A land space planning data collection and analysis system and method

Abstract

The invention relates to the field of data analysis, and discloses a national land space planning data acquisition and analysis system and a method thereof, the method comprising the following steps: firstly, according to the use nature of the land in each region, image data of a to-be-detected area is collected regularly or on demand; then, the image data of the cultivated land area in each region is compared with the color of the surrounding natural vegetation, and after the comparison result is output, it is judged whether the current area is an abnormal cultivated land area; then, historical image data and a crop growth cycle knowledge base are combined to analyze the change of the abnormal cultivated land area over time to obtain an evaluation coefficient, the evaluation coefficient is compared with a preset evaluation threshold interval, and it is judged whether there is a disguised cultivated land according to the comparison result; if it is judged that there is a disguised cultivated land, an alarm is automatically triggered, and the location information of the abnormal cultivated land area is provided; finally, different review methods are executed according to the location and evaluation coefficient of each abnormal cultivated land area.

Description

Technical Field

[0001] The present invention relates to the field of data analysis, and in particular to a national land space planning data collection and analysis system and method. Background Art

[0002] The national land space planning is an arrangement in space and time for the development and protection of the land space in a certain area. It is a guide for national space development, a spatial blueprint for sustainable development, and the basic basis for all kinds of development, protection and construction activities.

[0003] At present, geographic information system (GIS) technology has been widely used in the field of national land space planning. GIS technology can store, manage, analyze and display geospatial data, providing strong technical support for national land space planning. However, traditional GIS data collection methods mainly rely on maps and remote sensing technology, and there are problems such as low collection efficiency and insufficient data accuracy. For example, some regions use ground camouflage to cope with land data collection, and the errors in actual land data collection have led to errors in regional use, which seriously restricted the actual national land space planning process in the later stage. Summary of the invention

[0004] The purpose of the present invention is to provide a national land space planning data collection and analysis system and method to solve the above technical problems.

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] A method for collecting and analyzing national land space planning data comprises the following steps:

[0007] S1. Use drone aerial photography and satellite remote sensing to collect image data of the area to be inspected regularly or as needed according to the nature of land use in each region;

[0008] S2, perform color analysis on the collected image data, input the image data of the cultivated land area in each region into the pre-trained convolutional neural network model to compare it with the color of the surrounding natural vegetation, and output the comparison results;

[0009] S3. If the comparison result is true, the current area is judged to be a real cultivated land area; if the comparison result is false, the current area is judged to be an abnormal cultivated land area;

[0010] S4. Combining historical image data and crop growth cycle knowledge base, analyzing the changes of abnormal cultivated land areas over time to obtain an evaluation coefficient, comparing the evaluation coefficient with a preset evaluation threshold interval, and judging whether there is disguised cultivated land based on the comparison result;

[0011] S5. If it is determined that there is disguised farmland, an alarm is automatically triggered and the location information of the abnormal farmland area is provided; different review methods are performed according to the location and evaluation coefficient of each abnormal farmland area.

[0012] In a further solution, the expression of the evaluation coefficient is:

[0013] ;in,

[0014] ;

[0015] In the formula, is the evaluation coefficient of the current abnormal cultivated land area, is the area change of the current abnormal cultivated land area within a monitoring period, is the vegetation index change of the current abnormal cultivated land area within a monitoring period, , , are the maximum values ​​of color contrast change, area change, and vegetation index change, respectively.

[0016] , , is the exponential coefficient, and , is the reference coefficient, Any abnormal cultivated land area;

[0017] is the color contrast change between the current abnormal cultivated land area and the surrounding natural vegetation within a monitoring period, is the intensity of the current abnormal cultivated land area and the surrounding natural vegetation in the red channel, is the intensity of the current abnormal cultivated land area in the green channel, is the intensity of the current abnormal cultivated land area in the blue channel, , , is the preset scale factor, is the total number of sampling times, For the Subsampling, , , They are the intensities of the surrounding natural vegetation in the red channel, green channel, and blue channel, respectively.

[0018] In a further embodiment, the evaluation coefficients in S4 are The preset evaluation threshold interval The comparison process is:

[0019] like , it is judged that there is no disguised farmland in the current abnormal farmland area;

[0020] like , it is judged that there is disguised farmland in the current abnormal farmland area.

