as follows:

[0064] Step one: city land remote sensing image classification model training:

[0065] Step 1.1: collect different city land type remote sensing image, and extract the attribute features of the remote sensing image about each land type and perform standard normalization processing, thereby obtaining a remote sensing image sample set of different city land types , wherein, represents the remote sensing image sample of the city land type, represents the attribute feature vector of the remote sensing image of the city land type, and has: , represents the attribute feature value of the remote sensing image of the city land type, represents the corresponding land type label, , d represents the dimension of the attribute feature vector, , is the number of land types.

[0066] Step 1.2: Divide the original sample set based on the one-to-many decomposition strategy ,get Two-class remote sensing image training subset { , ,…, ,…, },in, Indicates the A binary classification remote sensing image training subset, and ,in, represent Middle The attribute feature vector of the remote sensing image of urban land types, express Land type label; It will The land type label is The remote sensing image sample sets with the same land type labels are considered as positive samples, and the remote sensing image sample sets with the other land type labels are considered as negative samples;

[0067] Figure 1 Taking a four-class urban land remote sensing image problem as an example, the basic processing flow of one-to-many decomposition is demonstrated, where the binary classifier trained by the i-th binary remote sensing image training subset is used to distinguish the i-th type of urban land remote sensing image samples from urban land remote sensing image samples of all other types.

[0068] Step 1.3: Define the current number of iterations as , and initialize ;

[0069] Build sub-classifier groups, among which the The subclassifier group contains sub-classifiers, and encode each sub-classifier as an individual, thus obtaining the Daidi subclassifier population ={ , ,…, ,…, } and initialized randomly; among them, Indicates the Daidi subclassifier population Middle individuals, and the dimension of each individual is , .

[0070] Step 1.4: Initialization ;

[0071] Step 1.5: Based on In the corresponding A binary classification remote sensing image training subset Classification on In convergence state or Consider whether it is necessary to introduce external knowledge transfer or focus on self-evolution and iteration based on the state of the land remote sensing image classifier;

[0072] Step 1.5.1: Place A binary classification remote sensing image training subset All attribute feature vectors in enter Processed in Output of all land type prediction labels , and with the True land type labels in the training subset of the second-class remote sensing images Compare and calculate In the Second-class remote sensing image training subset True positive rate and the false positive rate And form coordinate points and map them to the target space grid, and then The true positive rate and false positive rate of each individual in the target space are mapped to the target space grid.

[0073] Step 1.5.2: According to True positive rate and false positive rate ,get The number of grids occupied by individuals on the Pareto front in the target space grid is taken as Diversity indicators ;

[0074] Step 1.5.3: Calculation The classification accuracy of each individual in , thus obtaining Average precision and as The accuracy index of

[0075] Step 1.5.4: According to and , discriminate In convergence state or :

[0076] when > and > When In an evolutionary state , this indicates that the optimization state of the land remote sensing image classifier is good and has made progress compared with the previous generation. Considering reducing the probability of introducing external knowledge, it focuses on self-evolution; among them, Indicates the -1st generation subclassifier population The accuracy index, Indicates the -1st generation subclassifier population Diversity indicators;

[0077] when < or < When In convergence state At this time, it shows that the optimization state of the land remote sensing image classifier has encountered a bottleneck in a certain indicator. Compared with the previous generation, no progress has been made. Consider increasing the probability of introducing external knowledge to make it jump out of the local optimum.

[0078] Step 1.6: Calculate using formula (1) The migration probability , and according to Determine whether knowledge transfer is necessary

[0079] (1)

[0080] In formula (1), express The migration probability is season is the set initial value; is the scaling factor;

[0081] Step 1.7: Randomly initialize the Generate the jth random number ,like , then use the differential evolution algorithm to Perform crossover mutation to obtain Generation first child collection , proceed to step 1.10; otherwise, proceed to step 1.8.

[0082] Step 1.8: Multi-state based source population selection strategy from Select a classifier population from the m-1 classifier populations outside as source populations;

[0083] Step 1.8.1: If In convergence state , more random and diverse knowledge sources may help Jump out of the local optimum and calculate The probability of selecting the i-th random task = , thus obtaining Random task selection vector ={ , ,…, ,…, },in,

[0084] like In an evolutionary state , Choosing more similar knowledge sources will help Accelerate the optimization process to converge faster and calculate The probability of selecting the i-th similar task , thus obtaining Similar task selection vector ={ , ,…, ,…, },in, Indicates the Daidi subclassifier populations, represents the cosine similarity,

[0085] Step 1.8.2: According to or , using the roulette algorithm from the remaining Select a classifier population from the m-1 classifier populations as source population.

