Grassland restoration state dynamic monitoring system and method based on multi-scale fusion

Abstract

The application provides a grassland recovery state dynamic monitoring system and method based on multi-scale fusion, which comprises: a multi-scale data acquisition module for acquiring macro monitoring data of the grassland by satellite remote sensing technology, and simultaneously acquiring micro monitoring data of the grassland by a UAV; a long time sequence data analysis module for collecting macro monitoring data and micro monitoring data in a long time, forming a time sequence and performing fusion analysis to determine the actual growth state and recovery effect of the grassland; a prediction and strategy adjustment module for predicting the grassland growth trend and adjusting the grassland recovery strategy based on the actual growth and recovery effect of the grassland, in combination with actual grassland environment data and historical time sequence. Through the combination of macro monitoring of satellite remote sensing and micro monitoring of the UAV, a dynamic monitoring system is established to realize accurate monitoring and prediction of the grassland growth trend, and to provide timely and accurate data support for management personnel to adjust the recovery strategy.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of satellite navigation positioning monitoring, in particular to a grassland restoration state dynamic monitoring system and method based on multi-scale fusion. BACKGROUND

[0002] Grassland, as an important part of the terrestrial ecosystem, its ecological restoration is of great significance to maintain ecological balance and promote the development of animal husbandry. Traditional grassland monitoring methods have limited monitoring range, insufficient accuracy and other problems, which cannot meet the needs of comprehensive, dynamic monitoring and accurate prediction of grassland restoration. With the development of satellite remote sensing and unmanned aerial vehicle technology, new means are provided for grassland monitoring, but how to effectively integrate these different scale data to realize scientific evaluation and accurate prediction of grassland restoration is still a problem to be solved. SUMMARY

[0003] The present application provides a grassland restoration state dynamic monitoring system and method based on multi-scale fusion, which combines macro monitoring of satellite remote sensing and micro monitoring of unmanned aerial vehicle to establish a dynamic monitoring system, realizes accurate monitoring and prediction of grassland growth trend, and provides reliable data support for adjustment of grassland restoration strategy.

[0004] The present application provides a grassland restoration state dynamic monitoring system based on multi-scale fusion, comprising:

[0005] A multi-scale data acquisition module is used to acquire macro monitoring data of the grassland by satellite remote sensing technology, and micro monitoring data of the grassland by unmanned aerial vehicle;

[0006] A long time series data analysis module is used to collect macro monitoring data and micro monitoring data over a long period of time, form a time series, and analyze the data in the current time series to determine the actual growth state and restoration effect of the grassland;

[0007] A prediction and strategy adjustment module is used to predict the growth trend of the grassland based on the actual growth and restoration effect of the grassland, combined with actual grassland environment data and historical time series, and adjust the grassland restoration strategy according to the prediction.

[0008] Preferably, in a grassland restoration state dynamic monitoring system based on multi-scale fusion, comprising:

[0009] A macro data acquisition unit is used to collect panoramic images and non-panoramic large-area grassland images of the grassland by satellite remote sensing technology to generate macro monitoring data;

[0010] A micro data acquisition unit is used to collect detailed images of local areas of the grassland by unmanned aerial vehicle to generate micro monitoring data.

[0011] Preferably, in a grassland restoration state dynamic monitoring system based on multi-scale fusion, the multi-scale data acquisition module comprises:

[0012] The time sequence generation unit is configured to store the macro monitoring data and the micro monitoring data in sequence based on the time axis, and generate a time sequence.

[0013] The sequence data fusion unit is configured to extract features of the macro monitoring data and the micro monitoring data in the time sequence respectively, obtain grassland topographic features and vegetation features, fuse the grassland topographic features and the vegetation features based on the corresponding relationship of the collection area, and obtain a current fusion feature image.

[0014] The fusion data analysis unit is configured to analyze the current fusion feature image, determine the actual growth condition and the growth state of the grassland vegetation, and compare the actual growth condition of the grassland vegetation with an actual growth condition corresponding to a previous fusion feature image to determine the restoration effect of the grassland.

[0015] The fusion data analysis unit is configured to analyze the current fusion feature image, determine the actual growth condition and the growth state of the grassland vegetation, and compare the actual growth condition of the grassland vegetation with an actual growth condition corresponding to a previous fusion feature image to determine the restoration effect of the grassland.

[0016] Preferably, in a grassland restoration state dynamic monitoring system based on multi-scale fusion, the sequence data fusion unit comprises:

[0017] The macro feature extraction subunit is configured to perform edge detection processing on the grassland panoramic image and the non-panoramic large-area image in the macro monitoring data respectively, and obtain texture features and color features of the grassland panoramic image and the non-panoramic large-area image respectively.

[0018] The macro feature fusion subunit is configured to fuse the texture features and the color features of the non-panoramic large-area image and the grassland panoramic image based on the collection area of the non-panoramic large-area image, and obtain a primary panoramic image.

[0019] The micro feature extraction subunit is configured to extract features of the detail image corresponding to the micro monitoring data, and obtain grassland topographic features and vegetation detail features.

[0020] The macro feature extraction subunit is configured to determine the region correspondence relationship between the detail image and the primary panoramic image based on the texture features of the primary panoramic image and the shooting position of the micro monitoring data, fuse the grassland topographic features and the vegetation detail features of the detail image with the corresponding region of the primary panoramic image based on the prime number region correspondence relationship, and obtain a fusion feature image.

[0021] Preferably, in a grassland restoration state dynamic monitoring system based on multi-scale fusion, the fusion data analysis unit comprises:

[0022] The prairie feature extraction unit is configured to perform feature extraction on the current fused feature image to obtain the topography features and the vegetation features of the prairie.

[0023] The vegetation determination unit is configured to determine the topography distribution of the prairie based on the prairie topography features, and determine the vegetation information corresponding to different topographies in combination with the prairie management data.

[0024] The growth analysis unit is configured to determine the main vegetation species of each region on the prairie based on the vegetation information and the topography distribution, obtain the standard vegetation growth data corresponding to different vegetation species in a database based on the vegetation species, determine the vegetation sub-features corresponding to each region on the prairie respectively, determine the actual growth conditions of the vegetation in each region based on the vegetation sub-features, and determine the growth state of the vegetation in each region on the prairie according to the actual growth conditions of the vegetation and the standard vegetation growth data corresponding to the vegetation.

[0025] The effect determination unit is configured to compare the actual growth conditions of the prairie vegetation with the actual growth conditions corresponding to the previous fused feature image, and obtain the growth difference corresponding to different topographies of the prairie respectively.

[0026] The effect determination unit is configured to compare the actual growth conditions of the prairie vegetation with the actual growth conditions corresponding to the previous fused feature image, and obtain the growth difference corresponding to different topographies of the prairie respectively.

[0027] The effect determination unit is configured to compare the actual growth conditions of the prairie vegetation with the actual growth conditions corresponding to the previous fused feature image, and obtain the growth difference corresponding to different topographies of the prairie respectively.

[0028] The effect determination unit is configured to compare the actual growth conditions of the prairie vegetation with the actual growth conditions corresponding to the previous fused feature image, and obtain the growth difference corresponding to different topographies of the prairie respectively.

[0029] The effect determination unit is configured to compare the actual growth conditions of the prairie vegetation with the actual growth conditions corresponding to the previous fused feature image, and obtain the growth difference corresponding to different topographies of the prairie respectively.