[0021] A further solution is to obtain the area change of the current abnormal cultivated land area within a monitoring period. The method is:

[0022] Fitting to obtain the curve of the area change of the current abnormal cultivated land area over time within a monitoring period ;

[0023] The end time is the time when the area is judged as an abnormal cultivated land area. , trace back to the preset period to get the start time , As a monitoring cycle;

[0024] By formula: Calculate the area change of the current abnormal cultivated land area ;

[0025] In the formula, This is the curve of the area change of the same type of cultivated land over time obtained based on big data; , is the weight factor.

[0026] A further solution is to obtain the reference coefficient The method is:

[0027] Mark the abnormal cultivated land areas in each region and obtain the area of ​​the marked abnormal cultivated land areas And the current total area ,

[0028] Substituting into the formula: Calculate the reference coefficient ;

[0029] In the formula, is the number of abnormal cultivated land areas in the current region, Indicates The area of ​​abnormally cultivated land areas.

[0030] A further solution is to perform different review methods according to the location and evaluation coefficient of each abnormal cultivated land area:

[0031] Each region is sorted in descending order according to the number of fake cultivated lands it contains and is used as the first priority sequence. The top 50% of the regions are classified as key review regions, and the bottom 50% of the regions are classified as secondary key review regions.

[0032] The secondary key review areas will be conducted in sequence by manual on-site inspection, while the key areas will be conducted in sequence by drone inspection.

[0033] A further solution is to cover each abnormal cultivated land area in each region with red and mark the midpoint of each abnormal cultivated land area;

[0034] Get the distance between the midpoints of adjacent abnormal cultivated land areas and number them in sequence , , ,..., ;

[0035] Set the spacing value to less than The two abnormal cultivated land areas are regarded as a cluster;

[0036] By formula Calculate the priority coefficient for each region ;

[0037] Arrange in descending order of priority coefficient as the review order;

[0038] In the formula, is the rank in the first priority sequence, is the number of clusters, , It is a preset fixed coefficient.

[0039] A national land space planning data collection and analysis system, comprising:

[0040] The land data acquisition module is used to collect image data of the area to be inspected regularly or on demand according to the nature of land use in each region;

[0041] The land color analysis module performs color analysis on the collected image data, inputs the image data of the cultivated land area in each region into the pre-trained convolutional neural network model, compares it with the color of the surrounding natural vegetation, and outputs the comparison results;

[0042] If the comparison result is true, the current area is judged to be a real cultivated land area; if the comparison result is false, the current area is judged to be an abnormal cultivated land area;

[0043] The land cycle monitoring module is used to combine historical image data and the crop growth cycle knowledge base to analyze the changes in the morphology of abnormal cultivated land areas over time to obtain an evaluation coefficient, compare the evaluation coefficient with a preset evaluation threshold interval, and determine whether there is disguised cultivated land based on the comparison results;

[0044] The alarm and verification module automatically triggers an alarm if there is any disguised farmland, and provides the location information of the abnormal farmland area. Different verification methods are performed according to the location and evaluation coefficient of each abnormal farmland area.

[0045] Database and knowledge base, used to store color and morphological characteristic data of normal cultivated land and different crop growth cycles.

[0046] Beneficial effects of the present invention:

[0047] The present invention performs preliminary screening by comparing the colors of the cultivated land area with the surrounding vegetation area, so as to quickly determine the abnormal cultivated land area with abnormalities, narrow the scope for subsequent further analysis, reduce computing resources, and then perform comprehensive analysis through the obtained multiple parameters to avoid the situation where the cultivated land is determined to be cultivated land by rough comparison of image data. It not only relies on the surface image of the cultivated land, but also comprehensively establishes an evaluation model for the abnormal cultivated land area based on the changes of the cultivated land area over time, so as to achieve the purpose of accurately judging whether the abnormal cultivated land area is disguised cultivated land. Once it is judged to be disguised cultivated land, an alarm is triggered to focus on the investigation of the abnormal cultivated land area, thereby reducing the probability of disguised cultivated land and providing accurate conditions for subsequent national land space planning. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The present invention will be further described below in conjunction with the accompanying drawings.

