Image data processing method based on big data and artificial intelligence generation system technology

Abstract

The invention relates to the technical field of remote sensing image processing, and discloses an image data processing method based on a big data and artificial intelligence generation system technology, and the method comprises the following steps: 100, collecting motion estimation information and glacier animal habitat information; step 200, coding the motion estimation information into first graph structure data; and step 300, inputting the first graph structure data into the motion estimation model, outputting a result representing the motion pattern classification of the glacier by a first output layer, and outputting a result representing the surface feature movement of the glacier by a second output layer. According to the method, the glacier motion mode and the surface feature movement result are comprehensively analyzed, and the dynamic features, the material balance state, the response to environment change and the influence on regional water resources of the glacier can be comprehensively understood.

Description

Technical Field

[0001] The present invention relates to the field of remote sensing image processing, and more specifically, to an image data processing method based on a generation system technology of big data and artificial intelligence. Background Art

[0002] Glaciers refer to natural ice bodies that exist on the surface of polar or alpine areas for many years and move along the ground. Glaciers are formed by years of snow accumulation, compaction, recrystallization, refreezing and other ice-forming processes. They have certain shapes and layers, and are plastic. Under gravity and pressure, they produce plastic flow and block sliding, and are important freshwater resources on the surface.

[0003] In particular, polar glaciers have complex nonlinear motion behaviors such as glacier surges and tidal influences, which are difficult to describe with simple models. At the same time, the movement in areas such as glacier edges and cracks is discontinuous, resulting in instability in the estimation of the motion field. In addition, some areas such as the bottom of ice shelves have large-scale three-dimensional flows. Simple surface motion analysis may not be comprehensive enough, which will cause difficulties in glacier research, disaster risk assessment and water resources management research. Summary of the invention

[0004] The present invention provides an image data processing method based on a generation system technology of big data and artificial intelligence to solve technical problems in related technologies.

[0005] The present invention provides an image data processing method based on big data and artificial intelligence generation system technology, comprising the following steps:

[0006] Step 100, collecting motion estimation information and glacier animal habitat information; the motion estimation information includes optical and SAR satellite image information, satellite image imaging information, glacier information and bed information; the glacier animal habitat information includes animal species, animal habitat location, animal habitat area and animal number;

[0007] Step 200, encoding the motion estimation information into first graph structure data, the first graph structure data including nodes and edges connecting the nodes, a node representing a satellite image, a glacier, a bottom bed or a satellite;

[0008] Constructing second graph structure data, the second graph structure data includes nodes and edges connecting the nodes, one node represents one animal or habitat; encoding the glacier animal habitat information of each node into sequence data, the sequence data includes n sequence units, and the t-th sequence unit represents the glacier animal habitat information of a node at the t-th moment;

[0009] Step 300, inputting the first graph structure data into a motion estimation model, the motion estimation model comprising a first hidden layer, a first output layer, and a second output layer, the first output layer outputting a result representing the classification of the motion pattern of the glacier, and the second output layer outputting a result representing the movement of the surface features of the glacier;

[0010] The sequence data, the output results of the first output layer, and the output results of the second output layer are input into the input habitat assessment model, and the third output layer outputs the rate of change of the species core area;

[0011] Step 400, based on the output results of the motion pattern classification and the results of the movement of the surface features of the glacier, evaluate the sensitivity of the glacier to environmental changes, the stability and potential risks of the glacier, or the impact of glacier changes on regional water resources.

[0012] Furthermore, the satellite image imaging information includes: satellite orbit parameters, imaging geometry information, and ground control point information of the imaging satellite;

[0013] Glacier information: glacier boundary, glacier surface elevation, glacier surface velocity, glacier thickness, glacier mass balance, glacier surface temperature, glacier surface albedo and glacier surface morphology;

[0014] Substrate information: bedrock topography, bedrock elevation, bedrock slope, bedrock roughness, bedrock type, bedrock temperature, and bedrock stress.

[0015] Furthermore, there are edges between the node representing the animal and all the nodes representing the habitat, there are edges between the nodes corresponding to all the animals connected to the same node representing the habitat, and there are edges between all the nodes representing the habitat.

[0016] Furthermore, there is an edge between the node representing the glacier and the node representing the bed in the same satellite image;

[0017] There is an edge between a node representing a glacier in a satellite image and a node representing a glacier in a satellite image at a previous timestamp;

[0018] If the similarity between a node representing a glacier in a satellite image and a node representing a glacier in a satellite image at a previous timestamp exceeds a set threshold, it is determined that there is an edge between the two;

[0019] If the similarity between the nodes representing the bed in the satellite image exceeds the set threshold, it is determined that there is an edge between the two;

[0020] Satellite Imagery There is an edge between the image and the satellite.

[0021] Furthermore, there are edges between nodes representing animals and all nodes representing habitats, there are edges between nodes corresponding to all animals in the same node representing habitats, and there are edges between all nodes representing habitats;

[0022] The habitat assessment model was run twice. In the first run, the first feature fusion layer input the second hidden state and the default feature. The size of the default feature was consistent with the second hidden state. The first output layer output the initial species core area change rate.