[0030] The effect determination unit is configured to compare the actual growth conditions of the prairie vegetation with the actual growth conditions corresponding to the previous fused feature image, and obtain the growth difference corresponding to different topographies of the prairie respectively.

[0031] The effect determination unit is configured to compare the actual growth conditions of the prairie vegetation with the actual growth conditions corresponding to the previous fused feature image, and obtain the growth difference corresponding to different topographies of the prairie respectively.

[0032] The effect determination unit is configured to compare the actual growth conditions of the prairie vegetation with the actual growth conditions corresponding to the previous fused feature image, and obtain the growth difference corresponding to different topographies of the prairie respectively.

[0033] The effect determination unit is configured to compare the actual growth conditions of the prairie vegetation with the actual growth conditions corresponding to the previous fused feature image, and obtain the growth difference corresponding to different topographies of the prairie respectively.

[0034] The growth trend prediction unit is configured to acquire actual environmental data of the grassland and historical time series, generate a vegetation growth curve for each region of the grassland based on the vegetation growth in the historical time series corresponding to each region of the grassland, and determine the vegetation growth dynamics corresponding to each region.

[0035] According to the generated curve and the current vegetation growth state, the growth trend of the vegetation in each region of the grassland and the change in the coverage of the grassland are predicted.

[0036] The intelligent strategy adjustment unit is configured to adjust the current grassland restoration strategy based on the growth trend and the change in the coverage of the grassland and in combination with the actual environmental data of the grassland.

[0037] Preferably, in the grassland restoration state dynamic monitoring system based on multi-scale fusion, the growth trend prediction unit comprises:

[0038] The growth curve generation subunit is configured to determine the vegetation height and the vegetation color based on the vegetation growth, predict the peak point corresponding to the optimal vegetation growth state of the vegetation in the corresponding region based on the vegetation height change and in reference to the height interval corresponding to the optimal vegetation growth state of the corresponding vegetation, and generate the vegetation growth curve for the corresponding region based on the vegetation height change and in combination with the predicted peak value and the time interval corresponding to the predicted peak value.

[0039] The peak point is taken as the predicted peak value of the vegetation growth curve for the corresponding region, and the time interval corresponding to the predicted peak value is determined in combination with the time series.

[0040] The vegetation growth curve for the corresponding region is generated based on the vegetation height change and in combination with the predicted peak value and the time interval corresponding to the predicted peak value, and the vegetation color at each point on the vegetation growth curve is marked.

[0041] The trend prediction subunit is configured to acquire the direction vector corresponding to the end of the curve of the vegetation growth curve for each region, and obtain the predicted vegetation growth trend direction.

[0042] Meanwhile, the target growth data corresponding to the target grassland vegetation is acquired based on big data, and the first average growth speed or the second average decay speed of the target vegetation under the current solar term is determined based on the target growth data.

[0043] The actual average growth speed or the actual decay speed of the actual vegetation is determined based on the vegetation growth curve corresponding to the target grassland vegetation, and the environmental impact growth difference is obtained by comparing the first average growth speed or the actual average growth speed or the first average decay speed and the actual first average decay speed.

[0044] In the vegetation growth curve for the corresponding region, a plurality of target curve segments corresponding to similar vegetation growth trends are intercepted, and the second average growth speed or the second average decay speed corresponding to the plurality of target curve segments is calculated.

[0045] Based on the second average growth rate or the second average decline rate, the growth difference combined with the environmental influence is determined to determine the change of the vegetation in the corresponding region in the predicted vegetation growth trend direction, and a vegetation growth trend prediction result of the corresponding region is obtained.

[0046] The application provides a grassland restoration state dynamic monitoring method based on multi-scale fusion, comprising:

[0047] Step 1: macroscopic monitoring data of the grassland is obtained by using satellite remote sensing technology, and microcosmic monitoring data of the grassland is obtained by using a UAV;

[0048] Step 2: macroscopic monitoring data and microcosmic monitoring data in a long time are collected to form a time sequence, and fusion analysis is performed on the data in the current time sequence to determine the actual growth state and restoration effect of the grassland;

[0049] Step 3: based on the actual growth and restoration effect of the grassland, actual grassland environmental data and historical time sequences are combined to predict the growth trend of the grassland, and the grassland restoration strategy is adjusted according to the prediction.

[0050] Preferably, in the grassland restoration state dynamic monitoring method based on multi-scale fusion, step 2 comprises:

[0051] The macroscopic monitoring data and the microcosmic monitoring data are stored based on a time axis sequence to generate a time sequence;

[0052] The macroscopic monitoring data and the microcosmic monitoring data in the time sequence are respectively subjected to feature extraction to obtain grassland topographic features and vegetation features, the grassland topographic features and the vegetation features are fused based on the corresponding relationship of the collection region to obtain a current fusion feature image;

[0053] The current fusion feature image is analyzed to determine the actual growth condition and growth state of the grassland vegetation;

[0054] The actual growth condition of the grassland vegetation is compared with the actual growth condition corresponding to a previous fusion feature image to determine the restoration effect of the grassland.

[0055] Preferably, in the grassland restoration state dynamic monitoring method based on multi-scale fusion, the current fusion feature image is analyzed to determine the actual growth condition and growth state of the grassland vegetation; and the actual growth condition of the grassland vegetation is compared with the actual growth condition corresponding to a previous fusion feature image to determine the restoration effect of the grassland, comprising:

[0056] The current fusion feature image is subjected to feature extraction to obtain the topographic features and the vegetation features of the grassland;

[0057] Based on the grassland topographic features, the topographic distribution of the grassland is determined, and the vegetation information corresponding to different topographies is determined in combination with the grassland management data;

[0058] Based on the vegetation information and the topographic distribution, the main vegetation species of each region on the grassland is determined, and based on the vegetation species, the standard vegetation growth data corresponding to different vegetation is obtained in the database;

[0059] The vegetation sub-features corresponding to each region of the grassland are determined respectively, the actual growth of the vegetation of each region is determined based on the vegetation sub-features, and the growth state of the vegetation of each region on the grassland is determined according to the actual growth of the vegetation and the standard vegetation growth data corresponding to the vegetation;

[0060] The actual growth of the grassland vegetation is compared with the actual growth corresponding to the fusion feature image of the last time, and the growth difference corresponding to different topographies of the grassland is obtained respectively;

[0061] Based on the growth difference, the actual growth state of each region is combined to determine whether the vegetation growth trend of the grassland is good, if so, it is determined that the current region has good recovery effect;

[0062] Otherwise, based on the current solar term change of the grassland and the vegetation growth data corresponding to each region, it is determined whether the current region vegetation growth trend is not good or in normal state;

[0063] If not, it is determined that the current region has poor recovery effect;

[0064] If so, based on the fusion feature image corresponding to the best vegetation growth state of the current region, the vegetation coverage corresponding to the current region is determined, and whether the current region coverage meets the standard is determined in combination with the preset coverage threshold of the topography corresponding to the current region;

[0065] If the current region coverage meets the standard, it is determined that the current region has good recovery effect;

[0066] Otherwise, it is determined that the current region has poor recovery effect.