[0049] Figure 1 It is a diagram of the method steps of the present invention;

[0050] Figure 2 It is a system structure block diagram of the present invention. DETAILED DESCRIPTION

[0051] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0052] See also Figure 1-Figure 2 As shown, the present invention is a method for collecting and analyzing land space planning data, comprising the following steps:

[0053] S1. Use drone aerial photography and satellite remote sensing to collect image data of the area to be inspected regularly or as needed according to the nature of land use in each region;

[0054] S2, perform color analysis on the collected image data, input the image data of the cultivated land area in each region into the pre-trained convolutional neural network model to compare it with the color of the surrounding natural vegetation, and output the comparison results;

[0055] S3. If the comparison result is true, the current area is judged to be a real cultivated land area; if the comparison result is false, the current area is judged to be an abnormal cultivated land area;

[0056] S4. Combining historical image data and crop growth cycle knowledge base, analyzing the changes of abnormal cultivated land areas over time to obtain an evaluation coefficient, comparing the evaluation coefficient with a preset evaluation threshold interval, and judging whether there is disguised cultivated land based on the comparison result;

[0057] S5. If it is determined that there is disguised farmland, an alarm is automatically triggered and the location information of the abnormal farmland area is provided; different review methods are performed according to the location and evaluation coefficient of each abnormal farmland area.

[0058] In order to solve the problem of camouflaged cultivated land raised in the background technology, the present invention first performs preliminary screening by comparing the colors of the cultivated land area with the surrounding vegetation area, so as to quickly determine the abnormal cultivated land area with abnormalities, narrow the scope for subsequent further analysis, and reduce computing resources. Subsequently, a comprehensive analysis is performed through the obtained multiple parameters to avoid the situation where the cultivated land is determined to be cultivated land by rough comparison of image data. It not only relies on the surface image of the cultivated land, but also comprehensively establishes an evaluation model for the abnormal cultivated land area based on the changes of the cultivated land area over time, so as to achieve the purpose of accurately judging whether the abnormal cultivated land area is camouflaged cultivated land. Once it is judged to be camouflaged cultivated land, an alarm is triggered to conduct a key investigation on the abnormal cultivated land area, thereby reducing the probability of camouflaged cultivated land and providing accurate conditions for subsequent national land space planning.

[0059] The expression of the evaluation coefficient is:

[0060] ;in,

[0061] ;

[0062] In the formula, is the evaluation coefficient of the current abnormal cultivated land area, is the area change of the current abnormal cultivated land area within a monitoring period, is the vegetation index change of the current abnormal cultivated land area within a monitoring period, , , are the maximum values ​​of color contrast change, area change, and vegetation index change, respectively.

[0063] , , is the exponential coefficient, and , is the reference coefficient, Any abnormal cultivated land area;

[0064] is the color contrast change between the current abnormal cultivated land area and the surrounding natural vegetation within a monitoring period, is the intensity of the current abnormal cultivated land area and the surrounding natural vegetation in the red channel, is the intensity of the current abnormal cultivated land area in the green channel, is the intensity of the current abnormal cultivated land area in the blue channel, , , is the preset scale factor, is the total number of sampling times, For the Subsampling, , , They are the intensities of the surrounding natural vegetation in the red channel, green channel, and blue channel, respectively.

[0065] In the field of image processing, the red channel, green channel and blue channel are the three basic elements that constitute the color of an image. They represent the intensity of the three colors red, green and blue in the image respectively. These three channels together determine the color of each pixel in the image.

[0066] 1. Red channel;

[0067] The red channel represents the intensity of the red component in the image. When evaluating the color contrast between abnormal cultivated land areas and surrounding natural vegetation, changes in the red channel can reflect the difference in red components between the two. If the vegetation growth conditions in abnormal cultivated land areas are different from those in natural areas due to human intervention (such as fertilization, irrigation, etc.), then this difference will be reflected in the red channel. Specifically, if the vegetation in abnormal cultivated land areas is more lush, its red component may be more prominent, resulting in a relatively high value for the red channel; conversely, if the vegetation grows poorly, the red component may be weakened, and the value of the red channel will be relatively low.

[0068] 2. Green channel;

[0069] The green channel represents the intensity of the green component in the image. Green is one of the important indicators of vegetation growth, so the green channel is particularly important when evaluating the color contrast between abnormal cultivated land areas and natural vegetation. By comparing the values ​​of the two on the green channel, the growth and health of the vegetation can be judged. If the vegetation in the abnormal cultivated land area is growing vigorously, the value of the green channel may be high; if the vegetation is damaged or growing poorly, the value of the green channel may be low.

[0070] 3. Blue channel;

[0071] The blue channel represents the intensity of the blue component in the image. Changes in the blue channel may reflect the impact of environmental factors, such as soil moisture and light conditions, on vegetation growth. In addition, the blue channel can also be used to assist in determining whether there is water or other non-vegetation cover in abnormal cultivated land areas.