[0023] During the second run, the first feature fusion layer inputs the second hidden state, the fourth hidden state, and the fifth hidden state, and the third output layer outputs the area change rate of the core area of ​​the current species.

[0024] Furthermore, the calculation formula of the first hidden layer is as follows:

[0025] ;

[0026] in represents the first hidden state of the vth node, represents the graph weight parameter in the first hidden layer, and Respectively represent the node features of the vth and uth nodes of the first graph structure data, is the set of nodes that have edges with the node v of the first graph structure data, express The total number of nodes in ;

[0027] ;

[0028] in represents the aggregation coefficient between the vth and uth nodes, tanh represents the hyperbolic tangent function, T represents the transpose, Indicates adjustable parameters; The default value of is 1 / N, where N is the dimension of the node feature;

[0029] The calculation formula of the first output layer is as follows:

[0030] ;

[0031] in represents the first output vector, and the i-th component of the first output vector represents the probability that the movement mode of the glacier belongs to the i-th state. represents the vth first hidden state of the input first output layer, It means concatenating the first hidden states of all nodes in the first output layer. represents the set of all nodes, is the weight parameter in the first output layer, is the bias parameter in the first output layer, represents the sigmoid function;

[0032] The calculation formula of the second output layer is as follows:

[0033] ;

[0034] in represents the second output vector, and the i-th component of the second output vector represents the probability value of the i-th action; the set of all actions is called the action space, which contains all possible actions that may be selected;

[0035] In one embodiment of the present invention, an action is represented as a vector, and the jth component of the vector represents the jth representation parameter of the movement of the surface feature of the glacier;

[0036] An action is represented as a matrix. The first row and the Lth column of the matrix represent the representation parameters of the Lth displacement vector, the second row and the Lth column represent the representation parameters of the Lth velocity field, the third row and the Lth column represent the representation parameters of the Lth strain rate, the fourth row and the Lth column represent the representation parameters of the Lth surface elevation change, and the fifth row and the Lth column represent the representation parameters of the Lth surface morphology change. represents the first hidden state of the vth node in the second output layer, represents the concatenation of the first hidden states of all nodes in the second output layer, Z represents the set of nodes related to the recognition results, is the weight parameter in the second output layer, is the bias parameter in the second output layer, Represents a sigmoid function.

[0037] Furthermore, the calculation formula of the second hidden layer is as follows:

[0038] ;

[0039] in, The second graph structure data identification feature of the dth node of the lth layer of the tth sequence unit of the sequence data, The second graph structure data identification feature of the u-th node of the t-th sequence unit of the sequence data of the l-1-th layer, and They represent the set of nodes with high association with node d and node u respectively, represents the cardinality of a set, The weight matrix representing the second graph structure data of the lth layer, , E represents the total number of layers, when l=1 , The feature representation of the object representing the u-th node index of the t-th sequence unit of the sequence data, is the sigmoid function;

[0040] The calculation formula for the fourth hidden layer is as follows:

[0041] ;

[0042] ;

[0043] in, represents the fourth hidden state, Information representing the movement pattern of the input glacier, represents the first convolution function;

[0044] The calculation formula for the fifth hidden layer is as follows:

[0045] ;

[0046] ;

[0047] in, represents the fifth hidden state, Information representing the movement of the surface features of the input glacier, represents the second convolution function.

[0048] Furthermore, the calculation formula of the first feature fusion layer is as follows:

[0049] ;

[0050] in, represents the first fusion state, The second graph structure data identification feature of the last layer of the dth node output when the tth sequence unit of the sequence data is input into the second hidden layer, represents the set of all units of the second graph structure data, represents the concatenation function, represents the sum weight matrix, represents the summation bias parameter;

[0051] The calculation formula for the fifth hidden layer is as follows:

[0052] ;

[0053] ;

[0054] ;

[0055] ;

[0056] in, , , represents the first, second, and third weight matrices in the fifth hidden layer, , , represents the first, second, and third bias parameters in the fifth hidden layer, represents the dot product, , and Respectively represent the first, second, and third intermediate states of the fifth hidden layer, , and They represent the t-th and t-1-th fifth hidden states, respectively, n≥t≥1, n represents the total number of sequence units of the sequence data, and when t=1 , tanh is the hyperbolic tangent function, Represents the sigmoid function;

[0057] The calculation formula of the third output layer is as follows:

[0058] ;

[0059] in Represents the output matrix, an element of which represents the change rate of the core area of ​​a species near a glacier. represents the g-th fifth hidden state, Indicates the concatenation of the fifth hidden states of all nodes. represents the set of all nodes, is the weight parameter in the third output layer, is the bias parameter in the third output layer, Represents a sigmoid function.

