Big data-based drainage basin glacier change information reliability evaluation method and system

Abstract

The invention relates to the technical field of electric digital data processing, in particular to a drainage basin glacier change information reliability evaluation method and system based on big data, and the method comprises the steps: obtaining a marked and sorted target region glacier standard form image set; segmenting the standard form image set to obtain a training set, a first test set and a second test set; uploading the training set and the first test set to a hierarchical iteration model for training to obtain a type 1 evaluation model; uploading the training set and the second test set to a hierarchical iteration model for training to obtain a type-2 evaluation model; fusing the first type evaluation model and the second type evaluation model to obtain a reliability evaluation model; and according to the reliability evaluation model, executing reliability evaluation of the drainage basin glacier change information. According to the invention, reliability evaluation of drainage basin glacier change information is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of electrical digital data processing, in particular to a method for evaluating the reliability of watershed glacier change information based on big data, and also relates to an evaluation system. Background Art

[0002] Glaciers are an important part of the cryosphere. They are not only one of the important driving factors of climate change, but also recorders and early warning devices that reflect climate change. It is of great significance to quickly and accurately detect glacier change areas. At present, the main methods for glacier change detection based on remote sensing images include simple algebraic algorithms, image classification methods, structural feature analysis methods, etc. However, due to the continuous warming of the global climate, the melting of glaciers has accelerated, and the positions of glaciers have changed continuously. The methods of the existing technologies can no longer well meet the effective prediction of changing glaciers, which leads to loopholes and uncontrollable risks in the prediction of global climate change.

[0003] There is a Chinese patent application for invention with a publication date of March 21, 2023, a publication number of CN115830466A, and a title of "A Remote Sensing Detection Method for Glacier Changes Based on a Deep Siamese Neural Network", which discloses preprocessing dual-temporal glacier remote sensing images to obtain training data for glacier change detection; inputting the dual-temporal training data into a deep siamese neural network for training to obtain a trained deep siamese neural network; inputting dual-temporal glacier remote sensing image data into the trained deep siamese neural network to obtain a binary prediction image for glacier change detection, and vectorizing the obtained binary prediction image for glacier change detection to obtain a vector map of glacier change areas.

[0004] The aforementioned technology can effectively enhance the spatial features of subtle change areas, obtain a glacier remote sensing image change detection map, and improve the accuracy of glacier change detection. However, it is unable to effectively integrate and analyze multiple data, and it is difficult to comprehensively and accurately evaluate the reliability of glacier change information. Summary of the Invention

[0005] The inventors found through research that: with the change of the global climate, glaciers, as sensitive indicators of climate change, have a profound impact on aspects such as sea-level rise, water resource supply, and ecological system balance. Accurately obtaining and evaluating watershed glacier change information is of great significance for predicting climate change trends, rationally planning water resource utilization, and protecting the ecological environment. Traditional glacier monitoring methods, such as field measurements, although able to obtain relatively accurate data, are limited by geographical conditions, with a limited measurement range and low efficiency. Remote sensing monitoring methods, although able to achieve large-area monitoring, have certain limitations in terms of accuracy and the capture of glacier detail changes.

[0006] The object of the present invention is to provide a method and system for evaluating the reliability of watershed glacier change information based on big data. By fusing a type-I evaluation model and a type-II evaluation model to obtain a reliability evaluation model, the reliability of watershed glacier change information is evaluated, so as to solve the technical problem that the prior art cannot provide a method for evaluating the reliability of multiple data in glacier change.

[0007] According to one aspect of the present invention, there is provided a method for evaluating the reliability of watershed glacier change information based on big data, which is executed by a processor. Obtain a glacier standard morphology atlas of the target area with marked sorting. Perform segmentation on the standard morphology atlas to obtain a training set, a first test set, and a second test set. Upload the training set and the first test set to the hierarchical iterative model for training to obtain a type-I evaluation model. Upload the training set and the second test set to the hierarchical iterative model for training to obtain a type-II evaluation model. Fuse the type-I evaluation model and the type-II evaluation model to obtain a reliability evaluation model. According to the reliability evaluation model, perform the reliability evaluation of the watershed glacier change information.

