Land use change detection method and system based on remote sensing image

Abstract

This invention discloses a land use change detection method and system based on remote sensing imagery, belonging to the field of remote sensing imagery technology. The method involves attaching label attributes, including time label attributes and ground area label attributes, to the acquired multi-temporal remote sensing images. Based on these label attributes, image feature markers are added to the multi-temporal remote sensing images. Ground area label attributes are selected as the image flow guide, and the image features are sorted and organized to obtain image flow clusters. Based on the image features, feature states are generated and characterized by feature capture functions. Based on the image flow clusters and feature capture functions, land use change state values ​​are analyzed and calculated. Based on the land use change state values, the correlation of land use features in different geographical areas is analyzed and detected to generate detection groups, which are then uploaded to the detection log. This enables the perception of land use behavior in multi-regional coordinated planning adjustments, providing an effective land use data association detection strategy.

Description

Technical Field

[0001] This invention relates to the field of remote sensing image technology, specifically to a method and system for detecting land use change based on remote sensing images. Background Technology

[0002] Multi-temporal remote sensing imagery technology can acquire image data of the same area at different times using satellite or airborne remote sensing platforms. These image data cover a variety of features of the land surface, such as spectrum, texture, and shape. By comparing and analyzing the changes of these features over time, changes in land use types can be identified, such as farmland being converted into construction land or forest land being converted into grassland.

[0003] In existing technologies, multi-temporal remote sensing imagery can be used to capture and accurately reflect subtle changes in land use in real time, thus making up for traditional land use monitoring methods, such as manual field surveys and cadastral surveys.

[0004] However, with the acceleration of urbanization and the adjustment of agricultural land use, changes in land use types are becoming more frequent and complex. Multi-temporal remote sensing imagery technology can only detect single ground areas and cannot cope with the land use modal perception in the context of coordinated planning and adjustment of multiple ground areas. Summary of the Invention

[0005] The purpose of this invention is to provide a land use change detection method and system based on remote sensing imagery to solve the problems mentioned in the background art.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] A land use change detection system based on remote sensing imagery, comprising: a remote sensing module, an image data management module, and a cloud data analysis module;

[0008] The remote sensing module is used to remotely control the drone and acquire multi-temporal remote sensing images of the area.

[0009] The image data management module is used to attach label attributes to the acquired multi-temporal remote sensing images and perform image feature processing to generate image stream clusters. It is also used to characterize the feature state corresponding to the image features through a feature capture function.

[0010] The cloud-based data analysis module is used to analyze and calculate the change status value of land use, and also to analyze and detect the correlation of land use characteristics in different geographical areas to generate detection groups and upload them to the detection log.

[0011] Furthermore, the remote sensing module is used to acquire the multi-temporal remote sensing images by acquiring multi-temporal remote sensing images of different geographical areas through a remote sensing platform mounted on a drone.

[0012] Furthermore, the tag attributes include time tag attributes and ground area tag attributes. The time tag is the acquisition time node of the multi-temporal remote sensing image, and the ground area tag is the geographical range displayed by the multi-temporal remote sensing image. The geographical range is set by pre-dividing the surface boundary areas, and one surface boundary area corresponds to one geographical range. The image feature processing includes time tag dimension processing, ground area tag dimension processing, and crop type dimension processing, and generates time tag serial numbers, ground area tag serial numbers, and crop type numbers respectively. Based on the crop type numbers in ascending order, an initial arrangement of each image feature is performed. In response to the completion of the initial arrangement, based on the time tag serial numbers in ascending order, an optimized arrangement is performed on the result of the initial arrangement. Through the two methods of initial arrangement and optimized arrangement, the image stream cluster is generated.