[0060] The present invention also proposes an image data processing system based on big data and artificial intelligence generation system technology, which executes the steps in the aforementioned image data processing method based on big data and artificial intelligence generation system technology, including:

[0061] Data collection module: A generation system based on big data and artificial intelligence technology collects movement estimation information and glacier animal habitat information;

[0062] Data encoding module: constructing first graph structure data, sequence data and second graph structure data, including sequences or nodes and edges, and establishing corresponding motion estimation models and habitat assessment models as inputs for each glacier and bottom bed and habitat and animal;

[0063] Data processing module: input the encoded first graph structure data, sequence data and second graph structure data into the motion estimation model and habitat assessment model, and output the corresponding motion pattern classification results and the surface feature movement results of the glacier;

[0064] Result evaluation module: The dynamic characteristics, material balance state, response to environmental changes and impact on regional water resources of the glacier are evaluated through the corresponding movement pattern classification results and the results of the movement of the glacier's surface features. At the same time, it can also be combined with the glacier animal habitat information to obtain the area change rate of the species core area and evaluate the impact of glacier changes on species near the glacier.

[0065] A storage medium storing non-temporary computer-readable instructions for executing the steps in the aforementioned image data processing method based on big data and artificial intelligence generation system technology.

[0066] The beneficial effects of the present invention are:

[0067] The present invention comprehensively analyzes the results of glacier movement patterns and surface feature shifts, which can provide a more comprehensive understanding of the glacier's dynamic characteristics, material balance state, response to environmental changes, and impact on regional water resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Figure 1 It is a flow chart of an image data processing method based on a generation system technology of big data and artificial intelligence proposed by the present invention.

[0069] Figure 2 It is a structural block diagram of an image data processing system based on big data and artificial intelligence generation system technology proposed by the present invention.

[0070] In the figure: 101, data acquisition module; 102, data encoding module; 103, data processing module; 104, result evaluation module. DETAILED DESCRIPTION

[0071] The subject matter described herein will now be discussed with reference to example embodiments. It should be understood that the discussion of these embodiments is only to enable those skilled in the art to better understand and implement the subject matter described herein, and the functions and arrangements of the elements discussed may be changed without departing from the scope of protection of the contents of this specification. Each example may omit, replace or add various processes or components as needed. In addition, the features described relative to some examples may also be combined in other examples.

[0072] like Figure 1 As shown, an image data processing method based on big data and artificial intelligence generation system technology includes.

[0073] Step 100, collecting motion estimation information and glacier animal habitat information; the motion estimation information includes optical and SAR satellite image information, satellite image imaging information, glacier information and bed information; the glacier animal habitat information includes animal species, animal habitat location, animal habitat area and animal number;

[0074] Satellite imagery imaging information includes: satellite orbit parameters, imaging geometry information, and ground control point information of the imaging satellite;

[0075] Glacier information: glacier boundary, glacier surface elevation, glacier surface velocity, glacier thickness, glacier mass balance, glacier surface temperature, glacier surface albedo and glacier surface morphology;

[0076] Substrate information: bedrock topography, bedrock elevation, bedrock slope, bedrock roughness, bedrock type, bedrock temperature and bedrock stress;

[0077] Step 200, encoding the motion estimation information into first graph structure data, the first graph structure data including nodes and edges connecting the nodes, a node representing a satellite image, a glacier, a bottom bed or a satellite;

[0078] There are edges between nodes representing glaciers and nodes representing the bed in the same satellite image;

[0079] There is an edge between a node representing a glacier in a satellite image and a node representing a glacier in a satellite image at a previous timestamp;

[0080] If the similarity between a node representing a glacier in a satellite image and a node representing a glacier in a satellite image at a previous timestamp exceeds a set threshold, it is determined that there is an edge between the two;

[0081] If the similarity between the nodes representing the bed in the satellite image exceeds the set threshold, it is determined that there is an edge between the two;

[0082] There are edges between satellite imagery images and satellites;

[0083] constructing second graph structure data, the second graph structure data including nodes and edges connecting the nodes, wherein one node represents one animal or habitat;

[0084] There are edges between the nodes representing animals and all the nodes representing habitats, there are edges between the nodes corresponding to all animals in the same node representing habitat, and there are edges between all the nodes representing habitats.

[0085] The glacier animal habitat information of each node is encoded into sequence data, the sequence data includes n sequence units, and the t-th sequence unit represents the glacier animal habitat information of a node at the t-th moment;

[0086] Step 300, inputting the first graph structure data into a motion estimation model, the motion estimation model comprising a first hidden layer, a first output layer, and a second output layer, the first output layer outputting a result representing the classification of the motion pattern of the glacier, and the second output layer outputting a result representing the movement of the surface features of the glacier;

[0087] In one embodiment of the present invention, the glacier movement mode classification results include block sliding, internal deformation, creep and periodic surge;

[0088] It should be added that block sliding means that the bottom of the glacier slides along the bottom surface under the action of gravity, which is the main movement mode of valley glaciers;

[0089] Internal deformation refers to plastic deformation inside the glacier, where the upper ice layer moves relative to the lower ice layer, causing the entire glacier to deform.

[0090] Creep refers to the slow creep deformation of the ice body under the stress of the glacier's own weight;

[0091] Periodic surges refer to the phenomenon that some glaciers exhibit periodic alternations of rapid advance and relative stillness.

