Methods, devices, electronic equipment and products for monitoring the implementation of territorial spatial planning

Abstract

This invention discloses a method, device, electronic equipment, and product for monitoring the implementation of land spatial planning, relating to the field of image classification technology. The method includes acquiring aerial images of the target area of ​​the approved plan; extracting SIFT features from the aerial images and local color features of the neighborhood of each feature point in the SIFT features; fusing the SIFT features with the local color features of the neighborhood of each feature point in the SIFT features to obtain a first fused feature; extracting global color features from the aerial images; fusing the first fused feature with the global color features to obtain a second fused feature; and using the second fused feature as input to a pre-trained classification model to obtain the land use type of the target area, so as to monitor the implementation of land spatial planning in the target area based on the land use type. This invention can automatically and dynamically monitor the implementation of land spatial planning in real time, saving human resources.

Description

Technical Field

[0001] This invention belongs to the field of image classification technology, specifically relating to a method, device, electronic equipment, and product for monitoring the implementation of land spatial planning. Background Technology

[0002] Territorial spatial planning serves as a guide for national spatial development, aiming to optimize the spatial development pattern of the country and ensure ecological security, food security, and sustainable urban and rural development by scientifically delineating "three zones and three lines" (ecological protection red line, permanent basic farmland, and urban development boundary). However, the effectiveness of the plan's implementation highly depends on strict control over land use. Illegal occupation of arable land, unauthorized development of forest land, or disorderly expansion of urban land will directly threaten resource security and ecological balance. Therefore, dynamic monitoring of the implementation of territorial spatial planning to ensure that land use is carried out according to the plan has become a crucial link in the modernization of the national governance system and governance capacity.

[0003] For monitoring the implementation of national spatial planning, traditional monitoring methods mainly rely on manual monitoring, depending on regular on-site inspections or visual interpretation of remote sensing images. Although this method can directly identify changes in land features, it has significant limitations: on the one hand, manual monitoring is inefficient and has limited coverage, making it difficult to cope with the challenges of a vast and frequently changing national spatial area; on the other hand, manual monitoring requires a large amount of human resources, which can easily lead to monitoring delays or missed reports.

[0004] Therefore, how to provide an effective solution to facilitate dynamic monitoring of the implementation of national spatial planning has become a pressing problem to be solved in existing technologies. Summary of the Invention

[0005] The purpose of this invention is to provide a method, device, electronic equipment, and product for monitoring the implementation of national spatial planning, in order to solve the above-mentioned problems existing in the prior art.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: In a first aspect, the present invention provides a method for monitoring the implementation of territorial spatial planning, comprising: Obtain aerial images of the target area in the approved plan; Extract the SIFT features of the aerial image and the local color features of the neighborhood of each feature point in the SIFT features; The SIFT features are fused with the local color features of the neighborhood of each feature point in the SIFT features to obtain the first fused feature; Extract the global color features of the aerial image; The first fusion feature is fused with the global color feature to obtain the second fusion feature; The second fusion feature is used as input to a pre-trained classification model to obtain the land use type of the target area, so as to monitor the implementation of the land use planning of the target area based on the land use type of the target area; The classification model is trained by using the first sample fusion feature of the sample aerial image and the second sample fusion feature obtained by fusing the sample global color feature of the sample aerial image as the sample input, and the land use type of the sample area corresponding to the sample aerial image as the sample output. The first sample fusion feature of the sample aerial image is obtained by fusing the SIFT feature of the sample aerial image with the local color features of the neighborhood of each feature point in the SIFT feature of the sample aerial image.

[0007] In one possible design, local color features of the neighborhood of each feature point in the SIFT features are extracted, including: Obtain the hue, saturation, and brightness of the neighboring pixels of each feature point; Based on the mapping relationship table, the hue, saturation, and brightness of the neighboring pixels of each feature point are mapped to binary data of a preset number of bits, thereby obtaining the hue mapping value, saturation mapping value, and brightness mapping value of the neighboring pixels of each feature point. The hue mapping value, saturation mapping value, and brightness mapping value of the neighboring pixels of each feature point are combined to obtain the local color features of the neighboring pixels of each feature point.

