Visual ecological intelligent monitoring system and method based on three-dimensional simulation

Abstract

The invention relates to the technical field of ecological monitoring, and provides a visual ecological intelligent monitoring system and method based on three-dimensional simulation, and the system comprises a data optimization module which is used for calculating an angle abnormal factor between data in an ecological data flow, carrying out the quality optimization of the ecological data flow, and obtaining a target data flow; the image partitioning module is used for performing image partitioning processing on the multi-temporal remote sensing image to obtain a multi-temporal water area map and a multi-temporal landform map; the landform analysis module is used for carrying out semantic segmentation on the landform structure to obtain landform types and calculating the landform transition amplitude between the landform types; the water area analysis module is used for calculating the element deviation amount corresponding to each element in the water area elements and analyzing the water area evolution trend of the ecological area; and the ecological monitoring module is used for constructing a three-dimensional simulation area corresponding to the ecological area and dynamically monitoring the three-dimensional simulation area to obtain an area ecological monitoring result. The invention aims to improve the accuracy of ecological monitoring.

Description

Technical Field

[0001] The present invention relates to the technical field of ecological monitoring, and particularly to a three-dimensional simulation visualization ecological intelligent monitoring system and method. Background Art

[0002] With the development of society and the intensification of human activities, ecological environment problems have become increasingly prominent. In order to understand the status of the ecosystem and timely discover the problems and threats faced by the ecosystem, it is necessary to monitor the ecosystem in real time and accurately to achieve the purpose of protecting the ecosystem.

[0003] However, the existing ecological monitoring uses two-dimensional planes and relies on manual observation methods. By constructing a two-dimensional plan of the ecological area and collecting ecological data within the area, through experimental analysis of the ecological data, and combining the analysis results with the two-dimensional plan, ecological monitoring of the ecological area is realized. However, in this method, there are certain deviations in the data collected manually, and the final calculated results will deviate. Moreover, the two-dimensional plan is difficult to comprehensively and intuitively reflect the dynamic changes of the ecosystem, which leads to a reduction in the accuracy of ecological monitoring. Summary of the Invention

[0004] The present invention provides a three-dimensional simulation visualization ecological intelligent monitoring system and method, and its main purpose is to improve the accuracy of ecological monitoring.

[0005] To achieve the above object, a three-dimensional simulation visualization ecological intelligent monitoring system provided by the present invention includes a data optimization module, an image segmentation module, a landform analysis module, a water area analysis module, and an ecological monitoring module.

[0006] The data optimization module is used to determine the ecological area to be monitored, collect the ecological data stream corresponding to the ecological area, calculate the angular anomaly factor between each data in the ecological data stream, and optimize the quality of the ecological data stream according to the angular anomaly factor to obtain a target data stream.

[0007] The image segmentation module is used to collect multi-temporal remote sensing images of the ecological area, identify the visual units in the multi-temporal remote sensing images, and perform image segmentation processing on the multi-temporal remote sensing images according to the visual units to obtain multi-temporal water area maps and multi-temporal landform maps.

[0008] The landform analysis module is used to identify the landform structures in the multi-temporal landform maps, extract the landform features corresponding to the landform structures, perform semantic segmentation on the landform structures according to the landform features to obtain landform types, calculate the landform change amplitude between the landform types according to the multi-temporal landform maps, and analyze the landform evolution trend of the ecological area according to the landform change amplitude.

[0009] The water area analysis module is used to construct the water area elements of the multi-temporal water area map, combine the water area elements and the multi-temporal water area map, calculate the element deviation amount corresponding to each element in the water area elements, and analyze the water area evolution trend of the ecological area according to the element deviation amount;

[0010] The ecological monitoring module is used to combine the multi-temporal geomorphic map and the multi-temporal water area map to construct a three-dimensional simulation area corresponding to the ecological area, and dynamically monitor the three-dimensional simulation area in combination with the geomorphic evolution trend, the water area evolution trend and the target data stream to obtain the regional ecological monitoring result.

[0011] Optionally, calculating the angular anomaly factor between each pair of data in the ecological data stream includes:

[0012] Performing clustering processing on each data in the ecological data stream to obtain a clustering data set;

[0013] Locating the central data corresponding to the clustering data set and calculating the data distance value between each data in the clustering data set and the central data;

[0014] Calculating the total distance value between each data in the ecological clustering data and the adjacent data sets;

[0015] Performing vectorization processing on the data and the central data in the clustering data set to obtain a data vector and a central vector;

[0016] Combining the data vector, the central vector, the total distance value and the distance value to calculate the angular anomaly factor between each pair of data in the ecological data stream.

[0017] Optionally, combining the data vector, the central vector, the total distance value and the data distance value to calculate the angular anomaly factor between each pair of data in the ecological data stream includes:

[0018] Calculating the vector angle between each vector in the data vector and the central vector;

[0019] Determining the data angle between each pair of data in the ecological data stream according to the vector angle;

[0020] Combining the total distance value and the data distance value to calculate the distance change amount between each data in the clustering data set and the adjacent data sets;

[0021] Combining the distance value, the total distance value, the distance change amount and the data angle to calculate the angular anomaly factor between each pair of data in the ecological data stream.

[0022] Optionally, the multi-temporal remote sensing image is subjected to image block processing according to the visual unit to obtain a multi-temporal water area map and a multi-temporal landform map, including:

[0023] Perform radiometric correction processing on the multi-temporal remote sensing image to obtain a corrected multi-temporal map;

[0024] Perform geometric correction on the corrected multi-temporal map to obtain a target multi-temporal map;

[0025] Extract the visual features corresponding to the visual unit from the target multi-temporal map;

[0026] Construct a visual label corresponding to the visual unit according to the visual features, and identify the water area label and the landform label in the visual label;

[0027] Combine the water area label and the landform label to perform image block processing on the target multi-temporal map to obtain a multi-temporal water area map and a multi-temporal landform map.