[0092] In one embodiment of the present invention, the result of the movement of the surface features of the glacier can be measured by one or more combinations of displacement vector, velocity field, strain rate, surface elevation change and surface morphology change. The result can be expressed as one or more parameter combinations of the movement direction and distance of the glacier as a whole, the movement speed and direction of the glacier surface, the degree and direction of the glacier deformation, the change of the morphological features of the glacier surface, and the change of the height or volume of the glacier surface.

[0093] It should be added that the displacement vector uses a vector to represent the direction and distance of movement of glacier surface features (such as cracks, glacial wells, etc.);

[0094] The starting point of the vector represents the initial position of the feature, and the end point represents the final position of the feature;

[0095] The length of the vector indicates how far the feature has moved, and the direction indicates the direction of movement.

[0096] Among them, the velocity field uses the velocity vector field to represent the movement speed and direction of different positions on the glacier surface.

[0097] Each grid point, or pixel, has a velocity vector that represents the speed and direction of movement of the glacier surface at that point;

[0098] Velocity fields can be used to analyze the spatial patterns and temporal variations of glacier surface motion.

[0099] Among them, the strain rate is the strain rate calculated in different areas of the glacier surface, which indicates the degree and direction of glacier deformation.

[0100] The strain rate can be decomposed into the lateral strain rate (indicating the lateral extension or compression of the glacier) and the longitudinal strain rate (indicating the longitudinal extension or compression of the glacier);

[0101] Positive values ​​of strain rate indicate tension, and negative values ​​indicate compression.

[0102] Among them, the surface elevation change is achieved by repeatedly measuring the glacier surface elevation and calculating the elevation changes between different periods.

[0103] Changes in elevation can be expressed as vertical displacements or thickness changes on the glacier surface;

[0104] Positive values ​​indicate that the glacier surface is rising or thickening, while negative values ​​indicate that the glacier surface is sinking or thinning.

[0105] Among them, surface morphological changes are the analysis of changes in the surface morphological characteristics of the glacier (such as glacier cracks, glacier wells, glacier springs, etc.).

[0106] Calculate the changes in the number, density, direction and other indicators of the glacier surface morphological features in different periods. The changes in surface morphology can reflect the changes in glacier dynamic processes and environmental conditions.

[0107] Among them, mass balance is the calculation of the mass balance of the glacier surface, which is the difference between accumulation (such as snowfall) and ablation (such as melting and sublimation).

[0108] Mass balance can be expressed as changes in glacier surface height or volume;

[0109] Positive values ​​indicate glacier mass gain, and negative values ​​indicate glacier mass loss.

[0110] In one example of the present invention, a glacier located in Antarctica needs to be studied. Through GPS measurement and remote sensing image analysis of the glacier surface, the following results of the movement of the glacier surface features are obtained:

[0111] Displacement vectors: Tracking of large cracks on the glacier surface revealed that the cracks moved about 200 meters downstream in a year; the direction of the displacement vectors pointed downstream of the glacier, indicating that the glacier as a whole was flowing downstream;

[0112] Velocity field: The velocity field on the glacier surface was calculated by repeated measurements of characteristic points on the glacier surface. The velocity field showed that the velocity was the highest at the center line of the glacier, reaching 300 meters per year, while the velocity was slower near the edge of the glacier, about 50 meters per year. The spatial pattern of the velocity field showed that the flow of the glacier was mainly affected by the topography and the glacier bed.

[0113] Strain rate: The strain rate of the glacier surface is calculated based on the velocity field; the results show that the lateral strain rate of the upstream part of the glacier is positive, indicating that the glacier is stretched in the lateral direction; the longitudinal strain rate of the downstream part of the glacier is negative, indicating that the glacier is compressed in the longitudinal direction.

[0114] Surface elevation changes: Changes in the glacier surface elevation were obtained through repeated aerial lidar measurements; the results showed that the surface elevation of the upstream part of the glacier dropped by 2 meters in one year, indicating that the glacier is thinning; the surface elevation of the downstream part of the glacier changed less, indicating that the glacier thickness is relatively stable.

[0115] Changes in surface morphology: Analysis of high-resolution satellite images revealed that the number and density of cracks on the glacier surface increased within a year; new cracks mainly appeared in the upstream part of the glacier, indicating that glacier movement in this area is more active; at the same time, a decrease in the number of glacial wells on the glacier surface was observed, indicating that the intensity of glacier ablation has increased.

[0116] Through the combination of the above parameters, the results of the movement of glacier surface features are comprehensively evaluated, and the dynamic process of the glacier and its response to environmental changes are inferred;

[0117] The first hidden layer inputs the first graph structure data, outputs the first hidden state to the first output layer, the first output layer outputs the classification result of the glacier's movement pattern, the first hidden state is input to the second output layer, and the second output layer outputs the result of the movement of the glacier's surface features.