[0072] In summary, the red channel, green channel, and blue channel each play different roles in evaluating the color contrast between abnormal cultivated land areas and the surrounding natural vegetation. Together, they provide important information about the growth status and health of vegetation, as well as the impact of environmental factors on vegetation. In practical applications, the true situation of abnormal cultivated land areas can be more accurately assessed by comprehensively analyzing the values ​​of these three channels.

[0073] The S4 will evaluate the coefficient The preset evaluation threshold interval The comparison process is:

[0074] like , it is judged that there is no disguised farmland in the current abnormal farmland area;

[0075] like , it is judged that there is disguised farmland in the current abnormal farmland area.

[0076] The present invention provides a method for obtaining the evaluation of each abnormal cultivated land area. Specifically, the method obtains the color contrast change, area change and vegetation index change between the current abnormal cultivated land area and the surrounding natural vegetation within a monitoring period of each abnormal cultivated land area, and then integrates the above multiple parameter items to establish an evaluation model expression as follows: Obviously, if it is a normal cultivated land area, the plants in the growing state in a monitoring period will inevitably show different color contrast changes. The cultivated land area will also change to a certain extent. Because in the seedling stage, there will be enough space for seedlings to grow on the cultivated land. As the vegetation grows gradually, it will cover a larger area. If it is camouflaged cultivated land, this kind of change will obviously not occur. Therefore, the closer the ratio of the color contrast change and the cultivated land area change to their respective maximum values ​​is, the more normal the abnormal cultivated land area is. Once it exceeds the maximum value, it means that the abnormal cultivated land area does not conform to the growth law of plants, so the possibility of camouflage is greater.

[0077] Therefore, by comparing the evaluation coefficient with the evaluation threshold interval, it is obvious that the abnormal cultivated land area falling into the evaluation threshold area is more likely to be normal cultivated land, while the abnormal cultivated land area that does not fall into the evaluation threshold interval is more likely to be disguised cultivated land. The vegetation index is a simple, effective and empirical measure of the surface vegetation condition. The change in the vegetation index can be obtained based on the existing technology. It is a parameter that can be directly obtained by those skilled in the art and will not be described in detail here. The larger the vegetation index value, the better the vegetation growth state and the higher the vegetation coverage. The smaller the change in the vegetation index value, the more stable the vegetation growth state is. Obviously, real plants will change during the growth process, so the probability of disguised cultivated land is greater.

[0078] At the same time, the calculation of different color contrast changes is given The specific formula is ,Through the above formula, the contrast change between the abnormal cultivated land area and the surrounding natural vegetation in RGB is calculated in the form of standard deviation, which is more accurate, so as to reflect the actual contrast between the abnormal cultivated land area and the surrounding natural vegetation, improve the accuracy of the overall assessment of the abnormal cultivated land area, and reduce the probability of misjudging disguised cultivated land.

[0079] Get the area change of the current abnormal cultivated land area within a monitoring period The method is:

[0080] Fitting to obtain the curve of the area change of the current abnormal cultivated land area over time within a monitoring period ;

[0081] The end time is the time when the area is judged as an abnormal cultivated land area. , trace back to the preset period to get the start time , As a monitoring cycle;

[0082] By formula: Calculate the area change of the current abnormal cultivated land area ;

[0083] In the formula, This is the curve of the area change of the same type of cultivated land over time obtained based on big data; , is the weight factor.

[0084] In this embodiment, a method for obtaining area change is provided, in which a curve of area change over time within a monitoring period is obtained by fitting. The cumulative change of area in a monitoring period is obtained by integration. Obviously, the larger the cumulative change, the better the vegetation growth state of the abnormal cultivated land area. Similarly, the greater the impact on the evaluation coefficient, the higher the probability that the abnormal cultivated land area is normal cultivated land. However, when calculating the cumulative change, in order to reduce the difference in the change state of vegetation in the abnormal cultivated land area and the same type of vegetation, there will be a situation of counterfeiting crops through other vegetation. Therefore, Comparing with the reference curve, it is obvious that when the two change in equal proportion, The closer it is to 1, the more consistent the vegetation in the abnormal cultivated land area is with the changes in the same type of crops. The smaller the overall area change The larger the index is, the higher the exponential coefficient can be determined based on historical data analysis. .

[0085] Get the reference coefficient The method is:

[0086] Mark the abnormal cultivated land areas in each region and obtain the area of ​​the marked abnormal cultivated land areas And the current total area ,

[0087] Substituting into the formula: Calculate the reference coefficient ;

[0088] In the formula, is the number of abnormal cultivated land areas in the current region, Indicates The area of ​​abnormally cultivated land areas.