[0008] In some embodiments, the process of obtaining the glacier standard morphology atlas of the target area with marked sorting is as follows: Connect to a high-resolution satellite to obtain at least images of the standard area, standard contour, and conventional surface features of the glaciers in the target area. Perform bubble sorting on the images of the standard area, standard contour, and conventional surface features of the glaciers to obtain time-sequence images from near to far. Group the images of the standard area, standard contour, and conventional surface features of the glaciers at the same time sequence. Perform data cleaning on each group of images to obtain the glacier standard morphology atlas of the target area.

[0009] In some embodiments, the process of performing segmentation on the standard morphology atlas to obtain a training set, a first test set, and a second test set is as follows: According to the machine learning algorithm, perform proportional data segmentation on the glacier standard morphology atlas of the target area to obtain a training set, a first test set, and a second test set, where the ratio of the training set, the first test set, and the second test set is at least 5:4:1.

[0010] In some embodiments, the process of uploading the training set and the first test set to the hierarchical iterative model for training to obtain a type-I evaluation model is as follows: Upload the training set to the hierarchical iterative model and perform optimization. Upload the first test set to the optimized hierarchical iterative model according to the sample definition, adjust the parameters of the hierarchical iterative model, and obtain the optimal hierarchical iterative model; Perform uncertainty analysis on the optimal hierarchical iterative model to obtain a type-I evaluation model.

[0011] In some embodiments, the process of uploading the training set and the second test set to the hierarchical iterative model for training to obtain a type-II evaluation model is as follows: Upload the training set to the hierarchical iterative model and perform optimization; Upload the second test set to the optimized hierarchical iterative model according to the sample definition, adjust the parameters of the hierarchical iterative model, and obtain the optimal hierarchical iterative model; Perform uncertainty analysis on the optimal hierarchical iterative model to obtain a type-II evaluation model.

[0012] In some embodiments, the process of fusing the type-I evaluation model and the type-II evaluation model to obtain a reliability evaluation model is as follows: Perform data fusion on the first test set and the second test set to obtain a combined test set; Perform data cleaning on the combined test set to obtain an interference-free test set, and upload this test set as the training set to the type-I evaluation model and the type-II evaluation model to obtain the first evaluation data; Construct a meta-model, use the first evaluation data as the input layer pending features, perform training, and obtain the second evaluation data; Compare the first evaluation data with the second evaluation data to obtain the optimal evaluation data; Use the optimal evaluation data as the training set, upload it to the meta-model, optimize the parameters, and obtain the reliability evaluation model.

[0013] In some embodiments, the process of performing reliability evaluation on the basin glacier change information according to the reliability evaluation model is as follows: Obtain the glacier characterization change image data corresponding to different time sequences in the target area, and perform reliability evaluation on the basin glacier change information based on the reliability evaluation model.

[0014] According to another aspect of the present invention, a reliability evaluation system for basin glacier change information based on big data is provided. The system includes a processor and further includes: An acquisition module, which is used to obtain the glacier standard morphology atlas of the target area that has been labeled and sorted; A preprocessing module, which is used to perform segmentation on the standard morphology atlas to obtain a training set, a first test set, and a second test set; A first training module, which is used to upload the training set and the first test set to the hierarchical iterative model for training to obtain a type-I evaluation model; The second training module is used to upload the training set and the second test set to the hierarchical iterative model for training to obtain a type-II evaluation model; The merging training module is used to fuse the type-I evaluation model and the type-II evaluation model to obtain a reliability evaluation model; The execution module is used to perform the reliability evaluation of the basin glacier change information according to the reliability evaluation model.

[0015] In some embodiments, there are a processor data connection acquisition module, a preprocessing module, a first training module, a second training module, a merging training module, and an execution module.