[0008] In one possible design, the SIFT features are fused with the local color features of the neighborhood of each feature point in the SIFT features to obtain a first fused feature, including: The feature descriptor of each feature point in the SIFT feature is concatenated with the local color features of its neighboring pixels to obtain the first combined feature of each feature point in the SIFT feature. The first combined feature of each feature point in the SIFT feature is compared with the feature words in the feature dictionary to determine the feature word that is closest to the first combined feature of each feature point in the SIFT feature; The feature word that is closest to the first combined feature of any feature point in the SIFT features is taken as the feature word corresponding to the first combined feature of any feature point. The number of first combined features corresponding to each feature word is counted, and the number of first combined features corresponding to each feature word is combined sequentially to obtain the first fused feature; The feature dictionary is obtained by clustering the first sample fusion features of all sample aerial images during the training process of the classification model.

[0009] In one possible design, the number of first combined features corresponding to each feature word is counted, and the number of first combined features corresponding to each feature word is combined sequentially to obtain the first fused feature, including: Count the number of first combination features corresponding to each feature word, and combine the number of first combination features corresponding to each feature word in turn to obtain the second combination features; The second combined feature is normalized to obtain the first fused feature.

[0010] In one possible design, global color features of the aerial image are extracted, including: Obtain the hue, saturation, and brightness of each pixel in the aerial image; Based on pre-divided hue, saturation, and brightness intervals, the number of pixels in the aerial image whose hue falls within each hue interval, whose saturation falls within each saturation interval, and whose brightness falls within each brightness interval are counted. Select the target hue range with the largest number of corresponding pixels, the target saturation range with the largest number of corresponding pixels, and the target brightness range with the largest number of corresponding pixels; The global color features of the aerial image are obtained by combining the interval numbers corresponding to the target hue interval, the target saturation interval, and the target brightness interval.

[0011] In one possible design, the interval numbers corresponding to the target hue interval, the target saturation interval, and the target brightness interval are combined to obtain the global color features of the aerial image, including: The interval number sequence is obtained by combining the interval number corresponding to the target hue interval, the interval number corresponding to the target saturation interval, and the interval number corresponding to the target brightness interval. The interval number sequence is normalized to obtain the global color features of the aerial image.

[0012] In one possible design, before acquiring the hue, saturation, and brightness of each pixel in the aerial image, the method further includes: Determine whether the aerial image is an HSV format image; If the aerial image is not in HSV format, then the aerial image will be converted to HSV format.

[0013] Secondly, the present invention provides a monitoring device for the implementation of land spatial planning, comprising: The acquisition unit is used to acquire aerial images of the target area in the approved plan. The first extraction unit is used to extract the SIFT features of the aerial image and the local color features of the neighborhood of each feature point in the SIFT features; The first fusion unit is used to fuse the SIFT features with the local color features of the neighborhood of each feature point in the SIFT features to obtain the first fused feature; The second extraction unit is used to extract the global color features of the aerial image; The second fusion unit is used to fuse the first fusion feature with the global color feature to obtain the second fusion feature; A classification unit is used to take the second fusion feature as input to a pre-trained classification model to obtain the land use type of the target area, so as to monitor the implementation of the land use planning of the target area based on the land use type of the target area; The classification model is trained by using the first sample fusion feature of the sample aerial image and the second sample fusion feature obtained by fusing the sample global color feature of the sample aerial image as the sample input, and the land use type of the sample area corresponding to the sample aerial image as the sample output. The first sample fusion feature of the sample aerial image is obtained by fusing the SIFT feature of the sample aerial image with the local color features of the neighborhood of each feature point in the SIFT feature of the sample aerial image.

[0014] Thirdly, the present invention provides an electronic device comprising a memory, a processor, and a transceiver connected in sequence, wherein the memory is used to store a computer program, the transceiver is used to send and receive messages, and the processor is used to read the computer program and execute the land spatial planning implementation monitoring method as described in the first aspect or any possible design of the first aspect.

[0015] Fourthly, the present invention provides a computer-readable storage medium storing instructions that, when executed on a computer, perform the land spatial planning implementation monitoring method described in the first aspect or any possible design of the first aspect.

[0016] Fifthly, the present invention provides a computer program product containing instructions that, when executed on a computer, cause the computer to perform the land spatial planning implementation monitoring method as described in the first aspect or any possible design of the first aspect.