[0089] The present invention obtains the ratio of abnormal cultivated land area in each region to the total area of ​​the current region as the reference coefficient, and finally expresses it in the form of Indicates the reference coefficient , obviously The larger the The smaller it is, the overall abnormal situation in the current area is reacted to through the above method. Obviously, if the abnormal proportion of the current area is higher, it means that the possibility of disguise in the current area is greater. Therefore, the overall evaluation coefficient is dynamically affected and adjusted to improve the accuracy of the evaluation coefficient.

[0090] The specific process of implementing different review methods according to the location and evaluation coefficient of each abnormal cultivated land area is as follows:

[0091] Each region is sorted in descending order according to the number of fake cultivated lands it contains and is used as the first priority sequence. The top 50% of the regions are classified as key review regions, and the bottom 50% of the regions are classified as secondary key review regions.

[0092] The secondary key review areas will be conducted in sequence by manual on-site inspection, while the key areas will be conducted in sequence by drone inspection.

[0093] Cover each abnormal cultivated land area in each region with red and mark the midpoint of each abnormal cultivated land area;

[0094] Get the distance between the midpoints of adjacent abnormal cultivated land areas and number them in sequence , , ,..., ;

[0095] Set the spacing value to less than The two abnormal cultivated land areas are regarded as a cluster;

[0096] By formula Calculate the priority coefficient for each region ;

[0097] Arrange in descending order of priority coefficient as the review order;

[0098] In the formula, is the rank in the first priority sequence, is the number of clusters, , It is a preset fixed coefficient.

[0099] In the present invention, through the above-mentioned sorting method, manual review is carried out for areas with a small number of disguised farmland, while drone review is used for areas with a large number of disguised farmland, thereby improving the efficiency of the review of the disguised farmland in each area. At the same time, the concentration of the disguised farmland in each area is taken into account. If there are more clusters, the disguised farmland in the area is more concentrated, and it is easier to review. Therefore, priority is given to review to improve the overall review efficiency, and the actual situation of the disguised farmland in each area is confirmed as quickly as possible, reserving sufficient time for subsequent processing.

[0100] A national land space planning data collection and analysis system, comprising:

[0101] The land data acquisition module is used to collect image data of the area to be inspected regularly or on demand according to the nature of land use in each region;

[0102] The land color analysis module performs color analysis on the collected image data, inputs the image data of the cultivated land area in each region into the pre-trained convolutional neural network model, compares it with the color of the surrounding natural vegetation, and outputs the comparison results;

[0103] If the comparison result is true, the current area is judged to be a real cultivated land area; if the comparison result is false, the current area is judged to be an abnormal cultivated land area;

[0104] The land cycle monitoring module is used to combine historical image data and the crop growth cycle knowledge base to analyze the changes in the morphology of abnormal cultivated land areas over time to obtain an evaluation coefficient, compare the evaluation coefficient with a preset evaluation threshold interval, and determine whether there is disguised cultivated land based on the comparison results;

[0105] The alarm and verification module automatically triggers an alarm if there is any disguised farmland, and provides the location information of the abnormal farmland area. Different verification methods are performed according to the location and evaluation coefficient of each abnormal farmland area.

[0106] Database and knowledge base, used to store color and morphological characteristic data of normal cultivated land and different crop growth cycles.

[0107] It should be noted that the calculation formulas and various parameters involved in the calculation in the present invention have been dimensionally processed in advance, and the process of dimensionless processing is well known in the industry and will not be described here.