[0028] Optionally, the identification of the water area label and the landform label in the visual label includes:

[0029] Calculate the label gain value corresponding to the visual label, and extract the characterization label in the visual label according to the label gain value;

[0030] Identify the label characters in the characterization label, and extract the key characters in the label characters;

[0031] Perform semantic analysis on the key characters to obtain character semantics;

[0032] Identify the water area label and the landform label in the visual label according to the character semantics.

[0033] Optionally, the extraction of the landform features corresponding to the landform structure includes:

[0034] Perform gray-scale processing on the landform structure to obtain a gray-scale landform structure;

[0035] Perform smoothing processing on the gray-scale landform structure to obtain a smoothed landform structure;

[0036] Extract the structural texture features corresponding to the smoothed landform structure;

[0037] Perform dimensionality reduction processing on the structural texture features to obtain dimensionality-reduced texture features;

[0038] Calculate the feature energy value corresponding to the dimensionality-reduced texture features;

[0039] Perform feature selection on the dimensionality-reduced texture features according to the feature energy value to obtain target texture features;

[0040] Use the target texture feature as the geomorphic feature corresponding to the geomorphic structure.

[0041] Optionally, calculating the feature energy value corresponding to the dimensionality-reduced texture feature includes:

[0042] Assign the feature weight corresponding to the dimensionality-reduced texture feature and calculate the texture feature value corresponding to the dimensionality-reduced texture feature;

[0043] Calculate the feature average value corresponding to the dimensionality-reduced texture feature according to the texture feature value;

[0044] Combine the feature weight, the feature average value and the texture feature value to calculate the feature energy value corresponding to the dimensionality-reduced texture feature.

[0045] Optionally, calculating the geomorphic change amplitude between the geomorphic types according to the multi-temporal geomorphic map includes:

[0046] Measure the pixel brightness value corresponding to each image in the multi-temporal geomorphic map;

[0047] Calculate the brightness average value corresponding to each image in the multi-temporal geomorphic map according to the pixel brightness value;

[0048] Calculate the pixel dispersion corresponding to each image in the multi-temporal geomorphic map;

[0049] Combine the brightness average value and the pixel dispersion to calculate the geomorphic change amplitude between the geomorphic types.

[0050] Optionally, combining the water area elements and the multi-temporal water area map to calculate the element deviation amount corresponding to each element in the water area elements includes:

[0051] Perform segmentation processing on the multi-temporal water area map according to the water area elements to obtain a water area element map;

[0052] Extract the image parameters corresponding to the water area element map;

[0053] Perform standardization processing on the image parameters to obtain standard image parameters;

[0054] Calculate the element deviation amount corresponding to each element in the water area elements according to the standard image parameters.

[0055] A three-dimensional simulation visualization ecological intelligent monitoring method, characterized in that the method includes:

[0056] Determine the ecological region to be monitored, collect the ecological data stream corresponding to the ecological region, calculate the angular anomaly factor between each data in the ecological data stream, and optimize the quality of the ecological data stream according to the angular anomaly factor to obtain the target data stream;

[0057] Collect multi-temporal remote sensing images of the ecological region, identify the visual units in the multi-temporal remote sensing images, and perform image block processing on the multi-temporal remote sensing images according to the visual units to obtain multi-temporal water area maps and multi-temporal geomorphic maps;

[0058] Identify the geomorphic structures in the multi-temporal geomorphic maps, extract the geomorphic features corresponding to the geomorphic structures, perform semantic segmentation on the geomorphic structures according to the geomorphic features to obtain geomorphic types, calculate the geomorphic change amplitude between the geomorphic types according to the multi-temporal geomorphic maps, and analyze the geomorphic evolution trend of the ecological region according to the geomorphic change amplitude;

[0059] Construct the water area elements of the multi-temporal water area maps, combine the water area elements and the multi-temporal water area maps, calculate the element deviation amount corresponding to each element in the water area elements, and analyze the water area evolution trend of the ecological region according to the element deviation amount;

[0060] Combine the multi-temporal geomorphic maps and the multi-temporal water area maps to construct a three-dimensional simulation area corresponding to the ecological region, and perform dynamic monitoring on the three-dimensional simulation area by combining the geomorphic evolution trend, the water area evolution trend and the target data stream to obtain the regional ecological monitoring results.

[0061] By calculating the angular anomaly factor between each pair of data in the ecological data stream, the present invention can obtain the degree of angular anomaly between each pair of data in the ecological data stream, so as to discover the abnormal data in the ecological data stream, providing a basis for subsequent quality optimization of the ecological data stream. By identifying the visual units in the multi-temporal remote sensing images, the present invention can obtain the image content in the multi-temporal remote sensing images, facilitating the analysis of the information in the images and providing a basis for subsequent image segmentation processing of the multi-temporal remote sensing images. By extracting the geomorphic features corresponding to the geomorphic structure, the present invention can obtain the morphological and structural features corresponding to the geomorphic structure, and according to the geomorphic features, perform semantic segmentation on the geomorphic structure to obtain the geomorphic categories corresponding to the geomorphic structure, thereby providing a basis for calculating the amplitude of geomorphic change between subsequent geomorphic types. By combining the water area elements and the multi-temporal water area maps, and calculating the element deviation amount corresponding to each element in the water area elements, the present invention can obtain the deviation degree corresponding to the water area elements, thereby providing a basis for subsequent analysis of the water area evolution trend. By combining the geomorphic evolution trend, the water area evolution trend and the target data stream, the present invention can perform dynamic monitoring on the three-dimensional simulation area, improving the accuracy of ecological monitoring in the ecological area. Therefore, a three-dimensional simulation visualization-based ecological intelligent monitoring system and method provided by an embodiment of the present invention can improve the accuracy of ecological monitoring. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 It is a functional module diagram of a three-dimensional simulation visualization-based ecological intelligent monitoring system provided by an embodiment of the present invention;

[0063] Figure 2 It is a flowchart of a three-dimensional simulation visualization-based ecological intelligent monitoring method provided by an embodiment of the present invention.