[0118] The sequence data, the output results of the first output layer and the output results of the second output layer are input into an input habitat assessment model, the habitat assessment model comprises a second hidden layer, a third hidden layer, a fourth hidden layer, a fifth hidden layer, a first feature fusion layer and a third output layer, and the third output layer outputs the area change rate of the species core area;

[0119] It should be noted that the rate of change of the species core area is the ratio of the difference between the current core area and the past core area to the past core area;

[0120] The second hidden layer inputs sequence data, outputs the second hidden state, and the output result of the first output layer and the output result of the second output layer are respectively input to the fourth hidden layer and the fifth hidden layer, and then the fourth hidden state and the fifth hidden state output by the fourth hidden layer and the fifth hidden layer are input together with the second hidden state into the first feature fusion layer, the first feature fusion layer outputs the first fusion state, and the output first fusion state is input to the third hidden layer, the third hidden layer outputs the third hidden state to the third output layer, and the third output layer outputs the result representing the area change rate of the core area of ​​the fifth middle school;

[0121] The habitat assessment model was run twice. In the first run, the first feature fusion layer input the second hidden state and the default feature. The size of the default feature was consistent with the second hidden state. The first output layer output the initial species core area change rate.

[0122] During the second run, the first feature fusion layer inputs the second hidden state, the fourth hidden state, and the fifth hidden state, and the third output layer outputs the area change rate of the core area of ​​the current species.

[0123] Step 400, based on the output results of the movement pattern classification and the results of the movement of the surface features of the glacier, assess the sensitivity of the glacier to environmental changes, the stability and potential risks of the glacier, or the impact of glacier changes on regional water resources;

[0124] Specifically, the sensitivity of glaciers to environmental changes (such as climate warming) can be assessed by comprehensively analyzing changes in glacier movement and surface characteristics.

[0125] Accelerated glacier flow, increased strain rates, and intensified surface ablation may all indicate an intensified response of glaciers to environmental changes.

[0126] The stability and potential risk of glaciers is to assess the overall stability of glaciers, which is very important for predicting glacier-related disasters (such as ice avalanches, glacier floods, etc.); accelerated glacier flow, increased crack density, unstable material balance, etc. may indicate that the glacier stability is reduced and the potential risk is increased.

[0127] The impact of glacier change on regional water resources is that glacier change will affect the regional water resource balance and hydrological processes; accelerated glacier melting may lead to an increase in river runoff in the short term, but may lead to a decrease in water resources in the long term; assessing the impact of glacier change on water resources is crucial to regional water resource management and planning.

[0128] By comprehensively analyzing the results of glacier movement patterns and surface feature movements, we can have a more comprehensive understanding of the glacier's dynamic characteristics, material balance status, response to environmental changes, and impact on regional water resources. This is of great significance to fields such as glacier research, disaster risk assessment, and water resources management.

[0129] Based on the above results, the habitat assessment model can also be used to output research on species in habitats near glaciers.

[0130] In one embodiment of the present invention, the calculation formula of the first hidden layer is as follows:

[0131] ;

[0132] in represents the first hidden state of the vth node, represents the graph weight parameter in the first hidden layer, and Respectively represent the node features of the vth and uth nodes of the first graph structure data, is the set of nodes that have edges with the node v of the first graph structure data, express The total number of nodes in ;

[0133] ;

[0134] in represents the aggregation coefficient between the vth and uth nodes, tanh represents the hyperbolic tangent function, T represents the transpose, Indicates adjustable parameters; The default value is 1 / N, where N is the dimension of the node features.

[0135] In one embodiment of the present invention, the calculation formula of the first output layer is as follows:

[0136] ;

[0137] in represents the first output vector, and the i-th component of the first output vector represents the probability that the movement mode of the glacier belongs to the i-th state. represents the vth first hidden state of the input first output layer, It means concatenating the first hidden states of all nodes in the first output layer. represents the set of all nodes, is the weight parameter in the first output layer, is the bias parameter in the first output layer, represents the sigmoid function;

[0138] In one embodiment of the present invention, the calculation formula of the second output layer is as follows:

[0139] ;

[0140] in represents the second output vector, and the i-th component of the second output vector represents the probability value of the i-th action; the set of all actions is called the action space, which contains all possible actions that may be selected;

[0141] In one embodiment of the present invention, an action is represented as a vector, and the jth component of the vector represents the jth representation parameter of the movement of the surface feature of the glacier;

[0142] An action is represented as a matrix. The first row and the Lth column of the matrix represent the representation parameters of the Lth displacement vector, the second row and the Lth column represent the representation parameters of the Lth velocity field, the third row and the Lth column represent the representation parameters of the Lth strain rate, the fourth row and the Lth column represent the representation parameters of the Lth surface elevation change, and the fifth row and the Lth column represent the representation parameters of the Lth surface morphology change. represents the first hidden state of the vth node in the second output layer, represents the concatenation of the first hidden states of all nodes in the second output layer, Z represents the set of nodes related to the recognition results, is the weight parameter in the second output layer, is the bias parameter in the second output layer, represents the sigmoid function;

[0143] In one embodiment of the present invention, the calculation formula of the second hidden layer is as follows:

[0144] ;

[0145] in, The second graph structure data identification feature of the dth node of the lth layer of the tth sequence unit of the sequence data, The second graph structure data identification feature of the u-th node of the t-th sequence unit of the sequence data of the l-1-th layer, and They represent the set of nodes with high association with node d and node u respectively, represents the cardinality of a set, The weight matrix representing the second graph structure data of the lth layer, , E represents the total number of layers, when l=1 , The feature representation of the object representing the u-th node index of the t-th sequence unit of the sequence data, is the sigmoid function.