[0067] Compared with the prior art, the present application has at least the following beneficial effects:

[0068] The present application combines macroscopic monitoring of satellite remote sensing and microscopic monitoring of unmanned aerial vehicles to dynamically monitor grassland restoration in multiple scales and in all directions. Macroscopic monitoring data and microscopic monitoring data over a long period of time are collected to form a time series, and the data in the current time series are fused and analyzed to determine the actual growth state and restoration effect of the grassland, so that the overall monitoring of the grassland restoration is automatically analyzed, the actual growth state and restoration effect of the grassland are obtained, the actual restoration of the grassland is quickly determined by the management personnel, and finally the growth trend of the grassland is predicted based on the actual growth and restoration effect of the grassland, combined with actual grassland environment data and historical time series, which is beneficial to timely find the adverse factors of grassland restoration, provide timely and accurate data support for the management personnel to adjust the restoration strategy, help to improve the efficiency and quality of grassland restoration, and promote the sustainable development of the grassland ecosystem.

[0069] Other features and advantages of the present application will be set forth in the following description, and in part will become apparent to those skilled in the art from the description, or can be learned by practice of the present application. The objects and other advantages of the present application can be realized and attained by the structure particularly pointed out in the specification.

[0070] The technical solutions of the present application will be further described in detail below by means of the drawings and examples. BRIEF DESCRIPTION OF DRAWINGS

[0071] The accompanying drawings are included to provide a further understanding of the present application, and constitute a part of the specification, together with the embodiments of the present application, to explain the present application, and do not constitute a limitation on the present application. In the drawings:

[0072] Figure 1 It is a structure diagram of a grassland restoration state dynamic monitoring system based on multi-scale fusion of the present application;

[0073] Figure 2 It is a structure diagram of a multi-scale data acquisition module of a grassland restoration state dynamic monitoring system based on multi-scale fusion of the present application;

[0074] Figure 3 It is a structure diagram of a long time series data analysis module of a grassland restoration state dynamic monitoring system based on multi-scale fusion of the present application;

[0075] Figure 4 It is a structure diagram of a prediction and strategy adjustment module of a grassland restoration state dynamic monitoring system based on multi-scale fusion of the present application;

[0076] Figure 5 It is a flowchart of a grassland restoration state dynamic monitoring method based on multi-scale fusion of the present application. DETAILED DESCRIPTION

[0077] The preferred embodiments of the present application are described below in conjunction with the accompanying drawings, in which it is understood that the preferred embodiments described below are merely intended to illustrate and explain the present application, and are not intended to limit the present application.

[0078] Embodiment 1

[0079] The present application provides a grassland restoration state dynamic monitoring system based on multi-scale fusion, as shown in Figure 1 The present application provides a grassland restoration state dynamic monitoring system based on multi-scale fusion, as shown in

[0080] The multi-scale data acquisition module is used to acquire macro monitoring data of the grassland by satellite remote sensing technology, and acquire micro monitoring data of the grassland by a UAV;

[0081] The long time sequence data analysis module is used to collect macro monitoring data and micro monitoring data in a long time, form a time sequence, and perform fusion analysis on data in the current time sequence to determine the actual growth state and restoration effect of the grassland.

[0082] The prediction and strategy adjustment module is used to predict the growth trend of the grassland based on the actual growth and restoration effect of the grassland, in combination with actual grassland environment data and historical time sequences, and adjust the grassland restoration strategy according to the prediction.

[0083] In this embodiment, the long time refers to a period of time, and the time is greater than or equal to 72 hours.

[0084] The above technical solution has the following beneficial effects: The present application combines macro monitoring by satellite remote sensing and micro monitoring by a UAV to perform multi-scale and all-around dynamic monitoring of grassland restoration. Macro monitoring data and micro monitoring data in a long time are collected to form a time sequence, and fusion analysis is performed on data in the current time sequence to determine the actual growth state and restoration effect of the grassland, so that comprehensive monitoring of the grassland restoration state is automatically analyzed to obtain the actual growth state and restoration effect of the grassland, which facilitates management personnel to quickly determine the actual restoration state of the grassland. Based on the actual growth and restoration effect of the grassland, in combination with actual grassland environment data and historical time sequences, the growth trend of the grassland is predicted, which is conducive to timely discovering adverse factors of grassland restoration, providing timely and accurate data support for management personnel to adjust the restoration strategy, and helping to improve the efficiency and quality of grassland restoration and promote the sustainable development of the grassland ecological system.

[0085] Embodiment 2

[0086] In the grassland restoration state dynamic monitoring system based on multi-scale fusion of embodiment 1, the multi-scale data acquisition module, as shown in Figure 2 The present application provides a grassland restoration state dynamic monitoring system based on multi-scale fusion, as shown in

[0087] A macro data acquisition unit is configured to collect panoramic images and non-panoramic large-area grassland images of the grassland by using satellite remote sensing technology, and generate macro monitoring data.

[0088] A micro data acquisition unit is configured to collect detail images of a local area of the grassland by using a UAV, and generate micro monitoring data.

[0089] The technical scheme has the following beneficial effects: The macro data acquisition unit collects panoramic images and non-panoramic large-area grassland images of the grassland by using satellite remote sensing technology, and generates macro monitoring data. Satellite remote sensing has a wide field of view, and can cover a large-area grassland region from a macro perspective, thereby providing a basis for grasping the overall topography, vegetation distribution, and land use of the grassland. For example, the panoramic images can be used to clearly understand the boundary range, overall morphology, and relationship with the surrounding geographical environment of the grassland. The non-panoramic large-area images can be used to more carefully observe the characteristic differences of different regions of the grassland, such as the distribution areas of different vegetation types, thereby providing comprehensive data support for macro ecological analysis and planning. Meanwhile, the micro data acquisition unit collects detail images of a local area of the grassland by using a UAV, and generates micro monitoring data. The UAV has high flexibility, and can be used to in-depth study the characteristics of a specific local area of the grassland, thereby obtaining high-resolution detail images, and focusing on micro elements in the grassland ecosystem, which is helpful for in-depth study of the microstructure and ecological processes of the grassland ecosystem.

[0090] Embodiment 3

[0091] Based on the embodiment 1, the multi-scale data acquisition module, as shown in Figure 3 includes:

[0092] A time sequence generation unit is configured to store the macro monitoring data and the micro monitoring data in a time sequence, and generate a time sequence.

[0093] A sequence data fusion unit is configured to respectively extract features of the macro monitoring data and the micro monitoring data in the time sequence, obtain grassland topographic features and vegetation features, fuse the grassland topographic features and the vegetation features based on a corresponding relationship of the collection area, and obtain a current fusion feature image.

[0094] A fusion data analysis unit is configured to analyze the current fusion feature image, determine an actual growth condition and a growth state of the grassland vegetation, and compare the actual growth condition of the grassland vegetation with an actual growth condition corresponding to a previous fusion feature image, and determine a recovery effect of the grassland.

[0095]

[0096] ​The beneficial effects of the above technical solutions are: the present application collects macro monitoring data and micro monitoring data of the grassland in time sequence, fuses the macro monitoring data and the micro monitoring data through a sequence data fusion unit, improves the accuracy of the grassland monitoring data, is beneficial to the managers to have a more comprehensive and in-depth understanding of the grassland vegetation restoration situation, and provides a more comprehensive and accurate data basis for determining the grassland restoration effect, finally, through a fused data analysis unit, the current fused feature image is analyzed to determine the actual growth situation and growth state of the grassland vegetation; and the actual growth situation of the grassland vegetation is compared with the actual growth situation corresponding to the previous fused feature image to determine the restoration effect of the grassland, so that the automatic analysis of the grassland restoration situation is realized, the monitoring workload of the managers is reduced, and the grassland restoration monitoring feedback efficiency is effectively improved.