[0086] Step 1.9: Analyze the population based on the global population model Source population Perform knowledge transfer to obtain The first child set of as well as The first child set of ,Different from the knowledge transfer between individuals, the global population model can ,better represent all the land remote sensing image feature knowledge learned by the ,current classifier population, and can perform knowledge transfer more comprehensively and ,efficiently;

[0087] Step 1.9.1: Make The element in row a and column b is recorded as ,but Middle dimensional column vector Recorded as , and obeys Gaussian distribution , which actually represents the distribution of classification decision variables learned from the b-th dimension features of land remote sensing image samples, and can represent All classifiers are trained on the b-th dimension features to obtain knowledge, and then construct Independent multivariate Gaussian model of ; Represents the land remote sensing image classifier population The classification knowledge obtained by training on the land features of all dimensions learned, , ;

[0088] make The element in row a and column b is recorded as ,but Middle dimensional column vector Recorded as , and obeys Gaussian distribution , thus constructing Independent multivariate Gaussian model of ;

[0089] Figure 2 The figure shows the independent multivariate Gaussian model and the Gaussian model that each dimensional decision variable obeys. Each dimension is independent of each other, so each dimensional decision variable (column vector) of the population obeys a Gaussian model. All Gaussian models together constitute a d-dimensional independent multivariate Gaussian model, which can characterize the parameter distribution range, peak or mean dispersion degree and other information of each dimensional land remote sensing image classifier.

[0090] Step 1.9.2: The mean encoding of , ;in, Represents population Middle The mean of the dimensional column vector, and , and The variance of , ;in, Represents population Middle The variance of the dimensional column vector, and .

[0091] Step 1.9.3: According to and The KL divergence between them can be used to accurately judge and The correlation between each dimension of decision variables. The more similar each dimension of decision variables is, the easier it is to benefit from similar remote sensing image features. Similarly, if they are not similar, the crossover probability of this dimension is reduced to prevent the occurrence of irrelevant land remote sensing image features leading to negative migration. Through formula (2), we can get and The crossover probability between , thus obtaining and The crossover probability vector between ={ };

[0092] (2)

[0093] Step 1.9.4: Initialize b=1;

[0094] Step 1.9.5: Randomly initialize Generate the jth random number of the bth dimension ,like , then in and in After the crossover, each of them mutates, and the two b-th dimension means after the crossover mutation are obtained accordingly. and At the same time, in and middle Cross and mutate each other, and get the two b-dimensional variances after cross mutation. and Otherwise, directly and Mutate separately to obtain the two b-th dimension means after mutation and , then and Mutate separately and get two variances after mutation and ;

[0095] Step 1.9.6: If b <d,则将b+1赋值给b后,返回步骤9.5;否则,表示得到交叉变异后的均值向量 and variance vector , and perform step 1.9.7.

[0096] Step 1.9.7: Decoding and The mean vector after cross-mutation and variance vector ,get New independent multivariate Gaussian model ,and New independent multivariate Gaussian model ;

[0097] Step 1.9.8: From and Randomly sample out / 2 individuals and form the g-th generation temporary population = ,in, represents the nth temporary individual of the gth generation;

[0098] Step 1.9.9: If and Euclidean distance Less than and Euclidean distance , then join in The first child set of Otherwise, join in The first child set of Thus, N temporary individuals are added to the first offspring set of the corresponding population. In this step, the source population and When transferring knowledge, It is also possible to benefit from this migration and obtain effective information. Similarly, in order to optimize efficiency, the temporary population of the g generation can be directly Join All The first child set of middle.