[0146] In one embodiment of the present invention, the calculation formula of the fourth hidden layer is as follows:

[0147] ;

[0148] ;

[0149] in, represents the fourth hidden state, Information representing the movement pattern of the input glacier, represents the first convolution function;

[0150] The calculation formula for the fifth hidden layer is as follows:

[0151] ;

[0152] ;

[0153] in, represents the fifth hidden state, Information representing the movement of the surface features of the input glacier, represents the second convolution function.

[0154] In one embodiment of the present invention, the calculation formula of the first feature fusion layer is as follows:

[0155] ;

[0156] in, represents the first fusion state, The second graph structure data identification feature of the last layer of the dth node output when the tth sequence unit of the sequence data is input into the second hidden layer, represents the set of all units of the second graph structure data, represents the concatenation function, represents the sum weight matrix, Represents the summed bias parameter.

[0157] In one embodiment of the present invention, the calculation formula of the fifth hidden layer is as follows:

[0158] ;

[0159] ;

[0160] ;

[0161] ;

[0162] in, , , represents the first, second, and third weight matrices in the fifth hidden layer, , , represents the first, second, and third bias parameters in the fifth hidden layer, represents the dot product, , and Respectively represent the first, second, and third intermediate states of the fifth hidden layer, , and They represent the t-th and t-1-th fifth hidden states, respectively, n≥t≥1, n represents the total number of sequence units of the sequence data, and when t=1 , tanh is the hyperbolic tangent function, Represents the sigmoid function.

[0163] In one embodiment of the present invention, the calculation formula of the third output layer is as follows:

[0164] ;

[0165] in Represents the output matrix, an element of which represents the change rate of the core area of ​​a species near a glacier. represents the g-th fifth hidden state, Indicates the concatenation of the fifth hidden states of all nodes. represents the set of all nodes, is the weight parameter in the third output layer, is the bias parameter in the third output layer, Represents a sigmoid function.

[0166] In one embodiment of the present invention, the first loss function of the motion estimation model is as follows:

[0167] ;

[0168] in represents the first loss value, M represents the total number of motion mode classification categories, represents the total number of training samples, It is The symbol value of the training sample, if the The actual motion mode of the training sample is the cth motion mode classification category and its value is 1, otherwise it is 0. is the first The probability that a training sample belongs to the cth motion pattern classification category, Represents the logarithmic function with the natural constant e as the base.

[0169] In one embodiment of the present invention, the second loss function of the habitat assessment model is as follows:

[0170] ;

[0171] in represents the second loss value, represents the total number of training samples, represents the real species core area change rate of the species near the glacier of the kth sample, Represents the rate of change in the core area of ​​species near the glacier of the kth sample output by the model.

[0172] like Figure 2 As shown, in one embodiment of the present invention, an image data processing system based on big data and artificial intelligence generation system technology includes:

[0173] Data collection module 101: collects motion estimation information and glacier animal habitat information based on big data and artificial intelligence generation system technology;

[0174] Data encoding module 102: constructing first graph structure data, sequence data and second graph structure data, including sequences or nodes and edges, to establish corresponding motion estimation models and habitat assessment models for each glacier and bottom bed and habitat and animal;

[0175] Data processing module 103: input the encoded first graph structure data, sequence data and second graph structure data into the motion estimation model and habitat assessment model, and output the corresponding motion pattern classification results and the surface feature movement results of the glacier;

[0176] Result evaluation module 104: The dynamic characteristics, material balance state, response to environmental changes and impact on regional water resources of the glacier are evaluated through the corresponding movement pattern classification results and the results of the movement of the glacier's surface features. At the same time, the glacier animal habitat information can also be combined to obtain the area change rate of the species core area and evaluate the impact of glacier changes on species near the glacier.

[0177] At least one embodiment of the present disclosure provides a storage medium storing non-temporary computer-readable instructions for executing one or more steps in the aforementioned image data processing method based on big data and artificial intelligence generation system technology.

[0178] The computer program may be stored / distributed on a suitable medium, such as an optical storage medium or a solid-state medium supplied together with or as part of other hardware, but may also be distributed in other forms, such as via the Internet or other wired or wireless telecommunication systems. Any reference signs in the claims should not be construed as limiting the scope.

[0179] The above describes an embodiment of the present embodiment, but the present embodiment is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present embodiment, ordinary technicians in this field can also make many forms, all of which are within the protection of the present embodiment.