[0064] The realization of the object, functional features and advantages of the present invention will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0065] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments of the present invention fall within the scope of protection of the present invention.

[0066] In addition, the step sequence in the following method embodiments is only an example, not strictly limited.

[0067] In fact, the server devices deployed based on the three-dimensional simulation visualization ecological intelligent monitoring system may consist of one or more devices. The above-mentioned three-dimensional simulation visualization ecological intelligent monitoring system can be implemented as: business instances, virtual machines, and hardware devices. For example, the three-dimensional simulation visualization ecological intelligent monitoring system can be implemented as a business instance deployed on one or more devices in a cloud node. Simply put, the system can be understood as a software deployed on a cloud node for providing services for three-dimensional simulation visualization ecological intelligent monitoring to each client. Alternatively, the three-dimensional simulation visualization ecological intelligent monitoring system can also be implemented as a virtual machine deployed on one or more devices in a cloud node. An application software for managing each client is installed in the virtual machine. Or, the three-dimensional simulation visualization ecological intelligent monitoring system can also be implemented as a server composed of many identical or different types of hardware devices, and one or more hardware devices are set to provide services for three-dimensional simulation visualization ecological intelligent monitoring to each client.

[0068] In terms of implementation form, the three-dimensional simulation visualization ecological intelligent monitoring system and the client adapt to each other. That is, if the three-dimensional simulation visualization ecological intelligent monitoring system is an application installed on a cloud service platform, then the client is a client that establishes a communication connection with the application; or if the three-dimensional simulation visualization ecological intelligent monitoring system is implemented as a website, then the client is implemented as a web page; or if the three-dimensional simulation visualization ecological intelligent monitoring system is implemented as a cloud service platform, then the client is implemented as a small program in an instant messaging application.

[0069] Refer to Figure 1 As shown, it is a functional module diagram of the three-dimensional simulation visualization ecological intelligent monitoring system provided by an embodiment of the present invention.

[0070] The three-dimensional simulation visualization ecological intelligent monitoring system 100 described in the present invention can be set in a cloud server. In terms of implementation form, it can be one or more service devices, or can be installed as an application on the cloud (such as a server or a server cluster based on the three-dimensional simulation visualization ecological intelligent monitoring system), or can also be developed into a website. According to the functions achieved, the three-dimensional simulation visualization ecological intelligent monitoring system 100 includes a data optimization module 101, an image segmentation module 102, a landform analysis module 103, a water area analysis module 104, and an ecological monitoring module 105.

[0071] In the embodiments of the present invention, in the tracking based on three-dimensional simulation visualization ecological intelligent monitoring, each of the above modules can be independently implemented and called by other modules. Here, the call can be understood as that a certain module can be connected to multiple modules of another type and provide corresponding services for the multiple modules it is connected to. In the three-dimensional simulation visualization ecological intelligent monitoring system provided by the embodiments of the present invention, without modifying the program code, the applicable range of the three-dimensional simulation visualization ecological intelligent monitoring architecture can be adjusted by adding modules and directly calling them, so as to achieve cluster-level horizontal expansion, so as to achieve the purpose of quickly and flexibly expanding the three-dimensional simulation visualization ecological intelligent monitoring system. In practical applications, the above modules can be set in the same device or different devices, or can be set in virtual devices, such as service instances in a cloud server.

[0072] The following will respectively describe each component and the specific working process of the three-dimensional simulation visualization ecological intelligent monitoring system in combination with specific embodiments.

[0073] The data optimization module 101 is used to determine the ecological area to be monitored, collect the ecological data stream corresponding to the ecological area, calculate the angular anomaly factor between each data in the ecological data stream, and optimize the quality of the ecological data stream according to the angular anomaly factor to obtain a target data stream.

[0074] By calculating the angular anomaly factor between each data in the ecological data stream, the present invention can obtain the degree of angular anomaly between each data in the ecological data stream to discover the abnormal data in the ecological data stream, providing a basis for subsequent quality optimization of the ecological data stream. Among them, the ecological area is a geographical area that needs to be monitored, such as a river basin, a mountain range or a forest. The angular anomaly factor represents the degree of angular anomaly between each data in the ecological data stream. The target data stream is the data obtained after removing the abnormal data in the ecological data stream. Optionally, the collection of the ecological data stream corresponding to the ecological area can be through a sensor network, deploying various sensors, such as temperature, humidity, light, water quality and other sensors, and collecting the ecological data stream of the ecological area through each sensor; comparing the angular anomaly factor with a preset anomaly threshold, and when the angular anomaly factor is greater than the anomaly threshold, removing the corresponding data to obtain a target data stream. The anomaly threshold can be set to 0.8 or can be set according to the actual application scenario.