[0099] Step 1.10: Considering that urban land classes have unbalanced properties, for example, there are fewer forest cover samples in cities, while there are more housing cover and artificial green land cover samples, in order to prevent the unbalanced land cover classes from interfering with the optimization of land remote sensing image classifiers, a bootstrap dataset is constructed. , generate guidance vectors for individuals with poor performance , used to get The second generation set of ;

[0100] Step 1.10.1: Calculation Each individual in A binary classification remote sensing image training subset on The area under the curve, and select The area under the curve is the smallest Individuals form the jth population to be guided in the gth generation ={ , ,…, ,…, },in, express The zth individual to be guided;

[0101] Step 1.10.2: Find In the Second-class remote sensing image training subset The remote sensing image samples with incorrect classification and the remote sensing image samples with minority class labels are combined to form the individuals to be guided The bootstrap dataset ;

[0102] Step 1.10.3: Calculate using formula (3) The guiding vector on ;

[0103] (3)

[0104] In formula (3), express The distance matrix between remote sensing image samples of different land categories, express The distance matrix of remote sensing image samples of the same land category;

[0105] Step 1.10.4: Using the crossover operator pair of differential evolution and Interact and get the zth new guided individual and join The second generation set of middle.

[0106] Step 1.11: Assign to ,like , then execute step 1.12, otherwise, return to step 1.5 and execute sequentially;

[0107] Step 1.12: of and Merge into The jth merged population of generation ,calculate Each individual in the Second-class remote sensing image training subset on The area under the curve, thus using the elite selection strategy from Select N outstanding individuals as the first Daidi subclassifier population ;

[0108] Step 1.13: Assign to ,judge Is it true? If so, go to step 1.14; otherwise, go back to step 1.4 and execute sequentially.

[0109] Step 1.14: From In each generation, select The individual with the largest area under the curve constitutes the optimal multi-classifier for urban land remote sensing image sample classification ={ , ,…, ,…, }, used to realize the classification of different urban land types, where Indicates the Daidi subclassifier population The best individual.

[0110] Step 2: Urban land remote sensing image prediction: Use the urban land remote sensing image classification model obtained in step 1 and the corresponding urban land remote sensing image attribute features to be tested to perform urban land cover classification:

[0111] Step 2.1: Input the urban remote sensing image data to be tested as the sample to be tested and extract the attribute features of urban land remote sensing image, which are recorded as ,in, Indicates the number of samples to be tested Attribute characteristics of urban land remote sensing images;

[0112] Step 2.2: Attribute characteristics of the sample to be tested Perform standard normalization to obtain the feature vector of the sample to be tested ;in, Indicates the normalized first Attribute characteristics of urban land remote sensing images;

[0113] Step 2.3: Input the feature vector of the sample to be tested into the urban land remote sensing image classification model obtained in Step 1.14 to obtain the classification result. If the output is 1, it means that the tth urban land remote sensing image sample is classified as artificial building cover. If the output is 2, it means that the tth urban land remote sensing image sample is classified as forest cover. If the output is 3, it means that the tth urban land remote sensing image sample is classified as water cover. If the output is 4, it means that the tth urban land remote sensing image sample is classified as agricultural / grassland cover.

[0114] In this embodiment, an electronic device includes a memory and a processor, wherein the memory is used to store a program that supports the processor to execute the above method, and the processor is configured to execute the program stored in the memory.

[0115] In this embodiment, a computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the above method are executed.

[0116] The specific implementation method of the present invention is illustrated and the effect of the method of the present invention is verified by using an example of a simulation data set.

[0117] 1) Prepare standard dataset:

[0118] This paper uses the Urban Land Cover dataset as a benchmark dataset to verify the effectiveness of the urban land remote sensing image classification model. The Urban Land Cover dataset is a multi-classification benchmark dataset widely used for imbalanced classification. The Ecoli dataset contains 307 data points, corresponding to 307 remote sensing image samples of land cover collected in different areas of cities. All of these samples are used to verify the performance of the present invention for urban land remote sensing image classification. These 307 training samples are divided into training and test sample data.

[0119] 2) Evaluation indicators:

[0120] The area under the ROC curve (AUC) was used as the evaluation metric in this example to assess the classification performance of the present invention on different breast tissue CT images. Higher values ​​indicate better classification performance and greater generalization when dealing with imbalanced samples.

[0121] 3) Experiment on standard datasets:

[0122] To verify the effectiveness of our proposed method, we compared our classification algorithm with the classic EMT-MC and EMaTO algorithms for modeling and prediction on the Urban Land Cover dataset. The experimental results are shown in Table 1.