Claims

1. An image data processing method based on big data and artificial intelligence generation system technology, characterized in that: The following steps are involved: Step 100, collecting motion estimation information and glacier animal habitat information; the motion estimation information includes optical and SAR satellite image information, satellite image imaging information, glacier information and bed information; the glacier animal habitat information includes animal species, animal habitat location, animal habitat area and animal number; Step 200, encoding the motion estimation information into first graph structure data, the first graph structure data including nodes and edges connecting the nodes, a node representing a satellite image, a glacier, a bottom bed or a satellite; Constructing second graph structure data, the second graph structure data includes nodes and edges connecting the nodes, one node represents one animal or habitat; encoding the glacier animal habitat information of each node into sequence data, the sequence data includes n sequence units, and the t-th sequence unit represents the glacier animal habitat information of a node at the t-th moment; Step 300, inputting the first graph structure data into a motion estimation model, the first output layer outputting a result representing the classification of the motion pattern of the glacier, and the second output layer outputting a result representing the movement of the surface features of the glacier; The sequence data, the output results of the first output layer, and the output results of the second output layer are input into the input habitat assessment model, and the third output layer outputs the rate of change of the species core area; Step 400, based on the output results of the motion pattern classification and the results of the movement of the surface features of the glacier, evaluate the sensitivity of the glacier to environmental changes, the stability and potential risks of the glacier, or the impact of glacier changes on regional water resources.

2. The image data processing method based on big data and artificial intelligence generation system technology according to claim 1 is characterized in that: Satellite imagery imaging information includes: satellite orbit parameters, imaging geometry information, and ground control point information of the imaging satellite; Glacier information: glacier boundary, glacier surface elevation, glacier surface velocity, glacier thickness, glacier mass balance, glacier surface temperature, glacier surface albedo and glacier surface morphology; Substrate information: bedrock topography, bedrock elevation, bedrock slope, bedrock roughness, bedrock type, bedrock temperature, and bedrock stress.

3. The image data processing method based on big data and artificial intelligence generation system technology according to claim 2 is characterized in that: There are edges between the nodes representing animals and all the nodes representing habitats, there are edges between the nodes corresponding to all animals in the same node representing habitat, and there are edges between all the nodes representing habitats.

4. The image data processing method based on big data and artificial intelligence generation system technology according to claim 3 is characterized in that: There are edges between nodes representing glaciers and nodes representing the bed in the same satellite image; There is an edge between a node representing a glacier in a satellite image and a node representing a glacier in a satellite image at a previous timestamp; If the similarity between a node representing a glacier in a satellite image and a node representing a glacier in a satellite image at a previous timestamp exceeds a set threshold, it is determined that there is an edge between the two; If the similarity between the nodes representing the bed in the satellite image exceeds the set threshold, it is determined that there is an edge between the two; There are edges between satellite imagery images and satellites; There are edges between the nodes representing animals and all the nodes representing habitats, there are edges between the nodes corresponding to all animals in the same node representing habitat, and there are edges between all the nodes representing habitats.

5. The image data processing method based on big data and artificial intelligence generation system technology according to claim 4 is characterized in that: The first hidden layer inputs the first graph structure data, outputs the first hidden state to the first output layer, the first output layer outputs the classification result of the movement pattern of the glacier, the first hidden state is input to the second output layer, and the second output layer outputs the result of the movement of the surface features of the glacier; The sequence data, the output results of the first output layer and the output results of the second output layer are input into an input habitat assessment model, the habitat assessment model comprises a second hidden layer, a third hidden layer, a fourth hidden layer, a fifth hidden layer, a first feature fusion layer and a third output layer, and the third output layer outputs the area change rate of the species core area; It should be noted that the rate of change of the species core area is the ratio of the difference between the current core area and the past core area to the past core area; The second hidden layer inputs sequence data, outputs the second hidden state, and the output result of the first output layer and the output result of the second output layer are respectively input to the fourth hidden layer and the fifth hidden layer, and then the fourth hidden state and the fifth hidden state output by the fourth hidden layer and the fifth hidden layer are input together with the second hidden state into the first feature fusion layer, the first feature fusion layer outputs the first fusion state, and the output first fusion state is input to the third hidden layer, the third hidden layer outputs the third hidden state to the third output layer, and the third output layer outputs the result representing the area change rate of the core area of ​​the fifth middle school; The habitat assessment model was run twice. In the first run, the first feature fusion layer input the second hidden state and the default feature. The size of the default feature was consistent with the second hidden state. The first output layer output the initial species core area change rate. During the second run, the first feature fusion layer inputs the second hidden state, the fourth hidden state, and the fifth hidden state, and the third output layer outputs the area change rate of the core area of ​​the current species.