[0016] Compared with the prior art, the present invention has the following advantages and beneficial effects: obtaining an atlas of standard glacier morphologies of target regions with labeled rankings can improve the evaluation accuracy of the method of the present invention during reliability evaluation; performing segmentation on the atlas of standard morphologies to obtain a training set, a first test set, and a second test set, different from the traditional machine learning segmentation method, adopting a more detailed test set division to achieve the precision of model training; uploading the training set and the first test set to the hierarchical iterative model for training to obtain a type-I evaluation model; uploading the training set and the second test set to the hierarchical iterative model for training to obtain a type-II evaluation model, by performing two different datasets of training to obtain different iterative training models, realizing the matching of multi-data information in glacier changes; fusing the type-I evaluation model and the type-II evaluation model to obtain a reliability evaluation model, and performing the reliability evaluation of the basin glacier change information according to the reliability evaluation model, fusing the type-I and type-II evaluation models to obtain a reliability evaluation model to achieve a more comprehensive and accurate analysis of the glacier change information, effectively solving the problems of incomplete and inaccurate results caused by single data processing in the prior art. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained according to these drawings.

[0018] Figure 1 is the flowchart of the evaluation method of the present invention; Figure 2 is the structural diagram of the evaluation system of the present invention; Figure 3 is the usage scenario diagram of the evaluation system of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0019] The following will be combined with the accompanying drawings in the embodiments of the present inventionFigures 1 - 3 The technical solutions in the embodiments of the present invention are clearly and completely described below. Apparently, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Embodiment

[0020] Figure 1 FIG. is a flowchart of a method for evaluating the reliability of watershed glacier change information based on big data provided in this embodiment. This method is executed by a processor. The so-called processor may be a central processing unit (CPU), or may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0021] The method specifically includes: Obtain a labeled and sorted atlas of standard glacier morphologies in the target area. Using the characterization image of the glacier in the target area as the image source, based on bubble sort, the characterization images are arranged in chronological order to obtain chronological labeled images with the time order from near to far, which is convenient for image management. It can be understood that the chronologically labeled images can help the model or observers find the standard characterization images of the glacier at different time points more quickly and accurately.

[0022] In some possible embodiments, a high-resolution satellite is connected to obtain at least the standard area, standard contour, and conventional surface feature images of the glaciers in the target area. For example, by using Landsat satellites in different periods, high-resolution images are obtained, which are at least the standard area, standard contour, and conventional surface feature images of the glaciers in the target area. These images can visually observe the changes in the glacier boundaries. At the same time, synthetic aperture radar satellite data is used to obtain information on the internal structure and thickness changes of the glaciers. Then, bubble sorting is performed on the images of the standard area, standard contour, and conventional surface features of the glaciers to obtain time-series images from near to far. For the images of the standard area, standard contour, and conventional surface features of the glaciers at the same time series, grouping is performed. In this embodiment, the images of different standard areas, standard contours, and conventional surface features of the glaciers are classified based on the same time, and at least the images of the past 5 years are classified. It should be noted that for some glaciers in geologically active areas, researchers may conduct short-term image collection for 3 - 5 years to quickly evaluate the impact of geological changes on the glaciers. Therefore, this embodiment selects at least 5 years as the period. Finally, data cleaning is performed on each group of images to obtain the standard morphological atlas of the glaciers in the target area. Among them, the data cleaning in this embodiment is at least to remove noise data, outliers, and duplicate data. For example, for abnormal pixels in satellite remote sensing data caused by cloud occlusion or sensor failures, interpolation algorithms are used for repair; for jump values in ground monitoring data, they are corrected or removed through comparative analysis with data in adjacent time periods.

[0023] Segment the standard morphological atlas to obtain a training set, a first test set, and a second test set. Here, the principle of segmentation adopts the segmentation principle of the training set and test set in machine learning, but this embodiment optimizes the segmentation ratio to further assist in improving the accuracy of the final reliability assessment.