[0123] Table 1 Comparative experimental results

[0124]

[0125] Table 1 shows that in the multi-class dataset Urban Land Cover, the classification algorithm of the present invention achieves better classification performance than the other two, demonstrating the effectiveness of the present invention. With the continuous advancement of urbanization, the monitoring of urban land cover types is of great significance for urban planning and disaster management. Therefore, more and more remote sensing images are being used to monitor and identify urban land use. This method can cover a wider range and improve accuracy. Based on this, the present invention can quickly and effectively classify sample states from a large number of urban land remote sensing image features, improving the efficiency of urban land cover type classification.

Claims

1. A classification method for urban land remote sensing images based on evolutionary multi-task optimization, characterized in that: The steps are as follows: Step 1: Collection The remote sensing images of different urban land types are extracted and the attribute characteristics of each land type are normalized to obtain Remote sensing image sample sets of different urban land types ,in, Indicates the Remote sensing image samples of urban land types, Indicates the The attribute feature vectors of remote sensing images of urban land types are: , Indicates the Remote sensing images of urban land types attribute feature values, express Land type label, , d represents the dimension of the attribute feature vector, , is the number of land types; Step 2: Divide the original sample set based on the one-to-many decomposition strategy ,get Two-class remote sensing image training subset { , ,…, ,…, },in, Indicates the A binary classification remote sensing image training subset, and ,in, represent Middle The attribute feature vector of the remote sensing image of urban land types, express Land type label; It will The land type label is Remote sensing image sample set, Lieutenant General The samples with class labels are considered as positive samples, and the remote sensing image sample sets with other land type labels are considered as negative samples; Step 3: Define the current number of iterations as , and initialize ; Build sub-classifier groups, among which the The subclassifier group contains sub-classifiers, and encode each sub-classifier as an individual, thus obtaining the Daidi subclassifier population ={ , ,…, ,…, } and initialized randomly; among them, Indicates the Daidi subclassifier population Middle individuals, and the dimension of each individual is , ; Step 4: Initialization ; Step 5: Based on In the corresponding A binary classification remote sensing image training subset Classification on In convergence state or ; Step 6: Calculate using formula (1) The migration probability ; (1) In formula (1), express The migration probability is season is the set initial value; is the scaling factor; Step 7: Randomly initialize the Generate the jth random number ,like , then use the differential evolution algorithm to Perform crossover mutation to obtain Generation first child collection , and go to step 10; otherwise, go to step 8; Step 8: Multi-state based source population selection strategy Select a classifier population from the m-1 classifier populations outside as source populations; Step 9: Based on the global population model Source population Perform knowledge transfer to obtain The first child set of as well as The first child set of ; Step 10: Build a bootstrap dataset and generate bootstrap vectors for individuals with poor performance , used to obtain The second generation set of ; Step 11: Assign to ,like , then execute step 12, otherwise, return to step 5 and execute sequentially; Step 12: of and Merge into The jth merged population of generation ,calculate Each individual in the On the second-class remote sensing image training subset The area under the curve, thus using the elite selection strategy from Select N outstanding individuals as the first Daidi subclassifier population ; Step 13: Assign to ,judge Is it true? If so, go to step 14; otherwise, go back to step 4 and execute sequentially; Step 14: In each generation, select The individual with the largest area under the curve constitutes the optimal multi-classifier for urban land remote sensing image sample classification ={ , ,…, ,…, }, used to realize the classification of different urban land types, where Indicates the Daidi subclassifier population The best individual.

2. The urban land remote sensing image sample classification method based on evolutionary multi-task optimization according to claim 1 is characterized in that: If the label of the t-th urban land remote sensing image sample , indicating that the t-th urban land remote sensing image sample is classified as artificial building cover, if , indicating that the t-th urban land remote sensing image sample is classified as forest cover, if , indicating that the t-th urban land remote sensing image sample is classified as water cover, if , indicating that the t-th urban land remote sensing image sample is classified as agriculture / grassland cover.

3. The urban land remote sensing image classification method based on evolutionary multi-task optimization according to claim 1 is characterized in that: Described step 5 is carried out as follows: Step 5.1: A binary classification remote sensing image training subset All attribute feature vectors in enter Processed in Output of all land type prediction labels , and with the True land type labels in the training subset of the second-class remote sensing images Compare and calculate In the Second-class remote sensing image training subset True positive rate and the false positive rate And form coordinate points and map them to the target space grid, and then The coordinate points composed of the true positive rate and false positive rate of each individual are mapped to the target space grid; Step 5.2: According to True positive rate and false positive rate The distribution in the target space is obtained The number of grids occupied by individuals on the Pareto front in the target space grid is taken as Diversity indicators ; Step 5.3: Calculation Each individual in the binary classification remote sensing image training subset The classification accuracy on Average precision and as The accuracy index of Step 5.4: According to and , discriminant In convergence state or : when > and > When In an evolutionary state ;in, Indicates the -1st generation subclassifier population The accuracy index, Indicates the -1st generation subclassifier population Diversity indicators; when < or < When In convergence state .