6. The image data processing method based on big data and artificial intelligence generation system technology according to claim 5 is characterized in that: The calculation formula for the first hidden layer is as follows: ； in represents the first hidden state of the vth node, represents the graph weight parameter in the first hidden layer, and Respectively represent the node features of the vth and uth nodes of the first graph structure data, is the set of nodes that have edges with the node v of the first graph structure data, express The total number of nodes in ; ； in represents the aggregation coefficient between the vth and uth nodes, tanh represents the hyperbolic tangent function, T represents the transpose, Indicates adjustable parameters; The default value of is 1 / N, where N is the dimension of the node feature; The calculation formula of the first output layer is as follows: ； in represents the first output vector, and the i-th component of the first output vector represents the probability that the movement mode of the glacier belongs to the i-th state. represents the vth first hidden state of the input first output layer, It means concatenating the first hidden states of all nodes in the first output layer. represents the set of all nodes, is the weight parameter in the first output layer, is the bias parameter in the first output layer, represents the sigmoid function; The calculation formula of the second output layer is as follows: ； in represents the second output vector, and the i-th component of the second output vector represents the probability value of the i-th action; the set of all actions is called the action space, which contains all possible actions that may be selected; In one embodiment of the present invention, an action is represented as a vector, and the jth component of the vector represents the jth representation parameter of the movement of the surface feature of the glacier; An action is represented as a matrix. The first row and the Lth column of the matrix represent the representation parameters of the Lth displacement vector, the second row and the Lth column represent the representation parameters of the Lth velocity field, the third row and the Lth column represent the representation parameters of the Lth strain rate, the fourth row and the Lth column represent the representation parameters of the Lth surface elevation change, and the fifth row and the Lth column represent the representation parameters of the Lth surface morphology change. represents the first hidden state of the vth node in the second output layer, It represents the concatenation of the first hidden states of all nodes in the second output layer, Z represents the set of nodes related to the recognition results, is the weight parameter in the second output layer, is the bias parameter in the second output layer, Represents a sigmoid function.

7. The image data processing method based on big data and artificial intelligence generation system technology according to claim 6 is characterized in that: The calculation formula for the second hidden layer is as follows: ； in, The second graph structure data identification feature of the dth node of the lth layer of the tth sequence unit of the sequence data, The second graph structure data identification feature of the u-th node of the t-th sequence unit of the sequence data of the l-1-th layer, and They represent the set of nodes with high association with node d and node u respectively, represents the cardinality of a set, The weight matrix representing the second graph structure data of the lth layer, , E represents the total number of layers, when l=1 , The feature representation of the object representing the u-th node index of the t-th sequence unit of the sequence data, is the sigmoid function; The calculation formula for the fourth hidden layer is as follows: ； ； in, represents the fourth hidden state, Information representing the movement pattern of the input glacier, represents the first convolution function; The calculation formula for the fifth hidden layer is as follows: ； ； in, represents the fifth hidden state, Information representing the movement of the surface features of the input glacier, represents the second convolution function.

8. The image data processing method based on big data and artificial intelligence generation system technology according to claim 7 is characterized in that: The calculation formula of the first feature fusion layer is as follows: ; in, represents the first fusion state, The second graph structure data identification feature of the last layer of the dth node output when the tth sequence unit of the sequence data is input into the second hidden layer, represents the set of all units of the second graph structure data, represents the concatenation function, represents the sum weight matrix, represents the summation bias parameter; The calculation formula for the fifth hidden layer is as follows: ； ； ； ； in, , , represents the first, second, and third weight matrices in the fifth hidden layer, , , represents the first, second, and third bias parameters in the fifth hidden layer, represents the dot product, , and Respectively represent the first, second, and third intermediate states of the fifth hidden layer, , and They represent the t-th and t-1-th fifth hidden states, respectively, n≥t≥1, n represents the total number of sequence units of the sequence data, and when t=1 , tanh is the hyperbolic tangent function, Represents the sigmoid function; The calculation formula of the third output layer is as follows: ； in Represents the output matrix, an element of which represents the change rate of the core area of ​​a species near a glacier. represents the g-th fifth hidden state, Indicates the concatenation of the fifth hidden states of all nodes. represents the set of all nodes, is the weight parameter in the third output layer, is the bias parameter in the third output layer, Represents a sigmoid function.

9. An image data processing system based on big data and artificial intelligence generation system technology, characterized in that: Executing the steps in the image data processing method based on big data and artificial intelligence generation system technology as described in any one of claims 1 to 8, including: Data collection module: A generation system based on big data and artificial intelligence technology collects movement estimation information and glacier animal habitat information; Data encoding module: constructing first graph structure data, sequence data and second graph structure data, including sequences or nodes and edges, and establishing corresponding motion estimation models and habitat assessment models for each glacier and bed and habitat and animal; Data processing module: input the encoded first graph structure data, sequence data and second graph structure data into the motion estimation model and habitat assessment model, and output the corresponding motion pattern classification results and the surface feature movement results of the glacier; Result evaluation module: Through the corresponding movement pattern classification results and the results of the movement of the glacier surface features, and also combined with the glacier animal habitat information, the area change rate of the species core area can be obtained.

10. A storage medium, characterized in that: Non-temporary computer-readable instructions are stored for executing the steps in the image data processing method based on big data and artificial intelligence generation system technology as described in any one of claims 1-8.

Images