[0075] As an embodiment of the present invention, calculating the angular anomaly factor between each pair of data in the ecological data stream includes: performing clustering processing on each piece of data in the ecological data stream to obtain a clustering data set, locating the central data corresponding to the clustering data set, calculating the data distance value between each piece of data in the clustering data set and the central data, and calculating the total distance value between each piece of data in the ecological clustering data and the adjacent data sets. Vectorize the data and the central data in the clustering data set to obtain a data vector and a central vector. Combine the data vector, the central vector, the total distance value, and the distance value to calculate the angular anomaly factor between each pair of data in the ecological data stream.

[0076] Among them, the clustering data set is formed by gathering data of the same nature in the ecological data stream. The central data is the core data in the clustering data set. The data distance value represents the degree of distance between each piece of data in the clustering data set and the central data. The total distance value represents the sum of the data distances between each piece of data in the ecological clustering data and the data in the adjacent data sets.

[0077] Optionally, the clustering processing of each piece of data in the ecological data stream can be implemented by a clustering algorithm, such as the K-Means algorithm. The location of the central data corresponding to the clustering data set can be implemented by the median method, and the data selected as the median is used as the central data. The data distance value between each piece of data in the clustering data set and the central data can be calculated by the Euclidean distance algorithm. The total distance value between each piece of data in the ecological clustering data and the adjacent data sets can be obtained by calculating the distance between each piece of data and each piece of data in the adjacent data sets and summing up the distances. The vectorization processing of the data and the central data in the clustering data set can be implemented by one-hot encoding.

[0078] Further, as an optional embodiment of the present invention, combining the data vector, the central vector, the total distance value, and the data distance value to calculate the angular anomaly factor between each pair of data in the ecological data stream includes: calculating the vector angle between each vector in the data vector and the central vector, determining the data angle between each pair of data in the ecological data stream according to the vector angle, combining the total distance value and the data distance value, calculating the distance change amount between each piece of data in the clustering data set and the adjacent data sets, and combining the distance value, the total distance value, the distance change amount, and the data angle to calculate the angular anomaly factor between each pair of data in the ecological data stream through the following formula:

[0079]

[0080] Among them, A represents the angular anomaly factor between each pair of data in the ecological data stream, D a represents the data distance value corresponding to the a-th data in the ecological data stream, B a represents the total distance value corresponding to the a-th data in the ecological data stream, E a represents the data included angle of the a-th data in the ecological data stream, F a represents the distance change amount of the a-th data in the ecological data stream.

[0081] Among them, the vector included angle represents the angle formed between each vector in the data vector and the central vector, and the distance change amount represents the change amount of the distance between each data in the clustering dataset and each data in the adjacent dataset. Further, the calculation steps for the distance change amount of each data in the clustering dataset and the adjacent dataset are as follows: count the number of data in the adjacent dataset, divide the total distance value by the number of data to obtain the average distance value, and calculate the difference between the data distance value and the average distance value to obtain the distance change amount of each data in the clustering dataset and the adjacent dataset.

[0082] Further, as an optional embodiment of the present invention, calculating the vector included angle between each vector in the data vector and the central vector includes:

[0083] Calculating the vector included angle between each vector in the data vector and the central vector through the following formula:

[0084]

[0085] Among them, G represents the vector included angle between each vector in the data vector and the central vector, represents the b-th vector in the data vector, b represents the serial number of the data vector, represents the central vector, and respectively represent the modulus of the b-th vector and the central vector in the data vector.

[0086] The image block module 102 is configured to collect multi-temporal remote sensing images of the ecological area, identify visual units in the multi-temporal remote sensing images, and perform image block processing on the multi-temporal remote sensing images according to the visual units to obtain multi-temporal water area maps and multi-temporal landform maps.

[0087] By identifying the visual units in the multi-temporal remote sensing images, the present invention can obtain the image content in the multi-temporal remote sensing images, so as to facilitate the analysis of the information in the images, providing a basis for subsequent image block processing of the multi-temporal remote sensing images. Among them, the multi-temporal remote sensing images are images of the ecological region taken at different times, and the visual units are the image elements in the multi-temporal remote sensing images. Further, the acquisition of the multi-temporal remote sensing images of the ecological region can be achieved by an unmanned aerial vehicle; the identification of the visual units in the multi-temporal remote sensing images can be achieved by a human-computer interaction method. Through the human-computer interaction method, the user is involved in the identification process of the visual units, thereby improving the accuracy and efficiency of the identification.

[0088] As an embodiment of the present invention, the image block processing of the multi-temporal remote sensing images according to the visual units to obtain a multi-temporal water area map and a multi-temporal landform map includes: performing radiometric correction processing on the multi-temporal remote sensing images to obtain a corrected multi-temporal map, performing geometric correction on the corrected multi-temporal map to obtain a target multi-temporal map, extracting the visual features corresponding to the visual units in the target multi-temporal map, constructing visual labels corresponding to the visual units according to the visual features, identifying the water area labels and landform labels in the visual labels, and combining the water area labels and the landform labels to perform image block processing on the target multi-temporal map to obtain a multi-temporal water area map and a multi-temporal landform map.

[0089] Among them, the corrected multi-temporal map is an image obtained after the brightness and contrast of the images in the multi-temporal remote sensing images are corrected, the target multi-temporal map is an image obtained after the geometric deformation of the corrected multi-temporal map is corrected, the visual features are the representations corresponding to the visual units in the target multi-temporal map, such as color features or morphological features, the visual labels are the visual description identifiers corresponding to the visual units, and the water area labels and landform labels are the labels related to water areas and landforms in the visual labels.