[0017] Beneficial effects: This invention discloses a monitoring scheme for the implementation of national spatial planning, which can automatically monitor the implementation of national spatial planning in real time. Specifically, firstly, aerial images of the target area of ​​the approved plan are acquired. SIFT features of the aerial images and local color features of the neighborhood of each feature point in the SIFT features are extracted. The SIFT features are then fused with the local color features of the neighborhood of each feature point in the SIFT features to obtain a first fused feature. This fuses SIFT features including edge, corner, and texture features with the local color features of the image. Next, global color features of the aerial images are extracted. The first fused feature is then fused with the global color features to obtain a second fused feature. This fuses SIFT features including edge, corner, and texture features with the local and global color features of the image, resulting in a more comprehensive extraction of image features. Finally, the second fused feature is used as input to a pre-trained classification model to obtain the land use type of the target area, enabling monitoring of the implementation of national spatial planning in the target area based on the land use type. Thus, by extracting and fusing SIFT features containing features such as edges, corners, and textures, local color features of the image, and global color features of the image, image features can be extracted more comprehensively, improving the feature extraction effect and enriching feature expression. This enables accurate classification of land use types in the target area, achieving dynamic real-time monitoring of the implementation of national land spatial planning. At the same time, it can save human resources, avoid monitoring delays, false alarms, and omissions caused by manual monitoring, and facilitate practical application and promotion. Attached Figure Description

[0018] Figure 1 A flowchart of the land and space planning implementation monitoring method provided in the embodiments of this application; Figure 2 A block diagram of the land and space planning implementation monitoring device provided in the embodiments of this application; Figure 3 This is a block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the present invention will be briefly introduced below in conjunction with the accompanying drawings and descriptions of the embodiments or the prior art. Obviously, the following description of the structure of the accompanying drawings is only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. It should be noted that the description of these embodiments is for the purpose of helping to understand the present invention, but does not constitute a limitation of the present invention.

[0020] It should be understood that although the terms first, second, etc., may be used herein to describe various units, these units should not be limited by these terms. These terms are only used to distinguish one unit from another. For example, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit, without departing from the scope of the exemplary embodiments of the invention.

[0021] It should be understood that the term "and / or" that may appear in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can mean: A exists alone, B exists alone, and A and B exist simultaneously. The term " / and" that may appear in this document describes another relationship between related objects, indicating that two relationships can exist. For example, A / and B can mean: A exists alone, and A and B exist alone. In addition, the character " / " that may appear in this document generally indicates that the related objects before and after it are in an "or" relationship.

[0022] It should be understood that specific details are provided in the following description to facilitate a complete understanding of the exemplary embodiments. However, those skilled in the art will understand that the exemplary embodiments can be implemented without these specific details. For example, the system may be shown in block diagrams to avoid obscuring the example with unnecessary details. In other instances, well-known processes, structures, and techniques may be shown without unnecessary details to avoid obscuring the exemplary embodiments.

[0023] To facilitate dynamic monitoring of the implementation of national spatial planning, this application provides a method, device, electronic equipment, and product for monitoring the implementation of national spatial planning. This method, device, electronic equipment, and product can automatically monitor the implementation of national spatial planning in real time, saving human resources.

[0024] The land spatial planning implementation monitoring method provided in this application can be applied to user terminals or servers. It is understood that the execution entity described does not constitute a limitation on the embodiments of this application.

[0025] The following will provide a detailed description of the land and space planning implementation monitoring method provided in the embodiments of this application.

[0026] like Figure 1 The diagram shown is a flowchart of a land spatial planning implementation monitoring method provided in the first aspect of the present application. The land spatial planning implementation monitoring method may include, but is not limited to, the following steps S101-S106.

[0027] Step S101. Obtain aerial images of the target area of ​​the approved plan.

[0028] In one or more embodiments, aerial images of the approved planned target area can be acquired via satellite or drone.

[0029] Step S102. Extract SIFT features from the aerial image and local color features of the neighborhood of each feature point in the SIFT features.

[0030] In one or more embodiments, extracting local color features of the neighborhood of each feature point in the SIFT features may include, but is not limited to, the following steps S1021-S1023.

[0031] Step S1021. Obtain the hue, saturation, and brightness of the neighboring pixels of each feature point.

[0032] The neighboring pixels of a feature point can be either the 8 neighboring pixels around the feature point or the neighboring pixels with a specified neighborhood radius of the feature point.

[0033] In practice, aerial images may not be in HSV format. Therefore, before obtaining the hue, saturation, and brightness of the neighboring pixels of each feature point, it is necessary to first determine whether the aerial image is in HSV format. If the aerial image is not in HSV format, it can be converted to HSV format first, and then the hue, saturation, and brightness of the neighboring pixels of each feature point can be extracted.