[0024] In some possible embodiments, according to the machine learning algorithm, the standard morphological atlas of the glaciers in the target area is segmented into fixed-ratio data to obtain a training set, a first test set, and a second test set, where the ratio of the training set, the first test set, and the second test set is at least 5:4:1. The ratio of the first test set to the second test set is 4:1 because the inventors found in actual work that the testing of general models, that is, the optimization of parameters, is very limited to single data and cannot be optimized multiple times, which leads to a high uncertainty in the model optimization results. Using a ratio of 4:1 can effectively avoid this problem. Of course, in other possible fields and situations, different ratios can be implemented to adapt to other training environments.

[0025] Upload the training set and the first test set to the hierarchical iterative model for training to obtain a type-I evaluation model. Upload the training set and the second test set to the hierarchical iterative model for training to obtain a type-II evaluation model. Upload the first test set and the second test set correspondingly to obtain different evaluation models. Here, it can be understood that the focuses of evaluation of the type-I evaluation model and the type-II evaluation model are different. Due to the differences in the test sets, the training focuses of the models are also different. Therefore, there is a complementary relationship between the two, but it is different from the existing unified test set training of the same model. The existing unified training set does not focus on a certain point for training, but shows a uniform distribution effect, which leads to the application after model training often not having obvious corresponding advantages. In the glacier change scenario of this embodiment, the model has a focus, which will greatly improve the accuracy of the subsequent reliability evaluation results.

[0026] In some possible implementation manners, upload the training set to the hierarchical iterative model and perform optimization; Upload the first test set to the optimized hierarchical iterative model according to the sample definition, adjust the parameters of the hierarchical iterative model to obtain the optimal hierarchical iterative model; perform uncertainty analysis on the optimal hierarchical iterative model to obtain a type-I evaluation model. Similarly, upload the training set to the hierarchical iterative model and perform optimization; upload the second test set to the optimized hierarchical iterative model according to the sample definition, adjust the parameters of the hierarchical iterative model to obtain the optimal hierarchical iterative model.

[0027] Perform uncertainty analysis on the optimal hierarchical iterative model to obtain a type-II evaluation model. Considering the uncertainty existing in the data collection of glacier change itself, in this embodiment, uncertainty analysis is performed on the optimal hierarchical iterative model, and the fluctuation range and confidence interval of the analysis results are obtained through multiple simulation calculations. For example, in this embodiment, the Monte Carlo simulation method is used to randomly perturb the standard morphological atlas of glaciers in the target area, run the optimal hierarchical iterative model multiple times, and statistically analyze the distribution of the results, so as to obtain the uncertainty range of the results.

[0028] Fuse the type-I evaluation model and the type-II evaluation model to obtain a reliability evaluation model. Model fusion combines the advantages of multiple models to improve the performance and generalization ability of the model. Common implementation methods include the averaging method, the voting method, the stacking method, the fusion tree, etc. In this embodiment, the stacking method is used to fuse the type-I evaluation model and the type-II evaluation model to obtain a reliability evaluation model.

[0029] In some possible implementation manners, perform data fusion on the first test set and the second test set to obtain a combined test set; perform data cleaning on the combined test set to obtain an interference-free test set, and upload this test set to the type-I evaluation model and the type-II evaluation model as the training set to obtain the first evaluation data; Construct a meta-model, use the first evaluation data as the input layer to-be-entered features, perform training, and obtain the second evaluation data; compare the first evaluation data with the second evaluation data to obtain the optimal evaluation data; use the optimal evaluation data as the training set, upload it to the meta-model, optimize the parameters, and obtain the reliability evaluation model.