[0097] Embodiment 4:

[0098] On the basis of embodiment 3, the sequence data fusion unit comprises:

[0099] The macro feature extraction subunit is configured to perform edge detection processing on the grassland panoramic image and the non-panoramic large-area image in the macro monitoring data respectively, and obtain the texture features and color features of the grassland panoramic image and the non-panoramic large-area image respectively.

[0100] The macro feature fusion subunit is configured to fuse the texture features and color features of the non-panoramic large-area image and the grassland panoramic image based on the collection area of the non-panoramic large-area image, and obtain a primary panoramic image.

[0101] The micro feature extraction subunit is configured to perform feature extraction on the detail image corresponding to the micro monitoring data, and obtain the grassland topography and vegetation detail features.

[0102] The macro feature extraction subunit is configured to determine the region correspondence relationship between the detail image and the primary panoramic image based on the texture features of the primary panoramic image and the shooting position of the micro monitoring data, fuse the grassland topography and vegetation detail features of the detail image with the corresponding region of the primary panoramic image based on the prime number region correspondence relationship, and obtain a fused feature image.

[0103] The beneficial effects of the above technical solutions are: firstly, the image features of the prairie panoramic image and the non-panoramic large-area image are fused by the macro feature extraction subunit and the macro feature fusion subunit, so that the image information at the macro level is more complete and rich, the overall field of view of the panoramic image and the local detailed features of the non-panoramic large-area image are combined, the macro state of the prairie can be more comprehensively reflected, and a foundation is laid for subsequent micro feature fusion; secondly, based on the texture features of the primary panoramic image, the shooting position of the micro monitoring data is combined to determine the regional correspondence between the detail image and the primary panoramic image, and then the prairie topography and vegetation detail features of the detail image are fused with the corresponding region of the primary panoramic image to obtain a fusion feature image, which organically combines macro and micro features, retains the macro information of the prairie as a whole, and integrates the micro details of the key region, so that the generated fusion feature image can comprehensively and meticulously reflect the real situation of the prairie.

[0104] Embodiment 5:

[0105] On the basis of embodiment 3, the fusion data analysis unit comprises:

[0106] The prairie feature extraction unit is configured to extract features from the current fusion feature image to obtain the topographic features and vegetation features of the prairie.

[0107] The vegetation determination unit is configured to determine the topographic distribution of the prairie based on the topographic features of the prairie, and determine the vegetation information corresponding to different topographies in combination with the prairie management data.

[0108] The growth analysis unit is configured to determine the main vegetation species in each region of the prairie based on the vegetation information and the topographic distribution, and obtain the standard vegetation growth data corresponding to different vegetation species in the database based on the vegetation species.

[0109] The vegetation sub-feature corresponding to each region of the prairie is determined respectively, the actual growth of the vegetation in each region is determined based on the vegetation sub-feature, and the growth state of the vegetation in each region of the prairie is determined based on the actual growth of the vegetation and the standard vegetation growth data corresponding to the vegetation.

[0110] The effect determination unit is configured to compare the actual growth of the prairie vegetation with the actual growth corresponding to the previous fusion feature image to obtain the growth difference corresponding to different topographies of the prairie respectively.

[0111] Based on the growth difference and the actual growth state of each region, it is determined whether the vegetation growth trend of the prairie is good, and if so, it is determined that the current region has good recovery effect.

[0112] Otherwise, based on the current solar term change of the prairie and the vegetation growth data corresponding to each region, it is determined whether the vegetation growth trend of the current region is not good and is in a normal state.

[0113] If not, it is determined that the current area has poor recovery effect;

[0114] If yes, based on the fusion feature image corresponding to the optimal vegetation growth state of the current area, the vegetation coverage corresponding to the current area is determined, and the preset coverage threshold of the topography corresponding to the current area is combined to determine whether the current area coverage meets the standard;

[0115] If the current area coverage meets the standard, it is determined that the current area has good recovery effect;

[0116] Otherwise, it is determined that the current area has poor recovery effect.

[0117] The beneficial effects of the above technical solutions are: firstly, the current fusion feature image is subjected to feature extraction to obtain the topographic features and vegetation features of the grassland, then the topographic distribution is determined based on the topographic features of the grassland, and the vegetation information corresponding to different topographies is determined in combination with the grassland management data, thereby establishing a close relationship between the topography and the vegetation, determining the vegetation types suitable for growth or target pre-planting under different topographic conditions, and providing an accurate basis for subsequent growth analysis of the grassland vegetation; thereafter, based on the vegetation information and the topographic distribution, the main vegetation types of each area on the grassland are determined, based on the vegetation types, the standard vegetation growth data corresponding to different vegetation in the database is obtained; the vegetation sub-features corresponding to each area of the grassland are determined respectively, based on the vegetation sub-features, the actual growth conditions of the vegetation of each area are determined respectively, and based on the actual growth conditions and the standard vegetation growth data corresponding to the vegetation, the growth state of the vegetation of each area on the grassland is determined respectively, which can objectively and accurately evaluate the growth state of the vegetation of each area; finally, by comparing the actual growth conditions of the vegetation corresponding to the current and the previous fusion feature images, the growth difference of different topographies is obtained, which directly reflects the growth trend of the grassland vegetation in a period of time, helps to discover abnormal conditions in the growth process of the vegetation in a timely manner, and provides an important basis for evaluating the recovery effect; in combination with the growth difference and the actual growth state, the growth trend of the vegetation is determined, the influence of the solar term change on the growth of the vegetation is considered, and the recovery effect is comprehensively judged, which fully considers the complexity and dynamics of the grassland ecosystem, combines the solar term and the vegetation growth data, can effectively avoid misjudgment, and further improves the scientificity and accuracy of the evaluation. The present application can clearly understand the recovery effect of different areas of the grassland, which helps to reasonably allocate the grassland management resources and adjust the subsequent grassland recovery strategy.

[0118] Example 6:

[0119] On the basis of example 5, the prediction and strategy adjustment module, as shown in Figure 4 , includes:

[0120] a growth trend prediction unit configured to acquire actual environment data of the grassland and historical time series, generate a vegetation growth curve diagram corresponding to each region of the grassland based on a vegetation growth condition of each region in the historical time series, and determine a vegetation growth dynamic corresponding to each region;

[0121] According to the vegetation growth curve diagram, the growth trend of the vegetation in each region of the grassland and the change of the grassland coverage are predicted respectively in combination with a current vegetation growth state;

[0122] an intelligent strategy adjustment unit configured to adjust a current grassland restoration strategy based on the growth trend and the change of the grassland coverage in combination with the actual environment data of the grassland.

[0123] In this embodiment, the intelligent strategy adjustment unit comprises:

[0124] a first adjustment subunit configured to acquire the growth trend of the vegetation in each region of the grassland and the change of the grassland coverage, respectively, determine whether the vegetation in each region tends to be in good condition, and if so, maintain a current grassland restoration strategy corresponding to the region;

[0125] Otherwise, the current region is taken as a target region, and whether the target region is caused by environmental impact is determined in combination with the current grassland environment data. If so, the vegetation growth optimal environment data of the target region is compared with the current grassland environment data to determine an environmental impact factor.