[0034] Step S1022. Based on the mapping relationship table, map the hue, saturation, and brightness of the neighboring pixels of each feature point to binary data of a preset number of bits, and obtain the hue mapping value, saturation mapping value, and brightness mapping value of the neighboring pixels of each feature point.

[0035] The hue range is 0°-360°, the saturation range is 0%-100%, and the brightness range is 0%-100%. If these are directly combined into features after being converted to binary data, the data dimensionality will be too high, making subsequent classification difficult. Therefore, a mapping table can be pre-defined. This table divides the hue, saturation, and brightness ranges into multiple intervals, mapping hue, saturation, and brightness belonging to different intervals to binary data of a preset bit depth.

[0036] For example, for hue, the hue range can be divided into eight intervals: [0°, 45°), [45°, 90°), [90°, 135°), [135°, 180°), [180°, 225°), [225°, 270°), [270°, 315°), and [315°, 360°). These eight intervals can be numbered sequentially as 0, 1, 2, 3, 4, 5, 6, and 7. The hue value of a neighboring pixel can be mapped to the three-digit binary code (000, 001, 010, 011, 100, 101, 110, or 111) corresponding to its interval number.

[0037] Step S1023. Combine the hue mapping value, saturation mapping value and brightness mapping value of the neighboring pixels of each feature point to obtain the local color features of the neighboring pixels of each feature point.

[0038] It is understandable that in some other embodiments, the R (red channel), G (green channel), and B (blue channel) components of the RGB image can be used instead of the mapping values ​​and saturation mapping values ​​for local color feature extraction.

[0039] Step S103. The SIFT features are fused with the local color features of the neighborhood of each feature point in the SIFT features to obtain the first fused feature.

[0040] In one or more embodiments, the SIFT features are fused with the local color features of the neighborhood of each feature point in the SIFT features, which may include, but is not limited to, the following steps S1031-S1034.

[0041] Step S1031. Concatenate the feature descriptor of each feature point in the SIFT feature with the local color features of its neighboring pixels to obtain the first combined feature of each feature point in the SIFT feature.

[0042] Step S1032. Compare the first combined feature of each feature point in the SIFT feature with the feature words in the feature dictionary, and determine the feature word that is closest to the first combined feature of each feature point in the SIFT feature.

[0043] In one or more embodiments, a classification model for land use types is pre-trained. The feature dictionary can be obtained by clustering the first sample fusion features (obtained by fusing the SIFT features of the sample aerial images with the local color features of the neighborhood of each feature point in the SIFT features of the sample aerial images) of all sample aerial images during the training process of the classification model. Here, the feature words can refer to the cluster centers obtained from the clustering.

[0044] After obtaining the first combined feature of each feature point in the SIFT feature, the distance between the first combined feature of each feature point in the SIFT feature and each feature word can be calculated, and the feature word closest to the first combined feature of each feature point in the SIFT feature can be determined.

[0045] Step S1033. Take the feature word that is closest to the first combined feature of any feature point in the SIFT features as the feature word corresponding to the first combined feature of any feature point.

[0046] For example, if the feature word closest to the first combination of features of feature point a is feature word A, then feature word A can be taken as the feature word corresponding to the first combination of features of feature point a.

[0047] Step S1034. Count the number of first combination features corresponding to each feature word, and combine the number of first combination features corresponding to each feature word in sequence to obtain the first fused feature.

[0048] Understandably, when combining the number of first combined features corresponding to each feature word in sequence, the number of first combined features corresponding to each feature word can be converted into binary before being combined to obtain the first fused feature.

[0049] For example, all feature words include feature word A, feature word B, feature word C, feature word D, feature word E, and feature word F. The number of first-combination features corresponding to feature word A is 10, the number of first-combination features corresponding to feature word B is 11, the number of first-combination features corresponding to feature word C is 5, the number of first-combination features corresponding to feature word D is 7, the number of first-combination features corresponding to feature word E is 7, and the number of first-combination features corresponding to feature word F is 9. Then, the number of first-combination features corresponding to feature word A and F can be sequentially converted to 01010, 01011, 00101, 00111, 01000, and 01001. The resulting first fusion feature after sequential combination is 010100101100101001110100001001.