[0030] A meta-model is a model about models, which defines the rules, concepts, relationships, etc. followed when constructing specific models. Since there may be multiple different models in a complex system or domain, the meta-model can provide unified standards and specifications for these models to ensure the consistency and compatibility among various models. Common existing meta-models are E-R meta-model, UML meta-model, and MOF meta-model. In this embodiment, the MOF meta-model is selected. The first evaluation data is used as the input layer to-be-entered features in the MOF meta-model. The meta-model will determine the standard, that is, determine the evaluation standard, and then use this standard as the output object to compare with the first evaluation data. Through comparison, the more complete and accurate data between the two is analyzed as the optimal evaluation data. For example, the first evaluation data includes a glacier corner collapse area of 2 square meters and a thickness change of 50 centimeters, and the second evaluation data includes a glacier corner collapse area of about 2 square meters and a thickness change of about 50 centimeters. At this time, since the first evaluation data is more accurate, the first evaluation data is determined as the optimal evaluation data, and so on. The optimal evaluation data is used as the training set, uploaded to the meta-model, and the corresponding parameters are optimized according to the machine learning model optimization logic to obtain the reliability evaluation model.

[0031] According to the reliability evaluation model, perform the reliability evaluation of the basin glacier change information.

[0032] In some possible implementation manners, obtain the glacier characterization change image data corresponding to different time sequences of the target area, and perform the reliability evaluation of the basin glacier change information based on the reliability evaluation model. It should be noted here that the reliability evaluation model will perform the reliability evaluation of the basin glacier change information of the corresponding target area. Embodiment

[0033] Figure 2This is the structural diagram of the evaluation system of the present invention. Exemplarily, the method can be divided into one or more modules. One or more modules are stored in the memory and executed by the processor to implement the present invention. One or more modules can be a series of computer program instruction segments capable of performing specific functions, and these instruction segments are used to describe the execution process of the computer program. For example, the computer program can be divided into an acquisition module, a preprocessing module, a first training module, a second training module, a combined training module, and an execution module. The specific functions of each module are as follows: The acquisition module is used to obtain a labeled and sorted atlas of the standard forms of glaciers in the target area; the preprocessing module is used to perform segmentation on the atlas of standard forms to obtain a training set, a first test set, and a second test set; the first training module is used to upload the training set and the first test set to the hierarchical iterative model for training to obtain a type-I evaluation model; the second training module is used to upload the training set and the second test set to the hierarchical iterative model for training to obtain a type-II evaluation model; the combined training module is used to fuse the type-I evaluation model and the type-II evaluation model to obtain a reliability evaluation model; the execution module is used to perform the reliability evaluation of the watershed glacier change information according to the reliability evaluation model.

[0034] In some possible implementation manners, the processor is data-connected to the acquisition module, the preprocessing module, the first training module, the second training module, the combined training module, and the execution module.

[0035] The specific example of the method for evaluating the reliability of watershed glacier change information based on big data in the foregoing Embodiment 1 is also applicable to the system for evaluating the reliability of watershed glacier change information based on big data in this embodiment. Through the foregoing detailed description of the method for evaluating the reliability of watershed glacier change information based on big data, those skilled in the art can clearly know the system for evaluating the reliability of watershed glacier change information based on big data in this embodiment. Therefore, for the sake of brevity of the specification, it will not be described in detail herein.

[0036] The above shows and describes the basic principles, main features, and advantages of the present invention. For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and without departing from the spirit or basic features of the present invention, the present invention can be implemented in other specific forms. Therefore, in any aspect, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, it is intended to embrace all changes within the meaning and scope of the equivalent elements of the claims in the present invention. Any reference signs in the claims should not be regarded as limiting the claimed rights.

[0037] In addition, it should be understood that although this specification is described in terms of embodiments, not every embodiment contains only an independent technical solution. This narrative style of the specification is only for clarity. Those skilled in the art should regard the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A reliability assessment method for basin glacier change information based on big data, the method being executed by a processor and characterized in that: Obtain a collection of standard glacier morphology atlases in the target area that have been annotated and sorted; Perform segmentation on the standard morphological atlas to obtain a training set, a first test set, and a second test set; Upload the training set and the first test set to the hierarchical iterative model training to obtain a type 1 evaluation model; Upload the training set and the second test set to the hierarchical iterative model training to obtain the type II evaluation model; The reliability assessment model is obtained by integrating the type 1 assessment model with the type 2 assessment model; Based on the reliability assessment model, the reliability assessment of the basin glacier change information is performed.