[0126] Based on the environmental impact data, a best intervention scheme is obtained by screening artificial intervention modes, and the current grassland restoration strategy is adjusted based on the best intervention scheme;

[0127] a second adjustment subunit configured to, when the target region is not caused by environmental impact, acquire biological activity data of the target region, determine all involved regions of the biological activity, compare the growth trend of the vegetation in a plurality of involved regions corresponding to the same biological activity and the grassland coverage of the regions, respectively, and determine an influence of each biological activity on the grassland restoration;

[0128] According to the biological activity data corresponding to the target region, a target biological activity contained in the target region is determined, and a restriction strategy of the target biological activity is generated;

[0129] The current grassland restoration strategy is adjusted based on the restriction strategy;

[0130] If the growth trend of the vegetation in the target region and the change of the grassland coverage are not improved after the biological activities are limited, the preset database is screened based on the environmental data of the target region and the growth data of different vegetation in the target region to determine the best growth vegetation in the target region, and the current grassland restoration strategy is adjusted based on the best growth vegetation and the corresponding planting mode thereof.

[0131] In this embodiment, the biological activity data refers to animal activity data on the grassland, including the change of the grassland data caused by human grazing.

[0132] The present application determines whether to adjust the strategy by judging the vegetation growth of each region. For the region with good vegetation growth, the existing strategy is maintained to avoid unnecessary adjustment and ensure the stability and continuity of the management work. For the region with poor vegetation growth, it is set as a target region, and the grassland restoration strategy of the target region is adjusted accordingly to realize the partition management of the entire grassland, making the strategy adjustment more accurate and enabling the resources to be concentrated in the areas that really need to be improved. When determining the target region is affected by environmental factors, the influencing factors are found out by comparing the best environmental data with the current environmental data, and the best intervention scheme is selected to adjust the restoration strategy, realizing the accurate analysis of the environmental factors of the target region and quickly solving the vegetation growth problems caused by the unsuitable environment. For example, if it is found that the soil fertility of a certain region affects the vegetation growth, the fertilization scheme can be adjusted accordingly to improve the soil fertility and promote the healthy growth of the vegetation. When the target region is affected by non-environmental factors, the vegetation growth trend and grassland coverage of multiple regions involved in the same biological activity are compared by obtaining biological activity data to comprehensively determine the influence of various biological activities on grassland restoration, which helps to deeply understand the complex relationship between biological activities and grassland vegetation growth and provides a basis for formulating a reasonable biological activity management strategy. For example, through analysis, it is found that the overgrazing of a certain type of animal leads to inhibited vegetation growth, so measures can be taken to limit the activity range or number of the animal, and the target biological activity is determined according to the biological activity data of the target region to generate a restriction strategy, and then the restoration strategy is adjusted to realize the customization of the restoration strategy of the region affected by biological activities, effectively control the negative impact of biological activities on grassland vegetation, and protect the balance of the grassland ecosystem. For example, for harmful biological activities that destroy vegetation, appropriate restriction strategies such as setting up protective fences and using biological control methods can be formulated to reduce the damage to the vegetation. If the vegetation growth in the target region does not improve after limiting the biological activity, the best growing vegetation and planting method are selected from the preset database to adjust the strategy by combining the environmental data and different vegetation growth data, which adopts a multi-dimensional comprehensive adjustment method, fully considers the complexity of the grassland ecosystem, analyzes from the aspects of environmental adaptability, biological activity influence and vegetation selection, and ensures that the restoration strategy can promote the growth and restoration of the grassland vegetation to the greatest extent. For example, when it is found that the existing vegetation cannot grow well in the current environment, a more suitable vegetation variety can be selected for planting to improve the survival rate and growth quality of the vegetation. The present application adjusts the grassland restoration strategy based on the actual situation and multiple factors, effectively solves various problems encountered in the growth process of the grassland vegetation, significantly improves the management effect of the grassland restoration work, promotes the healthy and sustainable development of the grassland ecosystem, and realizes the partition management of the grassland, improves the accuracy of the grassland restoration strategy adjustment, avoids blind investment of resources, improves the resource utilization efficiency, and realizes the optimal allocation of resources.

[0133] The beneficial effects of the above technical solutions are: the actual environment data and the historical time sequence of the grassland are acquired, the vegetation growth curve graph is generated for each region, the growth of the vegetation at different time points is intuitively displayed, the dynamic change process of the vegetation growth can be clearly observed by the management personnel, the growth rate of the vegetation in a certain time period can be judged through the slope of the curve, and the activity degree of the vegetation growth in the region is analyzed, which helps to deeply understand the internal law of the vegetation growth in different regions, provides a solid foundation for subsequent growth trend prediction, improves the prediction accuracy, the growth trend of the vegetation in each region of the grassland and the change of the grassland coverage rate are predicted based on the vegetation growth curve graph and the current vegetation growth state, the vegetation growth dynamics of different regions may be different due to the differences in environmental factors such as terrain, soil and illumination, the historical data and the current actual situation of the vegetation growth are fully considered, the future growth trend of the vegetation and the macro development trend of the grassland vegetation can be more accurately predicted, and the current grassland restoration strategy is adjusted based on the growth trend and the change of the grassland coverage rate and the actual environment data of the grassland, which is beneficial to the reasonable allocation of grassland restoration resources, improves the efficiency and quality of grassland management, for example, for the region where the vegetation grows slowly and the coverage rate may decrease, the restoration strategy can be adjusted, the irrigation frequency is increased, the soil fertility is improved or the planted vegetation species is adjusted to promote the growth of the vegetation and improve the coverage rate. The actual environment data of the grassland is considered for strategy adjustment, which ensures that the restoration strategy can adapt to the dynamic change of the grassland environment, makes the restoration strategy more suitable for the current environmental conditions, improves the effectiveness and sustainability of the restoration strategy, for example, in dry years, the irrigation strategy is adjusted according to the actual environment data of the reduced precipitation, the water resource utilization is optimized, and the normal growth of the vegetation is ensured.

[0134] Embodiment 7:

[0135] Based on the embodiment 5, the growth trend prediction unit comprises:

[0136] The growth curve generation subunit is configured to determine the vegetation height and the vegetation color based on the vegetation growth condition, predict the peak point corresponding to the best vegetation growth state of the vegetation in the corresponding region based on the vegetation height change and in reference to the height interval corresponding to the best vegetation growth state of the corresponding vegetation.

[0137] The peak point is taken as the predicted peak value of the vegetation growth curve graph of the corresponding region, and the time interval corresponding to the predicted peak value is determined in combination with the time sequence;

[0138] The vegetation growth curve of the corresponding region is generated based on the vegetation height change in combination with the predicted peak value and the time interval corresponding to the predicted peak value, and the vegetation color at each point on the vegetation growth curve is marked;

[0139] a trend prediction subunit configured to respectively acquire a direction vector corresponding to an end of a vegetation growth curve of each region, and obtain a predicted vegetation growth trend direction;

[0140] Meanwhile, target growth data corresponding to target grassland vegetation is acquired based on big data, and a first average growth speed or a second average decay speed of the target vegetation under the current solar term is determined based on the target growth data;

[0141] Based on the vegetation growth curve corresponding to the target grassland vegetation, an actual average growth speed or an actual decay speed of the actual vegetation is determined, and the first average growth speed or the actual average growth speed or the first average decay speed or the actual first average decay speed is compared to obtain an environmental impact growth difference;

[0142] In the vegetation growth curve of the corresponding region, a plurality of target curve segments corresponding to similar vegetation growth trends are intercepted, and a second average growth speed or a second average decay speed corresponding to the plurality of target curve segments is calculated;

[0143] Based on the second average growth speed or the second average decay speed, the change of the vegetation in the corresponding region in the predicted vegetation growth trend direction is determined in combination with the environmental impact growth difference, and a vegetation growth trend prediction result of the corresponding region is obtained.