[0050] In one or more embodiments, after counting the number of first combined features corresponding to each feature word and sequentially combining the number of first combined features corresponding to each feature word, the second combined features obtained by sequential combination can be normalized to obtain the first fused feature.

[0051] Step S104. Extract global color features from aerial images.

[0052] In one or more embodiments, extracting the global color features of an aerial image may include, but is not limited to, the following steps S1041-S1044.

[0053] Step S1041. Obtain the hue, saturation, and brightness of each pixel in the aerial image.

[0054] Step S1042. Based on the pre-divided hue intervals, saturation intervals, and brightness intervals, count the number of pixels in the aerial image whose hue is in each hue interval, whose saturation is in each saturation interval, and whose brightness is in each brightness interval.

[0055] The hue, saturation, and brightness ranges can be divided according to the actual situation. The hue, saturation, and brightness ranges can be numbered sequentially.

[0056] Step S1043. Select the target hue range with the largest number of corresponding pixels, the target saturation range with the largest number of corresponding pixels, and the target brightness range with the largest number of corresponding pixels.

[0057] Step S1044. Combine the interval numbers corresponding to the target hue interval, the target saturation interval, and the target brightness interval to obtain the global color features of the aerial image.

[0058] Specifically, the interval numbers corresponding to the target hue interval, the target saturation interval, and the target brightness interval can be combined to obtain an interval number sequence. Then, the interval number sequence is normalized to obtain the global color features of the aerial image.

[0059] Understandably, when combining the interval numbers corresponding to the target hue interval, the target saturation interval, and the target brightness interval, the interval numbers corresponding to the target hue interval, the target saturation interval, and the target brightness interval can all be set to binary numbers, so that the resulting interval number sequence is a binary encoded sequence.

[0060] Understandably, the order of steps S104 and steps S101-S103 is not limited.

[0061] Step S105. Perform feature fusion between the first fusion feature and the global color feature to obtain the second fusion feature.

[0062] In one or more embodiments, the first fusion feature can be directly combined with the global color feature in sequence to obtain the second fusion feature.

[0063] Step S106. Use the second fusion feature as input to the pre-trained classification model to obtain the land use type of the target area, so as to monitor the implementation of the land use planning of the target area based on the land use type of the target area.

[0064] The classification model is trained by using the first sample fusion feature of the sample aerial image and the second sample fusion feature obtained by fusing the sample global color feature of the sample aerial image as the sample input, and the land use type of the sample area corresponding to the sample aerial image as the sample output. The first sample fusion feature of the sample aerial image is obtained by fusing the SIFT feature of the sample aerial image with the local color features of the neighborhood of each feature point in the SIFT feature of the sample aerial image.

[0065] The classification model may be, but is not limited to, a Support Vector Machine (SVM) model or a Convolutional Neural Network (CNN) model. The land use type of the target area may be, but is not limited to, forest land, cultivated land, urban development land, etc.

[0066] The present invention provides a method for monitoring the implementation of land spatial planning. This method involves acquiring aerial images of the target area for approved planning; extracting SIFT features from the aerial images and local color features of the neighborhood of each feature point in the SIFT features; fusing the SIFT features with the local color features of the neighborhood of each feature point in the SIFT features to obtain a first fused feature; extracting global color features from the aerial images; fusing the first fused feature with the global color features to obtain a second fused feature; and using the second fused feature as input to a pre-trained classification model to obtain the land use type of the target area, thereby enabling monitoring of the implementation of land spatial planning in the target area based on the land use type. By extracting and fusing SIFT features containing features such as edges, corners, and textures, local color features of the image, and global color features of the image, a more comprehensive extraction of image features can be achieved, improving the feature extraction effect and enriching feature expression. This allows for accurate classification of land use types in the target area, enabling dynamic real-time monitoring of the implementation of land spatial planning. Simultaneously, it saves human resources, avoids monitoring delays, false alarms, and missed alarms that occur with manual monitoring, and facilitates practical application and promotion.