[0090] Further, the radiometric correction processing of the multi-temporal remote sensing images can be achieved through the Gamma correction method; the geometric correction of the multi-temporal corrected images can be achieved through the polynomial correction method, using a polynomial function to fit the geometric deformation of the images, and then performing correction to obtain the target multi-temporal images; extracting the visual features corresponding to the visual units in the target multi-temporal images can be achieved through a feature extraction algorithm, and the feature extraction algorithm is compiled by a scripting language, such as the JS scripting language; extracting the feature descriptors of the visual features, generating the visual labels corresponding to the visual units according to the feature descriptors, where the feature descriptors are the descriptive information corresponding to the visual features and can be extracted through corresponding algorithms, such as the Sobel operator; the image block processing of the target multi-temporal images can be achieved through a segmentation algorithm, such as the threshold segmentation method.

[0091] Optionally, as an alternative embodiment of the present invention, identifying the water area label and the landform label in the visual label includes: calculating the label gain value corresponding to the visual label, extracting the representative label in the visual label according to the label gain value, identifying the label characters in the representative label, extracting the key characters in the label characters, performing semantic analysis on the key characters to obtain the character semantics, and identifying the water area label and the landform label in the visual label according to the character semantics.

[0092] Among them, the label gain value represents the importance corresponding to the visual label, the representative label is the representative label in the visual label, the label characters are the text information in the representative label, and the key characters are the core characters in the label characters. Further, the calculation of the label gain value corresponding to the visual label can be achieved through the information gain algorithm, and the label gain value is determined by calculating the increased amount of information brought by the visual label; the representative label in the visual label can be extracted according to the numerical size corresponding to the label gain value, and the label with an increasing label gain value is selected from the visual labels of the same category; the label characters in the representative label can be achieved through ocr recognition technology; the extraction of the key characters in the label characters can be achieved through the N-gram model; the semantic analysis of the key characters can be achieved through the semantic analysis method; by identifying the water-related and landform-related vocabulary in the character semantics, such as ocean, lake or mountain, etc., the water area label and the landform label in the visual label are further determined.

[0093] The landform analysis module 103 is used to identify the landform structure in the multi-temporal landform map, extract the landform features corresponding to the landform structure, perform semantic segmentation on the landform structure according to the landform features to obtain the landform types, calculate the landform change amplitude between the landform types according to the multi-temporal landform map, and analyze the landform evolution trend of the ecological region according to the landform change amplitude.

[0094] By extracting the geomorphic features corresponding to the geomorphic structure, the present invention can obtain the morphological and structural features corresponding to the geomorphic structure, and according to the geomorphic features, semantic segmentation of the geomorphic structure can be performed to obtain the geomorphic categories corresponding to the geomorphic structure, thereby providing a basis for calculating the amplitude of geomorphic change between subsequent geomorphic types. Among them, the geomorphic structure is the geomorphic form in the multi-temporal geomorphic map. Optionally, the recognition of the geomorphic structure in the multi-temporal geomorphic map can be realized by an edge detection algorithm.

[0095] As an embodiment of the present invention, the extraction of the geomorphic features corresponding to the geomorphic structure includes: performing gray-scale processing on the geomorphic structure to obtain a gray-scale geomorphic structure, performing smoothing processing on the gray-scale geomorphic structure to obtain a smoothed geomorphic structure, extracting the structural texture features corresponding to the smoothed geomorphic structure, performing dimensionality reduction processing on the structural texture features to obtain dimensionality-reduced texture features, calculating the feature energy value corresponding to the dimensionality-reduced texture features, and according to the feature energy value, performing feature selection on the dimensionality-reduced texture features to obtain target texture features, and taking the target texture features as the geomorphic features corresponding to the geomorphic structure.

[0096] Among them, the smoothed geomorphic structure is a structure diagram obtained after removing noise interference in the gray-scale geomorphic structure, the structural texture features are the texture representations corresponding to the smoothed geomorphic structure, the dimensionality-reduced texture features are the features obtained after removing redundant features in the structural texture features, and the feature energy value represents the feature intensity corresponding to the dimensionality-reduced texture features.

[0097] Further, the gray-scale processing of the geomorphic structure can be realized by the average value method, taking the average value of the RGB values of each pixel of the geomorphic structure to obtain a gray-scale value, and performing gray-scale processing on the geomorphic structure through the gray-scale value to obtain a gray-scale geomorphic structure; the smoothing processing of the gray-scale geomorphic structure can be realized by a low-pass filter; the extraction of the structural texture features corresponding to the smoothed geomorphic structure can be realized by a gray-level co-occurrence matrix; the dimensionality reduction processing of the structural texture features can be realized by the principal component analysis method; comparing the feature energy value with a preset energy value, if the feature energy value is greater than the preset energy value, then retaining the features of the dimensionality-reduced texture features to obtain target texture features.

[0098] Further, as an alternative embodiment of the present invention, calculating the feature energy value corresponding to the dimensionality-reduced texture feature includes: allocating the feature weight corresponding to the dimensionality-reduced texture feature, calculating the texture feature value corresponding to the dimensionality-reduced texture feature, calculating the feature average value corresponding to the dimensionality-reduced texture feature according to the texture feature value, and combining the feature weight, the feature average value, and the texture feature value to calculate the feature energy value corresponding to the dimensionality-reduced texture feature through the following formula:

[0099]

[0100] where L represents the feature energy value corresponding to the dimensionality-reduced texture feature, ω i represents the feature weight corresponding to the i-th feature in the dimensionality-reduced texture feature, K i represents the texture feature value corresponding to the i-th feature in the dimensionality-reduced texture feature, represents the feature average value, i represents the feature serial number corresponding to the dimensionality-reduced texture feature, and q represents the number of features of the dimensionality-reduced texture feature.