[0144] In this embodiment, the growth trend prediction unit further comprises:

[0145] A coverage prediction subunit configured to respectively determine vegetation coverage pixel points and non-vegetation coverage pixel points of each region of the grassland based on color features of the current fusion feature image;

[0146] According to the pixel point numbers corresponding to the vegetation coverage pixel points and the non-vegetation coverage pixel points of each region, a current vegetation coverage of each region is respectively obtained;

[0147] Based on the vegetation features corresponding to the current fusion feature image, the number of vegetation in each region of the grassland is determined, and in combination with the vegetation growth trend prediction result of the corresponding region, a vegetation coverage prediction trend corresponding to each region is respectively determined;

[0148] Based on the current vegetation coverage, in combination with the vegetation coverage prediction trend corresponding to each region, a vegetation coverage change of each region is respectively determined, and a vegetation coverage prediction result corresponding to each region of the grassland is obtained.

[0149] The present application can realize accurate quantification of the grassland vegetation coverage condition by determining the vegetation coverage pixel points and non-vegetation coverage pixel points of each region of the grassland based on the color features of the current fusion feature image, and then calculating the current vegetation coverage according to the number of pixel points. Then, the present application determines the vegetation quantity of each region of the grassland based on the vegetation features corresponding to the current fusion feature image, and determines the vegetation coverage prediction trend in combination with the vegetation growth trend prediction result of the corresponding region. The present application not only considers the actual quantity of the current vegetation, but also combines the dynamic trend of the vegetation growth, comprehensively considers the influence of various factors on the change of the vegetation coverage, makes the prediction more scientific and reasonable, and for example, when predicting the future vegetation coverage of a certain region, if the current vegetation quantity of the region is more and the growth trend is good, the prediction result will tend to increase the vegetation coverage; on the contrary, if the vegetation quantity decreases and the growth trend is not good, the vegetation coverage will be predicted to decrease. The multi-factor comprehensive prediction mode can more accurately reflect the future development trend of the grassland vegetation, provide strong support for formulating countermeasures in advance, and finally, based on the current vegetation coverage, in combination with the vegetation coverage prediction trend of each region, the present application determines the vegetation coverage change of each region corresponding to the vegetation coverage prediction result of each region of the grassland, so as to timely understand the dynamic change of the grassland vegetation coverage and discover potential problems in advance. For example, when the prediction result shows that the vegetation coverage of a certain region will continue to decrease, the manager can quickly take measures such as adjusting the grazing strategy, strengthening vegetation protection, etc., to avoid further reduction of the vegetation coverage, so as to effectively maintain the ecological balance of the grassland. Real-time monitoring and prediction of the change of the vegetation coverage are helpful to realize sustainable utilization of the grassland resources and stable development of the ecological system.

[0150] The beneficial effects of the above technical solutions are: the present application predicts the peak point according to the change of vegetation height and the height interval corresponding to the optimal vegetation growth state, takes the predicted peak point as the predicted peak, determines the corresponding time interval in combination with the time interval, provides an important time framework for the construction of the entire growth curve, helps to more intuitively understand the stage characteristics of the vegetation growth period, provides a basis for the curvature transformation of the curve of the natural growth and decay of the vegetation, generates the vegetation growth curve based on the change of the vegetation height, the predicted peak and the time interval thereof, and marks the vegetation color of each point on the curve, wherein the change of the vegetation height reflects the longitudinal dimension of the growth, and the vegetation color can be used to judge the health condition and growth stage of the vegetation, and can comprehensively and intuitively display the comprehensive growth information of the vegetation at different time points, provide rich data visualization expression for the management personnel, and facilitate in-depth analysis of the vegetation growth process; the direction vector at the end of the vegetation growth curve is acquired by the trend prediction subunit to determine the predicted vegetation growth trend direction, then the target growth data (such as the average growth or decay speed under the current solar term) of the target grassland vegetation is compared with the average growth or decay speed of the actual vegetation to obtain the environmental influence growth difference, the influence degree of the current environmental factor on the growth of the vegetation is determined, then the target curve segment of the similar vegetation growth trend is intercepted and the average growth or decay speed thereof is calculated, the change of the vegetation in the predicted trend direction is determined in combination with the environmental influence growth difference, the influence of the grassland environment on the growth of the vegetation is fully considered, the future growth trend of the vegetation can be more comprehensively and accurately predicted, and strong support is provided for the grassland management decision.

[0151] Embodiment 8:

[0152] The present application provides a kind of grassland recovery state dynamic monitoring method based on multiscale fusion, as shown in Figure Figure 5 It includes:

[0153] Step 1: macro monitoring data of grassland is acquired using satellite remote sensing technology, while micro monitoring data of grassland is acquired using unmanned aerial vehicle;

[0154] Step 2: collect macro monitoring data and micro monitoring data in a long time, form time sequence, and carry out fusion analysis on the data in the current time sequence, to determine the actual growth state of grassland and recovery effect;

[0155] Step 3: based on the actual growth of grassland and recovery effect, in combination with actual grassland environment data and historical time sequence, the growth trend of grassland is predicted, and the grassland recovery strategy is adjusted according to the prediction.

[0156] The beneficial effects of the above technical solutions are: the present application combines macroscopic monitoring of satellite remote sensing and microscopic monitoring of unmanned aerial vehicles to perform multi-scale and all-around dynamic monitoring of grassland restoration. Macroscopic monitoring data and microscopic monitoring data over a long period of time are collected to form a time sequence, and the data in the current time sequence are fused and analyzed to determine the actual growth state and restoration effect of the grassland, so that the automatic analysis of the overall monitoring of the grassland restoration is realized, the actual growth state and restoration effect of the grassland are obtained, the actual restoration of the grassland is quickly determined by the management personnel, and finally the grassland growth trend is predicted based on the actual growth and restoration effect of the grassland, combined with actual grassland environment data and historical time sequences, which is beneficial to timely discovering the adverse factors of grassland restoration, providing timely and accurate data support for the management personnel to adjust the restoration strategy, and is helpful to improve the efficiency and quality of grassland restoration and promote the sustainable development of the grassland ecosystem.

[0157] Embodiment 9:

[0158] On the basis of embodiment 8, step 2 comprises:

[0159] The macroscopic monitoring data and the microscopic monitoring data are stored based on the time sequence to generate a time sequence;

[0160] The macroscopic monitoring data and the microscopic monitoring data in the time sequence are respectively subjected to feature extraction to obtain grassland topographic features and vegetation features, and the grassland topographic features and the vegetation features are fused based on the corresponding relationship of the collection area to obtain a current fused feature image;

[0161] The current fused feature image is analyzed to determine the actual growth of the grassland vegetation and the growth state thereof;

[0162] The actual growth of the grassland vegetation is compared with the actual growth corresponding to the previous fused feature image to determine the restoration effect of the grassland.