[0067] Please see Figure 2 The second aspect of this application provides a land spatial planning implementation monitoring device, which includes: The acquisition unit is used to acquire aerial images of the target area in the approved plan. The first extraction unit is used to extract the SIFT features of the aerial image and the local color features of the neighborhood of each feature point in the SIFT features; The first fusion unit is used to fuse the SIFT features with the local color features of the neighborhood of each feature point in the SIFT features to obtain the first fused feature; The second extraction unit is used to extract the global color features of the aerial image; The second fusion unit is used to fuse the first fusion feature with the global color feature to obtain the second fusion feature; A classification unit is used to take the second fusion feature as input to a pre-trained classification model to obtain the land use type of the target area, so as to monitor the implementation of the land use planning of the target area based on the land use type of the target area; The classification model is trained by using the first sample fusion feature of the sample aerial image and the second sample fusion feature obtained by fusing the sample global color feature of the sample aerial image as the sample input, and the land use type of the sample area corresponding to the sample aerial image as the sample output. The first sample fusion feature of the sample aerial image is obtained by fusing the SIFT feature of the sample aerial image with the local color features of the neighborhood of each feature point in the SIFT feature of the sample aerial image.

[0068] The working process, working details and technical effects of the land and space planning implementation monitoring device provided in the second aspect of this embodiment can be found in the first aspect of the embodiment, and will not be repeated here.

[0069] Please see Figure 3 The third aspect of this application provides an electronic device, including a memory, a processor, and a transceiver that are sequentially and communicatively connected, wherein the memory is used to store a computer program, the transceiver is used to send and receive messages, and the processor is used to read the computer program and execute the land spatial planning implementation monitoring method as described in the first aspect of the application.

[0070] Specifically, the memory may include, but is not limited to, random access memory (RAM), read-only memory (ROM), flash memory, first-in-first-out (FIFO) memory, and / or last-in-first-out (FILO) memory, etc.; the processor may not be limited to microprocessors of the STM32F105 series, ARM (Advanced RISC Machines), x86 architecture processors, or processors with integrated NPU (neural-network processing units); the transceiver may be, but is not limited to, WiFi (Wireless Fidelity) wireless transceivers, Bluetooth wireless transceivers, General Packet Radio Service (GPRS) wireless transceivers, ZigBee (a low-power LAN protocol based on the IEEE 802.15.4 standard), 3G transceivers, 4G transceivers, and / or 5G transceivers, etc.

[0071] This fourth aspect of the embodiment provides a computer-readable storage medium storing instructions containing the land spatial planning implementation monitoring method described in the first aspect of the embodiment. Specifically, the computer-readable storage medium stores instructions that, when executed on a computer, perform the land spatial planning implementation monitoring method as described in the first aspect. The computer-readable storage medium refers to a data storage medium, which may include, but is not limited to, floppy disks, optical disks, hard disks, flash memory, USB flash drives, and / or memory sticks. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices.

[0072] The fifth aspect of this embodiment provides a computer program product containing instructions that, when executed on a computer, cause the computer to perform the land spatial planning implementation monitoring method as described in the first aspect of this embodiment, wherein the computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device.

[0073] Finally, it should be noted that the above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for monitoring the implementation of territorial spatial planning, characterized in that, include: Obtain aerial images of the target area in the approved plan; Extract the SIFT features of the aerial image and the local color features of the neighborhood of each feature point in the SIFT features; The SIFT features are fused with the local color features of the neighborhood of each feature point in the SIFT features to obtain the first fused feature; Extract the global color features of the aerial image; The first fusion feature is fused with the global color feature to obtain the second fusion feature; The second fusion feature is used as input to a pre-trained classification model to obtain the land use type of the target area, so as to monitor the implementation of the land use planning of the target area based on the land use type of the target area; The classification model is trained by using the first sample fusion feature of the sample aerial image and the second sample fusion feature obtained by fusing the sample global color feature of the sample aerial image as the sample input, and the land use type of the sample area corresponding to the sample aerial image as the sample output. The first sample fusion feature of the sample aerial image is obtained by fusing the SIFT feature of the sample aerial image with the local color features of the neighborhood of each feature point in the SIFT feature of the sample aerial image.

2. The method for monitoring the implementation of territorial spatial planning according to claim 1, characterized in that, Extracting local color features from the neighborhood of each feature point in the SIFT features, including: Obtain the hue, saturation, and brightness of the neighboring pixels of each feature point; Based on the mapping relationship table, the hue, saturation, and brightness of the neighboring pixels of each feature point are mapped to binary data of a preset number of bits, thereby obtaining the hue mapping value, saturation mapping value, and brightness mapping value of the neighboring pixels of each feature point. The hue mapping value, saturation mapping value, and brightness mapping value of the neighboring pixels of each feature point are combined to obtain the local color features of the neighboring pixels of each feature point.