[0013] A land use change detection method based on remote sensing imagery, comprising the following steps:

[0014] S1. Add label attributes to the acquired multi-temporal remote sensing images, the label attributes including time label attributes and ground area label attributes;

[0015] S2. Based on the label attributes, add image feature labels to the multi-temporal remote sensing images, select the ground area label attributes as image flow guides, sort and organize the various image features to obtain image flow clusters;

[0016] S3. Based on image features, generate feature states and characterize the feature states using a feature capture function;

[0017] S4. Based on image stream clusters and feature capture functions, analyze and calculate the land use change status values;

[0018] S5. Based on the land use change status value, analyze and detect the correlation of land use characteristics in different geographical areas to generate detection groups and upload them to the detection log.

[0019] Furthermore, the specific implementation process of the additional tag attribute includes:

[0020] Multi-temporal remote sensing images of different geographical areas are collected by a remote sensing platform mounted on a drone. The multi-temporal remote sensing images have time label attributes and ground area label attributes. The time label is the time node of the acquisition of the multi-temporal remote sensing images, and the ground area label is the geographical area displayed by the multi-temporal remote sensing images. The geographical area is set by pre-dividing the surface boundary area, and one surface boundary area corresponds to one geographical area.

[0021] Furthermore, the image stream cluster is generated in the following manner:

[0022] Based on the time-label sequence, the multi-temporal remote sensing images are numbered, and the i-th multi-temporal remote sensing image is denoted as RS. i When the a-th ground region label G a The crop type planted at the i-th time label is C. e At that time, for multi-temporal remote sensing images RS i Additional image feature markers, denoted as RS i (G a C e ), where e represents the crop type number;

[0023] Based on image features, using ground area labels G a To guide the image stream, the various image features are sorted and organized to obtain an image stream cluster, denoted as FL(G). a )={RS i (G a C e )|i∈[1,I],e∈[1,E]}; The sorting and sorting method includes initial sorting and sorting and optimized sorting and sorting, and the optimized sorting and sorting responds to the completion of the initial sorting and sorting; The initial sorting and sorting is: to initialize and arrange each image feature in ascending order according to the crop type number; The optimized sorting and sorting is: to optimize and arrange each image feature in ascending order according to the time tag number based on the initial sorting and sorting result.

[0024] Furthermore, the specific implementation process of characterizing the feature state through the feature capture function includes:

[0025] For the image features RS i (G a C e Feature capture is performed and feature states are generated, denoted as i:e. Based on the feature states, a feature capture function is constructed, denoted as f(y:x) = y, where x is the independent variable and the value of x corresponds to the crop type number, y is the dependent variable and the value of y corresponds to the time tag number. When the feature state is i:e, then f(y:x) = f(i:e) = i.

[0026] Furthermore, the specific implementation process for analyzing and calculating the state values ​​of land use change includes:

[0027] Based on image stream clusters and feature capture functions, the change state values ​​of land use are analyzed and calculated.

[0028]

[0029] SV(G a ) indicates the ground area label G a The land use change state value, j:(e+1) represents the image feature RS. j (G a C e+1 The corresponding generated feature state, RS j Let C represent the j-th multi-temporal remote sensing image. e+1 Represents the (e+1)th crop type, NUM[FL(G a )] represents the image stream cluster FL(C a The total number of image features contained in );

[0030] According to the above method, the land use change detection in this invention includes the perception of two types of underlying data: crop type and time node. In particular, the change of crop type is subjectively affected by the change of time node. The essence of this subjective influence comes from policy adjustments or human cultivation behavior. Therefore, under the influence of both crop type change and time node change, the features of the collected multi-temporal remote sensing images exhibit irregularity. This phenomenon makes traditional land use change detection difficult to cope with the perception of land use behavior in the context of coordinated planning adjustments in multiple ground areas. This invention simplifies the recording of feature states by capturing the features of multi-temporal remote sensing images within the same ground area, and constructs a feature capture function based on this. The essence of the feature capture function is to capture the degree of jump in cultivation time caused by crop type change, that is, the land use change state value. The larger the change state value, the greater the degree of jump in cultivation time caused by crop type change, and the more obvious the change in land use behavior.