2. The method according to claim 1, characterized in that The process of obtaining the glacier standard morphology atlas of the target area that has been marked and sorted is as follows: Connect to high-resolution satellites to obtain at least standard glacier area, standard contours and conventional surface feature images of the target area; Bubble sorting is performed on the images of the standard area, standard outline and conventional surface features of the glacier to obtain time-series images from near to far; Grouping of glacier standard area, standard outline and conventional surface feature images of the same time series; Data cleaning is performed on each set of images to obtain a standard morphological atlas of glaciers in the target area.

3. The method according to claim 2, characterized in that The process of performing segmentation on the standard morphological atlas to obtain the training set, the first test set and the second test set is as follows: According to the machine learning algorithm, the standard morphological atlas of glaciers in the target area is subjected to fixed-ratio data segmentation to obtain a training set, a first test set and a second test set, wherein the ratio of the training set, the first test set and the second test set is at least 5:4:

1.

4. The method according to claim 3, characterized in that The process of uploading the training set and the first test set to the hierarchical iterative model training to obtain a type of evaluation model is as follows: Upload the training set to the hierarchical iterative model and perform optimization; Upload the first test set to the optimized hierarchical iterative model according to the sample definition, adjust the hierarchical iterative model parameters, and obtain the optimal hierarchical iterative model; Uncertainty analysis is performed on the optimal hierarchy iteration model to obtain a type-one evaluation model.

5. The method according to claim 4, characterized in that The process of uploading the training set and the second test set to the hierarchical iterative model training to obtain the type II evaluation model is as follows: Upload the training set to the hierarchical iterative model and perform optimization; Upload the second test set to the optimized hierarchical iteration model according to the sample definition, adjust the hierarchical iteration model parameters, and obtain the optimal hierarchical iteration model; Uncertainty analysis was performed on the optimal hierarchical iteration model to obtain a type-II evaluation model.

6. The method according to claim 1, characterized in that The process of integrating the type 1 evaluation model with the type 2 evaluation model to obtain the reliability evaluation model is as follows: Perform data fusion on the first test set and the second test set to obtain a combined test set; Perform data cleaning on the combined test set to obtain a non-interference test set, and upload the test set as a training set to the type-one evaluation model and the type-two evaluation model to obtain first evaluation data; Constructing a meta-model, taking the first evaluation data as input layer features, performing training, and obtaining second evaluation data; Comparing the first evaluation data with the second evaluation data to obtain optimal evaluation data; The optimal evaluation data is used as a training set, uploaded to the meta-model, and the parameters are optimized to obtain a reliability evaluation model.

7. The method according to claim 6, characterized in that The process of performing reliability assessment of basin glacier change information according to the reliability assessment model is as follows: Acquire glacier characterization change image data corresponding to different time series in the target area, and perform reliability assessment of basin glacier change information based on the reliability assessment model.

8. A basin glacier change information reliability assessment system based on big data, the system comprising a processor, characterized in that: Also includes: An acquisition module, the acquisition module is used to obtain a standard morphological atlas of glaciers in a target area that has been marked and sorted; A preprocessing module, the preprocessing module is used to perform segmentation on the standard morphological atlas to obtain a training set, a first test set and a second test set; A first training module, wherein the first training module is used to upload a training set and a first test set to hierarchical iterative model training to obtain a type-one evaluation model; A second training module, wherein the second training module is used to upload the training set and the second test set to hierarchical iterative model training to obtain a type-two evaluation model; A combined training module, wherein the combined training module is used to merge the type 1 evaluation model with the type 2 evaluation model to obtain a reliability evaluation model; An execution module is used to perform reliability assessment of basin glacier change information according to a reliability assessment model.

9. The system according to claim 8, characterized in that The processor is data connected to the acquisition module, the preprocessing module, the first training module, the second training module, the combined training module and the execution module.

Images