3. The method for monitoring the implementation of territorial spatial planning according to claim 1, characterized in that, The SIFT features are fused with the local color features of the neighborhood of each feature point in the SIFT features to obtain the first fused feature, including: The feature descriptor of each feature point in the SIFT feature is concatenated with the local color features of its neighboring pixels to obtain the first combined feature of each feature point in the SIFT feature. The first combined feature of each feature point in the SIFT feature is compared with the feature words in the feature dictionary to determine the feature word that is closest to the first combined feature of each feature point in the SIFT feature; The feature word that is closest to the first combined feature of any feature point in the SIFT features is taken as the feature word corresponding to the first combined feature of any feature point. The number of first combined features corresponding to each feature word is counted, and the number of first combined features corresponding to each feature word is combined sequentially to obtain the first fused feature; The feature dictionary is obtained by clustering the first sample fusion features of all sample aerial images during the training process of the classification model.

4. The method for monitoring the implementation of territorial spatial planning according to claim 3, characterized in that, The number of first combined features corresponding to each feature word is counted, and the number of first combined features corresponding to each feature word is combined sequentially to obtain the first fused feature, including: Count the number of first combination features corresponding to each feature word, and combine the number of first combination features corresponding to each feature word in turn to obtain the second combination features; The second combined feature is normalized to obtain the first fused feature.

5. The method for monitoring the implementation of territorial spatial planning according to claim 1, characterized in that, Extracting the global color features of the aerial image, including: Obtain the hue, saturation, and brightness of each pixel in the aerial image; Based on pre-divided hue, saturation, and brightness intervals, the number of pixels in the aerial image whose hue falls within each hue interval, whose saturation falls within each saturation interval, and whose brightness falls within each brightness interval are counted. Select the target hue range with the largest number of corresponding pixels, the target saturation range with the largest number of corresponding pixels, and the target brightness range with the largest number of corresponding pixels; The global color features of the aerial image are obtained by combining the interval numbers corresponding to the target hue interval, the target saturation interval, and the target brightness interval.

6. The method for monitoring the implementation of territorial spatial planning according to claim 5, characterized in that, The global color features of the aerial image are obtained by combining the interval numbers corresponding to the target hue interval, the target saturation interval, and the target brightness interval, including: The interval number sequence is obtained by combining the interval number corresponding to the target hue interval, the interval number corresponding to the target saturation interval, and the interval number corresponding to the target brightness interval. The interval number sequence is normalized to obtain the global color features of the aerial image.

7. The method for monitoring the implementation of territorial spatial planning according to claim 5, characterized in that, Before acquiring the hue, saturation, and brightness of each pixel in the aerial image, the method further includes: Determine whether the aerial image is an HSV format image; If the aerial image is not in HSV format, then the aerial image will be converted to HSV format.

8. A monitoring device for the implementation of national land spatial planning, characterized in that, include: The acquisition unit is used to acquire aerial images of the target area in the approved plan. The first extraction unit is used to extract the SIFT features of the aerial image and the local color features of the neighborhood of each feature point in the SIFT features; The first fusion unit is used to fuse the SIFT features with the local color features of the neighborhood of each feature point in the SIFT features to obtain the first fused feature; The second extraction unit is used to extract the global color features of the aerial image; The second fusion unit is used to fuse the first fusion feature with the global color feature to obtain the second fusion feature; A classification unit is used to take the second fusion feature as input to a pre-trained classification model to obtain the land use type of the target area, so as to monitor the implementation of the land use planning of the target area based on the land use type of the target area; The classification model is trained by using the first sample fusion feature of the sample aerial image and the second sample fusion feature obtained by fusing the sample global color feature of the sample aerial image as the sample input, and the land use type of the sample area corresponding to the sample aerial image as the sample output. The first sample fusion feature of the sample aerial image is obtained by fusing the SIFT feature of the sample aerial image with the local color features of the neighborhood of each feature point in the SIFT feature of the sample aerial image.

9. An electronic device, characterized in that, The device includes a memory, a processor, and a transceiver that are sequentially and communicatively connected. The memory is used to store computer programs, the transceiver is used to send and receive messages, and the processor is used to read the computer programs and execute the land spatial planning implementation monitoring method as described in any one of claims 1 to 7.

10. A computer program product, comprising a computer program or instructions, characterized in that, When the computer program or the instructions are executed by the computer, they implement the land spatial planning implementation monitoring method as described in any one of claims 1 to 7.

Images