[0101] Among them, the feature weight represents the importance degree corresponding to the dimensionality-reduced texture feature, and the texture feature value is the expression value corresponding to the dimensionality-reduced texture feature. Further, the feature entropy value between the dimensionality-reduced texture features can be calculated, and the feature entropy value can be used as the feature weight corresponding to the dimensionality-reduced texture feature; the calculation of the texture feature value corresponding to the dimensionality-reduced texture feature can be realized by a linear function.

[0102] The present invention calculates the landform change amplitude between the landform types according to the multi-temporal landform map, and can understand the difference degree of the landform types on different time scales, so as to facilitate the subsequent analysis of the landform evolution trend. Among them, the landform change amplitude is the change degree between the landform types. Further, the difference value between the landform change amplitudes is calculated, and the landform evolution trend of the ecological region is analyzed according to the size of the difference value. For example, if the height difference is -3, it means that the height evolution trend of the landform is decreasing.

[0103] As an embodiment of the present invention, calculating the landform change amplitude between the landform types according to the multi-temporal landform map includes: measuring the pixel brightness value corresponding to each image in the multi-temporal landform map, calculating the brightness average value corresponding to each image in the multi-temporal landform map according to the pixel brightness value, and calculating the pixel dispersion corresponding to each image in the multi-temporal landform map. Combining the brightness average value and the pixel dispersion, the landform change amplitude between the landform types is calculated through the following formula:

[0104]

[0105] Among them, J represents the magnitude of geomorphic change between geomorphic types, φ d and φ d+1 respectively represent the average brightness values corresponding to the d-th and (d + 1)-th maps in the multi-temporal geomorphic map, β d and β d+1 respectively represent the pixel dispersion corresponding to the d-th and (d + 1)-th maps in the multi-temporal geomorphic map, α1 represents the brightness scaling factor, and α2 represents the dispersion scaling factor.

[0106] Furthermore, the measurement of the pixel brightness value corresponding to each image in the multi-temporal geomorphic map can be achieved through image processing software, such as Adobe Photoshop software; the pixel dispersion corresponding to each image in the multi-temporal geomorphic map can be calculated through the variance calculation formula; the brightness scaling factor and the dispersion scaling factor can be calculated through the data-driven method, by analyzing and statistically processing the historical data set to calculate the most suitable scaling factor.

[0107] The water area analysis module 104 is used to construct the water area elements of the multi-temporal water area map, combine the water area elements and the multi-temporal water area map, calculate the element deviation corresponding to each element in the water area elements, and analyze the water area evolution trend of the ecological region according to the element deviation.

[0108] In the present invention, by combining the water area elements and the multi-temporal water area map and calculating the element deviation corresponding to each element in the water area elements, the deviation degree corresponding to the water area elements can be obtained, thereby providing a basis for the subsequent analysis of the water area evolution trend. Among them, the water area elements are the water area indicators corresponding to the multi-temporal water area map, such as water area and shoreline, etc. Furthermore, the water area elements of the multi-temporal water area map can be constructed by combining the corresponding monitoring targets, determining the monitoring indicators through the detection targets, and selecting the elements corresponding to the monitoring indicators to construct the water area elements; the water area evolution trend of the ecological region is analyzed through the positive and negative values between the element deviations.

[0109] As an embodiment of the present invention, the combination of the water area elements and the multi-temporal water area map to calculate the element deviation corresponding to each element in the water area elements includes: performing segmentation processing on the multi-temporal water area map according to the water area elements to obtain a water area element map, extracting the image parameters corresponding to the water area element map, performing standardization processing on the image parameters to obtain standard image parameters, and calculating the element deviation corresponding to each element in the water area elements according to the standard image parameters.

[0110] Among them, the water area element map is the map corresponding to the water area element in the multi-temporal water area map, the image parameter is the parameter of the water area element in the water area element map, such as the number of pixels, and the standard image parameter is the parameter obtained after correcting the abnormal parameter in the image parameter. Further, the segmentation process of the multi-temporal water area map can be realized by a segmentation tool, and the segmentation tool is compiled by JAVA language; the extraction of the image parameter corresponding to the water area element map can be realized by the OpenCV tool; the standardization process of the image parameter can be realized by a corresponding standard algorithm, and the standard algorithm can be compiled by a programming language; calculate the parameter difference between the standard image parameters, and obtain the element deviation amount corresponding to each element in the water area element according to the parameter difference.

[0111] The ecological monitoring module 105 is used to combine the multi-temporal landform map and the multi-temporal water area map to construct a three-dimensional simulation area corresponding to the ecological area, and combine the landform evolution trend, the water area evolution trend and the target data stream to dynamically monitor the three-dimensional simulation area to obtain the regional ecological monitoring result.

[0112] The present invention can improve the ecological monitoring accuracy of the ecological area by dynamically monitoring the three-dimensional simulation area by combining the landform evolution trend, the water area evolution trend and the target data stream. Among them, the three-dimensional simulation area is a simulation model corresponding to the ecological area. Further, the construction of the three-dimensional simulation area corresponding to the ecological area can be realized by three-dimensional software, such as CAD software; combine the landform evolution trend and the water area evolution trend to determine the change trend of the three-dimensional simulation area, and combine the target data stream and the change trend to predict the three-dimensional simulation area, so as to timely discover the ecological problems in the ecological area and make corresponding solutions.