[0163] The beneficial effects of the above technical solutions are: the present application collects macroscopic monitoring data and microscopic monitoring data of the grassland based on time sequence, and fuses the macroscopic monitoring data and the microscopic monitoring data to improve the accuracy of the grassland monitoring data, which is beneficial to the management personnel to have a more comprehensive and in-depth understanding of the grassland vegetation restoration, and also provides a more comprehensive and accurate data basis for determining the restoration effect of the grassland. Finally, the current fused feature image is analyzed to determine the actual growth of the grassland vegetation and the growth state thereof, and the actual growth of the grassland vegetation is compared with the actual growth corresponding to the previous fused feature image to determine the restoration effect of the grassland, so that the automatic analysis of the grassland restoration is realized, the monitoring workload of the management personnel is reduced, and the feedback efficiency of the grassland restoration monitoring is effectively improved.

[0164] Embodiment 10:

[0165] On the basis of embodiment 9, the actual growth of the grassland vegetation and the growth state are determined by analyzing the current fusion feature image; and the actual growth of the grassland vegetation is compared with the actual growth corresponding to the previous fusion feature image to determine the recovery effect of the grassland, including:

[0166] The current fusion feature image is subjected to feature extraction to obtain the topographic features and vegetation features of the grassland;

[0167] Based on the topographic features of the grassland, the topographic distribution of the grassland is determined, and the vegetation information corresponding to different topographies is determined in combination with the grassland management data;

[0168] Based on the vegetation information and the topographic distribution, the main vegetation species of each region on the grassland are determined, and based on the vegetation species, the standard vegetation growth data corresponding to different vegetation is obtained in the database;

[0169] The vegetation sub-features corresponding to each region of the grassland are determined respectively, the actual growth of the vegetation of each region is determined based on the vegetation sub-features, and the growth state of the vegetation of each region on the grassland is determined according to the actual growth of the vegetation and the standard vegetation growth data corresponding to the vegetation;

[0170] The actual growth of the grassland vegetation is compared with the actual growth corresponding to the previous fusion feature image to obtain the growth difference corresponding to different topographies of the grassland respectively;

[0171] Based on the growth difference, the actual growth state of each region is combined to determine whether the vegetation growth trend of the grassland is good, if so, it is determined that the recovery effect of the current region is good;

[0172] Otherwise, based on the current solar term change of the grassland and the vegetation growth data corresponding to each region, it is determined whether the vegetation growth trend of the current region is not good is in a normal state;

[0173] If not, it is determined that the recovery effect of the current region is not good;

[0174] If so, the vegetation coverage corresponding to the current region is determined based on the fusion feature image corresponding to the best vegetation growth state of the current region, and it is determined whether the coverage of the current region meets the standard in combination with the preset coverage threshold of the topography corresponding to the current region;

[0175] If the coverage of the current region meets the standard, it is determined that the recovery effect of the current region is good;

[0176] Otherwise, it is determined that the recovery effect of the current region is not good.

[0177] The beneficial effects of the above technical scheme are: firstly, the current fused feature image is subjected to feature extraction to obtain the grassland topographic features and vegetation features, then the topographic distribution is determined based on the grassland topographic features, and the vegetation information corresponding to different topographies is determined in combination with the grassland management data, the close relationship between the topography and the vegetation is established, the vegetation type suitable for growth or target pre-planting under different topographic conditions is determined, and accurate basis is provided for subsequent growth analysis of the grassland vegetation, then based on the vegetation information and the topographic distribution, the main vegetation type of each region on the grassland is determined, based on the vegetation type, the standard vegetation growth data corresponding to different vegetation in the database is obtained; the vegetation sub-features corresponding to each region of the grassland are determined respectively, based on the vegetation sub-features, the actual growth conditions of each region are determined respectively, and according to the actual growth conditions of the vegetation and the standard vegetation growth data corresponding to the vegetation, the growth state of the vegetation of each region on the grassland is determined respectively, which can objectively and accurately evaluate the growth conditions of the vegetation of each region, finally, by comparing the actual growth conditions of the vegetation corresponding to the current and the previous fused feature image, the growth difference of different topographies is obtained, which intuitively reflects the growth change trend of the grassland vegetation in a period of time, which helps to discover abnormal conditions in the growth process of the vegetation in time, provides an important basis for evaluating the recovery effect, in combination with the growth difference and the actual growth state to judge the growth trend of the vegetation, considering the influence of solar term change on the growth of the vegetation, the recovery effect is comprehensively judged, the complexity and dynamic nature of the grassland ecosystem are fully considered, the solar term and the vegetation growth data are combined, which can effectively avoid misjudgment, in addition, the recovery effect is finally determined by comparing the current regional vegetation coverage rate with the preset threshold, which further improves the scientificity and accuracy of the evaluation. The present application can clearly understand the recovery effect of different regions of the grassland, which helps to reasonably allocate the grassland management resources and adjust the subsequent grassland recovery strategy.

[0178] Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the claims of the present application and their equivalent technologies, the present application also intends to include these modifications and variations.

Claims

1. A dynamic monitoring system for grassland restoration status based on multi-scale fusion, characterized in that, include: The multi-scale data acquisition module is used to acquire macroscopic monitoring data of grasslands using satellite remote sensing technology, and at the same time to acquire microscopic monitoring data of grasslands using drones. The long-term series data analysis module is used to collect macroscopic and microscopic monitoring data over a long period of time, form a time series, and perform fusion analysis on the data in the current time series to determine the actual growth status and restoration effect of the grassland. The prediction and strategy adjustment module is used to predict grassland growth trends based on the actual growth status and restoration effect of the grassland, combined with actual grassland environmental data and historical time series, and to adjust grassland restoration strategies according to the prediction. The multi-scale data acquisition module includes: The time series generation unit is used to store macro-monitoring data and micro-monitoring data in chronological order and generate time series. The sequence data fusion unit is used to extract features from macroscopic and microscopic monitoring data in the time series to obtain grassland landform features and vegetation features. Based on the correspondence of the collection areas, the grassland landform features and vegetation features are fused to obtain the current fused feature image. The fusion data analysis unit is used to analyze the current fused feature image to determine the actual growth status of grassland vegetation; The actual growth status of grassland vegetation is compared with the actual growth status corresponding to the previous fused feature image to determine the grassland restoration effect. The integrated data analysis unit includes: The grassland feature extraction unit is used to extract features from the current fused feature image to obtain the landform and vegetation features of the grassland. The vegetation determination unit is used to determine the landform distribution of grassland based on grassland landform characteristics, and to determine the vegetation information corresponding to different landforms in combination with grassland management data. The growth analysis unit is used to determine the main vegetation species in each area of ​​the grassland based on the vegetation information and landform distribution, and to obtain standard vegetation growth data corresponding to different vegetation species from the database based on the vegetation species. The vegetation sub-features corresponding to each region of the grassland are determined respectively. Based on the vegetation sub-features, the actual vegetation growth of each region is determined respectively. Based on the actual vegetation growth and the standard vegetation growth data corresponding to the vegetation, the vegetation growth status of each region of the grassland is determined respectively. The effect determination unit is used to compare the actual growth of grassland vegetation with the actual growth corresponding to the previous fused feature image, and obtain the growth differences corresponding to different grassland landforms. Based on the aforementioned growth differences and combined with the actual growth status of each region, it is determined whether the vegetation growth trend of the grassland is good. If so, it is determined that the current region has a good recovery effect. Otherwise, based on the current seasonal changes in the grassland and the corresponding vegetation growth data for each region, determine whether the current unfavorable vegetation growth trend in the region is in a normal state. If not, then the current area is considered to be poorly restored; If so, then based on the fused feature image corresponding to the best vegetation growth status in the current area, determine the vegetation coverage rate of the current area, and combine it with the preset coverage rate threshold of the landform in the current area to determine whether the coverage rate of the current area meets the standard. If the current area coverage rate meets the standard, then the current area is judged to have a good recovery effect; Otherwise, the current area is deemed to have poor recovery performance.