[0108] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A method for collecting and analyzing land space planning data, characterized in that: The steps include: S1. Use drone aerial photography and satellite remote sensing to collect image data of the area to be inspected regularly or as needed according to the nature of land use in each region; S2, perform color analysis on the collected image data, input the image data of the cultivated land area in each region into the pre-trained convolutional neural network model to compare it with the color of the surrounding natural vegetation, and output the comparison results; S3. If the comparison result is true, the current area is judged to be a real cultivated land area; If the comparison result is false, the current area is judged to be an abnormal cultivated land area; S4. Combining historical image data and crop growth cycle knowledge base, analyzing the changes of abnormal cultivated land areas over time to obtain an evaluation coefficient, comparing the evaluation coefficient with a preset evaluation threshold interval, and judging whether there is disguised cultivated land based on the comparison result; S5. If it is determined that there is disguised farmland, an alarm is automatically triggered and the location information of the abnormal farmland area is provided; different review methods are performed according to the location and evaluation coefficient of each abnormal farmland area; The expression of the evaluation coefficient is: ;in, ； In the formula, is the evaluation coefficient of the current abnormal cultivated land area, is the area change of the current abnormal cultivated land area within a monitoring period, is the vegetation index change of the current abnormal cultivated land area within a monitoring period, , , are the maximum values ​​of color contrast change, area change, and vegetation index change, respectively. , , is the exponential coefficient, and , is the reference coefficient, Any abnormal cultivated land area; is the color contrast change between the current abnormal cultivated land area and the surrounding natural vegetation within a monitoring period, is the intensity of the current abnormal cultivated land area and the surrounding natural vegetation in the red channel, is the intensity of the current abnormal cultivated land area in the green channel, is the intensity of the current abnormal cultivated land area in the blue channel, , , is the preset scale factor, is the total number of sampling times, For the Subsampling, , , are the intensities of the surrounding natural vegetation in the red channel, green channel, and blue channel respectively; Get the reference coefficient The method is: Mark the abnormal cultivated land areas in each region and obtain the area of ​​the marked abnormal cultivated land areas And the current total area , Substituting into the formula: Calculate the reference coefficient ; In the formula, is the number of abnormal cultivated land areas in the current region, Indicates The area of ​​abnormally cultivated land areas.

2. The method for collecting and analyzing land space planning data according to claim 1 is characterized in that: The S4 will evaluate the coefficient The preset evaluation threshold interval The comparison process is: like , it is judged that there is no disguised farmland in the current abnormal farmland area; like , it is judged that there is disguised farmland in the current abnormal farmland area.

3. The method for collecting and analyzing land space planning data according to claim 2 is characterized in that: Get the area change of the current abnormal cultivated land area within a monitoring period The method is: Fitting to obtain the curve of the area change of the current abnormal cultivated land area over time within a monitoring period ; The end time is the time when the area is judged as an abnormal cultivated land area. , trace back to the preset period to get the start time , As a monitoring cycle; By formula: Calculate the area change of the current abnormal cultivated land area ; In the formula, This is the curve of the area change of the same type of cultivated land over time obtained based on big data; , is the weight factor.

4. The method for collecting and analyzing land space planning data according to claim 1, characterized in that: The specific process of implementing different review methods according to the location and evaluation coefficient of each abnormal cultivated land area is as follows: Each region is sorted in descending order according to the number of fake cultivated lands it contains and is used as the first priority sequence. The top 50% of the regions are classified as key review regions, and the bottom 50% of the regions are classified as secondary key review regions. The secondary key review areas will be conducted in sequence by manual on-site inspection, while the key areas will be conducted in sequence by drone inspection.

5. The method for collecting and analyzing land space planning data according to claim 4 is characterized in that: Cover each abnormal cultivated land area in each region with red and mark the midpoint of each abnormal cultivated land area; Get the distance between the midpoints of adjacent abnormal cultivated land areas and number them in sequence , , ,..., ; Set the spacing value to less than The two abnormal cultivated land areas are regarded as a cluster; By formula Calculate the priority coefficient for each region ; Arrange in descending order of priority coefficient as the review order; In the formula, is the rank in the first priority sequence, is the number of clusters, , It is a preset fixed coefficient.

6. A national land space planning data collection and analysis system, characterized in that: The system is used to execute the land space planning data collection and analysis method described in any one of claims 1 to 5, including: The land data collection module is used to collect image data of the area to be inspected regularly or on demand according to the nature of land use in each region; The land color analysis module performs color analysis on the collected image data, inputs the image data of the cultivated land area in each region into the pre-trained convolutional neural network model, compares it with the color of the surrounding natural vegetation, and outputs the comparison results; If the comparison result is true, the current area is judged to be a real cultivated land area; if the comparison result is false, the current area is judged to be an abnormal cultivated land area; The land cycle monitoring module is used to combine historical image data and the crop growth cycle knowledge base to analyze the changes in the morphology of abnormal cultivated land areas over time to obtain an evaluation coefficient, compare the evaluation coefficient with a preset evaluation threshold interval, and determine whether there is disguised cultivated land based on the comparison results; The alarm and verification module automatically triggers an alarm if there is any disguised farmland, and provides the location information of the abnormal farmland area. Different verification methods are performed according to the location and evaluation coefficient of each abnormal farmland area. Database and knowledge base, used to store color and morphological characteristic data of normal cultivated land and different crop growth cycles.

Images