[0113] By calculating the angular anomaly factor between each pair of data in the ecological data stream, the present invention can obtain the degree of angular anomaly between each pair of data in the ecological data stream, so as to discover the abnormal data in the ecological data stream, providing a basis for subsequent quality optimization of the ecological data stream. By identifying the visual units in the multi-temporal remote sensing images, the present invention can obtain the image content in the multi-temporal remote sensing images, facilitating the analysis of the information in the images and providing a basis for subsequent image segmentation processing of the multi-temporal remote sensing images. By extracting the geomorphic features corresponding to the geomorphic structure, the present invention can obtain the morphological and structural features corresponding to the geomorphic structure, and according to the geomorphic features, perform semantic segmentation on the geomorphic structure to obtain the geomorphic categories corresponding to the geomorphic structure, thereby providing a basis for calculating the geomorphic change amplitude between the subsequent geomorphic types. By combining the water area elements and the multi-temporal water area maps, the present invention calculates the element deviation amount corresponding to each element in the water area elements, and can obtain the deviation degree corresponding to the water area elements, thereby providing a basis for subsequent analysis of the water area evolution trend. By combining the geomorphic evolution trend, the water area evolution trend and the target data stream, the present invention performs dynamic monitoring on the three-dimensional simulation area, which can improve the accuracy of ecological monitoring in the ecological area. Therefore, a three-dimensional simulation visualization-based ecological intelligent monitoring system and method provided by an embodiment of the present invention can improve the accuracy of ecological monitoring.

[0114] Referring to Figure 2 As shown, it is a schematic flowchart of a three-dimensional simulation visualization-based ecological intelligent monitoring method provided by an embodiment of the present invention. In this embodiment, the three-dimensional simulation visualization-based ecological intelligent monitoring method includes:

[0115] S1. Determine the ecological area to be monitored, collect the ecological data stream corresponding to the ecological area, calculate the angular anomaly factor between each pair of data in the ecological data stream, and optimize the quality of the ecological data stream according to the angular anomaly factor to obtain the target data stream;

[0116] S2. Collect multi-temporal remote sensing images of the ecological area, identify the visual units in the multi-temporal remote sensing images, and perform image segmentation processing on the multi-temporal remote sensing images according to the visual units to obtain multi-temporal water area maps and multi-temporal geomorphic maps;

[0117] S3. Identify the geomorphic structure in the multi-temporal geomorphic map, extract the geomorphic features corresponding to the geomorphic structure, perform semantic segmentation on the geomorphic structure according to the geomorphic features to obtain geomorphic types, calculate the geomorphic change amplitude between the geomorphic types according to the multi-temporal geomorphic map, and analyze the geomorphic evolution trend of the ecological area according to the geomorphic change amplitude;

[0118] S4. Construct the water area elements of the multi-temporal water area map. Combine the water area elements and the multi-temporal water area map, calculate the element deviation amount corresponding to each element in the water area elements, and analyze the water area evolution trend of the ecological area according to the element deviation amount;

[0119] S5. Combine the multi-temporal geomorphic map and the multi-temporal water area map to construct the three-dimensional simulation area corresponding to the ecological area. Combine the geomorphic evolution trend, the water area evolution trend and the target data stream to dynamically monitor the three-dimensional simulation area and obtain the regional ecological monitoring result.

[0120] In several embodiments provided by the present invention, it should be understood that the provided systems and methods can be implemented in other ways. For example, the system embodiments described above are only illustrative. For example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation.

[0121] In addition, each functional module in each embodiment of the present invention can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above integrated unit can be implemented in the form of hardware, or in the form of a hardware plus software functional module.

[0122] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. An ecological intelligent monitoring system based on three-dimensional simulation visualization, characterized in that: The data monitoring and processing system includes: a data optimization module, an image segmentation module, a landform analysis module, a water area analysis module, and an ecological monitoring module. The data optimization module is used to determine the ecological area to be monitored, collect the ecological data stream corresponding to the ecological area, calculate the angle anomaly factor between each data in the ecological data stream, and optimize the quality of the ecological data stream according to the angle anomaly factor to obtain the target data stream; The image segmentation module is used to collect multi-temporal remote sensing images of the ecological area, identify visual units in the multi-temporal remote sensing images, and perform image segmentation processing on the multi-temporal remote sensing images according to the visual units to obtain a multi-temporal water area map and a multi-temporal landform map; The geomorphic analysis module is used to identify the geomorphic structure in the multi-temporal geomorphic map, extract the geomorphic features corresponding to the geomorphic structure, perform semantic segmentation on the geomorphic structure according to the geomorphic features, obtain the geomorphic type, calculate the geomorphic change amplitude between the geomorphic types according to the multi-temporal geomorphic map, and analyze the geomorphic evolution trend of the ecological region according to the geomorphic change amplitude; The water area analysis module is used to construct the water area elements of the multi-temporal water area map, combine the water area elements and the multi-temporal water area map, calculate the element deviation corresponding to each element in the water area elements, and analyze the water area evolution trend of the ecological region according to the element deviation; The ecological monitoring module is used to combine the multi-temporal landform map and the multi-temporal water area map to construct a three-dimensional simulation area corresponding to the ecological area, and dynamically monitor the three-dimensional simulation area in combination with the landform evolution trend, the water area evolution trend and the target data stream to obtain regional ecological monitoring results.

2. The ecological intelligent monitoring system based on three-dimensional simulation visualization as claimed in claim 1 is characterized in that: The calculating of the angle anomaly factor between each data in the ecological data stream comprises: Performing clustering processing on each data in the ecological data stream to obtain a clustered data set; Locating the central data corresponding to the clustered data set, and calculating the data distance value between each data in the clustered data set and the central data; Calculate the total distance value between each data in the ecological clustering data and the adjacent data sets; Vectorizing the data in the clustering data set and the center data to obtain a data vector and a center vector; The data vector, the center vector, the total distance value and the distance value are combined to calculate the angle anomaly factor between each data in the ecological data stream.