4. The urban land remote sensing image classification method based on evolutionary multi-task optimization according to claim 3 is characterized in that: Described step 8 is carried out as follows: Step 8.1: If In convergence state , then calculate The probability of selecting the i-th random task = , thus obtaining Random task selection vector ={ , ,…, ,…, },in, ; like In an evolutionary state , then calculate The probability of selecting the i-th similar task , thus obtaining Similar task selection vector ={ , ,…, ,…, },in, Indicates the Daidi subclassifier populations, represents the cosine similarity, ; Step 8.2: According to or , using the roulette algorithm from the remaining Select a classifier population from the m-1 classifier populations as source population.

5. The urban land remote sensing image classification method based on evolutionary multi-task optimization according to claim 4 is characterized in that: Described step 9 is carried out as follows: Step 9.1: Make The element in row a and column b is recorded as ,but Middle dimensional column vector Recorded as , and obeys Gaussian distribution , thus constructing Independent multivariate Gaussian model of ;in, , ; make The element in row a and column b is recorded as ,but Middle dimensional column vector Recorded as , and obeys Gaussian distribution , thus constructing Independent multivariate Gaussian model of ; Step 9.2: The mean encoding of , ;in, express Middle The mean of the dimensional column vector, and , and The variance of , ;in, express Middle The variance of the dimensional column vector, and ; Step 9.3: According to and The KL divergence between is obtained by formula (2) and The crossover probability between , thus obtaining and The crossover probability vector between ={ }; (2) Step 9.4: Initialize b=1; Step 9.5: Randomly initialize the Generate the jth random number of the bth dimension ,like , then in and in After the crossover, each of them mutates, and the two b-th dimension means after the crossover mutation are obtained accordingly. and At the same time, in and middle Cross and mutate each other, and get the two b-dimensional variances after cross mutation. and Otherwise, directly and Mutate separately to obtain the two b-th dimension means after mutation and , then and Mutate separately and get two variances after mutation and ; Step 9.6: If b < d, then assign b + 1 to b and return to Step 9.5; otherwise, it means that the mean vector after crossover mutation is obtained and the variance vector , and execute Step 9.7; Step 9.7: Decoding and The mean vector after cross-mutation and variance vector ,get New independent multivariate Gaussian model and with New independent multivariate Gaussian model ; Step 9.8: From and Randomly sample out / 2 individuals and form the g-th generation temporary population = ,in, represents the nth temporary individual of the gth generation; Step 9.9: If and Euclidean distance Less than and Euclidean distance , then join in The first child set of Otherwise, join in The first child set of Thus, N temporary individuals are added to the first offspring set of the corresponding population.

6. The urban land remote sensing image classification method based on evolutionary multi-task optimization according to claim 5 is characterized in that: The step 10 is performed as follows: Step 10.1: Calculation Each individual in A binary classification remote sensing image training subset on The area under the curve, and select The area under the curve is the smallest Individuals form the jth population to be guided in the gth generation ={ , ,…, ,…, },in, express The zth individual to be guided; Step 10.2: Find In the Second-class remote sensing image training subset The remote sensing image samples with incorrect classification and the remote sensing image samples with minority class labels are combined to form the individuals to be guided The bootstrap dataset ; Step 10.3: Calculate using formula (3) The guiding vector on ; (3) In formula (3), express The distance matrix between remote sensing image samples of different land categories, express The distance matrix of remote sensing image samples of the same land category; Step 10.4: Using the crossover operator pair of differential evolution and Interact and get the zth new guided individual and join The second generation set of middle.

7. An electronic device comprising a memory and a processor, characterized in that: The memory is used to store a program that supports the processor to execute the urban land remote sensing image classification method according to any one of claims 1 to 6, and the processor is configured to execute the program stored in the memory.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the urban land remote sensing image classification method according to any one of claims 1 to 6 are executed.

Images