2. The grassland restoration status dynamic monitoring system based on multi-scale fusion according to claim 1, characterized in that, The multi-scale data acquisition module includes: The macro data acquisition unit is used to collect panoramic images and non-panoramic large-area grassland images using satellite remote sensing technology, and generate macro monitoring data. The micro data acquisition unit is used to collect detailed images of local areas of grassland using drones and generate micro monitoring data.

3. The grassland restoration status dynamic monitoring system based on multi-scale fusion according to claim 1, characterized in that, The sequence data fusion unit includes: The macro feature extraction subunit is used to perform edge detection processing on the grassland panoramic image and non-panoramic large-area image in the macro monitoring data, respectively, and to obtain the texture features and color features of the grassland panoramic image and non-panoramic large-area image. The macroscopic feature fusion subunit is used to fuse the texture and color features of the non-panoramic large-area image with the grassland panoramic image based on the acquisition area of ​​the non-panoramic large-area image to obtain a primary panoramic image. The micro-feature extraction subunit is used to extract features from the detailed images corresponding to the micro-monitoring data to obtain detailed features of grassland landforms and vegetation. The macroscopic feature extraction subunit is used to determine the regional correspondence between the detail image and the primary panoramic image based on the texture features of the primary panoramic image and the shooting location of the microscopic monitoring data. Based on the regional correspondence, the grassland landform and vegetation detail features of the detail image are fused with the corresponding regions of the primary panoramic image to obtain a fused feature image.

4. The grassland restoration status dynamic monitoring system based on multi-scale fusion according to claim 1, characterized in that, The prediction and strategy adjustment module includes: The growth trend prediction unit is used to acquire actual environmental data and historical time series of grassland. Based on the vegetation growth of each region of grassland in the historical time series, it generates vegetation growth curves for each region and determines the vegetation growth dynamics of each region. Based on the vegetation growth curve and the current vegetation growth dynamics, the growth trend of vegetation in various grassland areas and the changes in grassland coverage are predicted respectively. The intelligent strategy adjustment unit is used to adjust the current grassland restoration strategy based on the growth trend and changes in grassland coverage, combined with the actual environmental data of the grassland.

5. The grassland restoration status dynamic monitoring system based on multi-scale fusion according to claim 4, characterized in that, The growth trend prediction unit includes: The growth curve generation sub-unit is used to determine vegetation height and vegetation color based on vegetation growth. Based on the changes in vegetation height and referring to the height range corresponding to the optimal vegetation growth state of the corresponding vegetation, it predicts the peak point corresponding to the optimal vegetation growth state of the corresponding area. The peak points are used as the predicted peaks of the vegetation growth curves of the corresponding regions, and the time intervals corresponding to the predicted peaks are determined by combining the time series generation time intervals. Based on the changes in vegetation height, combined with the predicted peak value and its corresponding time interval, a vegetation growth curve for the corresponding region is generated, and the vegetation color is marked at each point corresponding to the vegetation growth curve. The trend prediction subunit is used to obtain the direction vector corresponding to the end of the vegetation growth curve for each region, and to obtain the predicted direction of vegetation growth trend. At the same time, based on big data, target growth data corresponding to the target grassland vegetation is obtained to determine the first average growth rate or the second average decay rate of the target vegetation under the current solar term. Based on the vegetation growth curve corresponding to the target grassland vegetation, the actual average growth rate or actual decay rate of the actual vegetation is determined. The first average growth rate is compared with the actual average growth rate or the first average decay rate to obtain the environmental impact growth difference. In the vegetation growth curve of the corresponding area, multiple target curve segments corresponding to similar vegetation growth trends are extracted, and the second average growth rate or the second average decay rate corresponding to the multiple target curve segments are calculated. Based on the second average growth rate or the second average decay rate, combined with the growth difference caused by environmental influences, the changes in vegetation in the predicted vegetation growth trend direction of the corresponding area are determined, and the vegetation growth trend prediction results of the corresponding area are obtained.

6. A method for dynamic monitoring of grassland restoration status based on multi-scale fusion, characterized in that, include: Step 1: Use satellite remote sensing technology to acquire macroscopic monitoring data of the grassland, and use drones to acquire microscopic monitoring data of the grassland. Step 2: Collect macroscopic and microscopic monitoring data over a long period of time to form a time series, and perform fusion analysis on the data in the current time series to determine the actual growth status and restoration effect of the grassland; Step 3: Based on the actual growth and restoration effect of the grassland, combined with actual grassland environmental data and historical time series, predict the grassland growth trend, and adjust the grassland restoration strategy according to the prediction; Step 2 includes: Based on the timeline order, macroscopic monitoring data and microscopic monitoring data are stored to generate time series; Feature extraction is performed on macroscopic and microscopic monitoring data in the time series to obtain grassland landform features and vegetation features. Based on the correspondence of the collection areas, the grassland landform features and vegetation features are fused to obtain the current fused feature image. Analyze the current fused feature images to determine the actual growth status and condition of the grassland vegetation; The actual growth of grassland vegetation is compared with the actual growth of the previous fused feature image to determine the grassland restoration effect. This process involves analyzing the current fused feature image to determine the actual growth status and condition of the grassland vegetation; and comparing the actual growth status of the grassland vegetation with that of the previous fused feature image to determine the grassland restoration effect, including: Feature extraction is performed on the current fused feature image to obtain the grassland's landform and vegetation features; Based on the characteristics of grassland landforms, the distribution of grassland landforms is determined, and combined with grassland management data, the vegetation information corresponding to different landforms is determined. Based on the vegetation information and landform distribution, the main vegetation types in each area of ​​the grassland are determined, and based on the vegetation types, standard vegetation growth data corresponding to different vegetation types are obtained from the database. The vegetation sub-features corresponding to each region of the grassland are determined respectively. Based on the vegetation sub-features, the actual vegetation growth of each region is determined respectively. Based on the actual vegetation growth and the standard vegetation growth data corresponding to the vegetation, the vegetation growth status of each region of the grassland is determined respectively. By comparing the actual growth of grassland vegetation with the actual growth of the previous fused feature image, the growth differences corresponding to different grassland landforms are obtained. Based on the aforementioned growth differences and combined with the actual growth status of each region, it is determined whether the vegetation growth trend of the grassland is good. If so, it is determined that the current region has a good recovery effect. Otherwise, based on the current seasonal changes in the grassland and the corresponding vegetation growth data for each region, determine whether the current unfavorable vegetation growth trend in the region is in a normal state. If not, then the current area is considered to be poorly restored; If so, then based on the fused feature image corresponding to the best vegetation growth status in the current area, determine the vegetation coverage rate of the current area, and combine it with the preset coverage rate threshold of the landform in the current area to determine whether the coverage rate of the current area meets the standard. If the current area coverage rate meets the standard, then the current area is judged to have a good recovery effect; Otherwise, the current area is deemed to have poor recovery performance.

Images