3. The ecological intelligent monitoring system based on three-dimensional simulation visualization as claimed in claim 2 is characterized in that: The combining the data vector, the center vector, the total distance value and the data distance value to calculate the angle anomaly factor between each data in the ecological data stream includes: Calculating the vector angle between each vector in the data vector and the central vector; Determine the data angle between each data in the ecological data stream according to the vector angle; Combining the total distance value and the data distance value, calculating the distance change between each data in the clustering data set and the adjacent data set; The angle anomaly factor between each data in the ecological data stream is calculated by combining the distance value, the total distance value, the distance change and the data angle.

4. The ecological intelligent monitoring system based on three-dimensional simulation visualization as claimed in claim 1 is characterized in that: The multi-temporal remote sensing image is subjected to image block processing according to the visual unit to obtain a multi-temporal water area map and a multi-temporal landform map, including: Performing radiation correction processing on the multi-temporal remote sensing image to obtain a corrected multi-temporal image; Performing geometric correction on the corrected multi-temporal phase image to obtain a target multi-temporal phase image; Extracting visual features corresponding to the visual unit in the target multi-temporal image; Constructing a visual label corresponding to the visual unit according to the visual feature, and identifying a water area label and a landform label in the visual label; The target multi-temporal map is processed by combining the water area label and the landform label to obtain a multi-temporal water area map and a multi-temporal landform map.

5. The ecological intelligent monitoring system based on three-dimensional simulation visualization as claimed in claim 4 is characterized in that: The identifying of the water area label and the landform label in the visual label comprises: Calculating a label gain value corresponding to the visual label, and extracting a representation label in the visual label according to the label gain value; Identify tag characters in the representation tag, and extract key characters in the tag characters; Performing semantic analysis on the key characters to obtain character semantics; According to the character semantics, a water area label and a landform label in the visual label are identified.

6. The ecological intelligent monitoring system based on three-dimensional simulation visualization as claimed in claim 1, characterized in that: The extracting of geomorphic features corresponding to the geomorphic structure comprises: Performing grayscale processing on the landform structure to obtain a grayscale landform structure; Smoothing the grayscale landform structure to obtain a smooth landform structure; Extracting structural texture features corresponding to the smooth landform structure; Performing dimensionality reduction processing on the structural texture features to obtain dimensionality reduced texture features; Calculating a feature energy value corresponding to the dimensionally reduced texture feature; According to the feature energy value, feature selection is performed on the dimension-reduced texture feature to obtain a target texture feature; The target texture feature is used as the landform feature corresponding to the landform structure.

7. The ecological intelligent monitoring system based on three-dimensional simulation visualization as claimed in claim 6 is characterized in that: The calculating the feature energy value corresponding to the dimension-reduced texture feature includes: Assigning a feature weight corresponding to the reduced-dimensionality texture feature, and calculating a texture feature value corresponding to the reduced-dimensionality texture feature; Calculating a feature average value corresponding to the reduced-dimensional texture feature according to the texture feature value; The feature energy value corresponding to the reduced-dimensional texture feature is calculated by combining the feature weight, the feature average value and the texture feature value.

8. The ecological intelligent monitoring system based on three-dimensional simulation visualization as claimed in claim 1 is characterized in that: Calculating the landform change amplitude between the landform types according to the multi-temporal landform map includes: Measuring the pixel brightness value corresponding to each image in the multi-temporal landform map; According to the pixel brightness values, calculating the average brightness value corresponding to each image in the multi-temporal landform map; Calculate the pixel discreteness corresponding to each image in the multi-temporal landform map; The landform change amplitude between the landform types is calculated by combining the brightness average value and the pixel discreteness.

9. The ecological intelligent monitoring system based on three-dimensional simulation visualization as claimed in claim 1, characterized in that: The combining the water area elements and the multi-temporal water area map to calculate the element deviation corresponding to each element in the water area elements includes: According to the water area elements, the multi-temporal water area map is segmented to obtain a water area element map; Extracting image parameters corresponding to the water element map; Performing standardization processing on the image parameters to obtain standard image parameters; The element deviation corresponding to each element in the water area element is calculated according to the standard image parameters.

10. A method for ecological intelligent monitoring based on three-dimensional simulation visualization, characterized in that: The method comprises: Determine an ecological area to be monitored, collect ecological data streams corresponding to the ecological area, calculate an angle anomaly factor between each data in the ecological data stream, and optimize the quality of the ecological data stream according to the angle anomaly factor to obtain a target data stream; Collecting multi-temporal remote sensing images of the ecological region, identifying visual units in the multi-temporal remote sensing images, and performing image block processing on the multi-temporal remote sensing images according to the visual units to obtain a multi-temporal water area map and a multi-temporal landform map; Identify the geomorphic structure in the multi-temporal geomorphic map, extract the geomorphic features corresponding to the geomorphic structure, perform semantic segmentation on the geomorphic structure according to the geomorphic features to obtain the geomorphic type, calculate the geomorphic change amplitude between the geomorphic types according to the multi-temporal geomorphic map, and analyze the geomorphic evolution trend of the ecological region according to the geomorphic change amplitude; Constructing water elements of the multi-temporal water map, combining the water elements and the multi-temporal water map, calculating the element deviation corresponding to each element in the water elements, and analyzing the water evolution trend of the ecological region according to the element deviation; In combination with the multi-temporal landform map and the multi-temporal water area map, a three-dimensional simulation area corresponding to the ecological area is constructed. In combination with the landform evolution trend, the water area evolution trend and the target data stream, the three-dimensional simulation area is dynamically monitored to obtain regional ecological monitoring results.

Images