[0031] Furthermore, the specific implementation process for generating the detection groups includes:

[0032] Based on land use change status values, the correlation of land use characteristics across different geographical regions was analyzed and detected.

[0033]

[0034] FC(G a →G b ) indicates the ground area label G aand ground area label G b Correlation of land use characteristics between them

[0035] SV(G b ) indicates the ground area label G b The land use change state value is given by , where b is the ground area label number, and a ≠ b, and μ represents the average value of the land use change state value. A represents the total number of ground areas, σ 2 The variance of the state values ​​representing changes in land use is given, and

[0036] A preset relevance threshold is set. If the relevance of land use characteristics is greater than or equal to the threshold, then the land area is labeled G. a and ground area label G b The samples were grouped together for testing and uploaded to the testing log. Staff then conducted on-site inspections according to the testing groups.

[0037] Based on the above method, the analysis of land use characteristic correlation is essentially a manifestation of correlation probability, as shown in the formula SV(G a )+SV(G b )-2μ is converted to obtain SV(G a )-μ+SV(G b )-μ, i.e., the state change value SV(G) a ) and the change state value SV(G b The difference between each and the average value is denoted as , and the smaller the difference is, the more similar the land use behavior of the two land areas is, or the more complementary the benefits are.

[0038] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, for causing the processor to execute the computer program via computer-executable instructions to implement a land use change detection method based on remote sensing imagery according to the present invention.

[0039] Compared with existing technologies, the beneficial effects achieved by this invention are as follows: In the land use change detection method and system based on remote sensing imagery provided by this invention, label attributes are added to the acquired multi-temporal remote sensing images, including time label attributes and ground area label attributes; based on the label attributes, image feature markers are added to the multi-temporal remote sensing images, and ground area label attributes are selected as image flow guides to sort and organize the various image features, obtaining image flow clusters; based on the image features, feature states are generated, and feature capture functions are used to characterize the feature states; based on the image flow clusters and feature capture functions, land use change state values ​​are analyzed and calculated; based on the land use change state values, the correlation of land use features in different geographical areas is analyzed and detected to generate detection groups, which are then uploaded to the detection log; thereby, land use behavior perception in multi-ground area linkage planning adjustments is realized, providing an effective land use data association detection strategy. Attached Figure Description

[0040] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof.

[0041] In the attached diagram: Figure 1 This is a schematic diagram illustrating the steps of a land use change detection method based on remote sensing imagery according to the present invention. Detailed Implementation

[0042] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0043] In this first embodiment: a land use change detection system based on remote sensing imagery is provided. The system includes: a remote sensing module, an image data management module, and a cloud data analysis module.

[0044] The remote sensing module is used to remotely control the drone and acquire multi-temporal remote sensing images of the area.

[0045] Preferably, multi-temporal remote sensing images of different geographical areas are collected through a remote sensing platform mounted on a drone;

[0046] The image data management module is used to attach label attributes to the acquired multi-temporal remote sensing images and perform image feature processing to generate image stream clusters. It is also used to characterize the feature state corresponding to the image features through a feature capture function.

[0047] Preferredly, the tag attributes include time tag attributes and ground area tag attributes. The time tag is the acquisition time node of the multi-temporal remote sensing image, and the ground area tag is the geographical range displayed by the multi-temporal remote sensing image. The geographical range is set by pre-dividing the surface boundary areas, and one surface boundary area corresponds to one geographical range. The image feature processing includes time tag dimension processing, ground area tag dimension processing, and crop type dimension processing, and generates time tag serial numbers, ground area tag serial numbers, and crop type numbers respectively. Based on the crop type numbers in ascending order, an initial arrangement of each image feature is performed. In response to the completion of the initial arrangement, based on the time tag serial numbers in ascending order, an optimized arrangement is performed on the result of the initial arrangement. The image stream cluster is generated through the two methods of initial arrangement and optimized arrangement.

[0048] The cloud-based data analysis module is used to analyze and calculate the change status value of land use, and also to analyze and detect the correlation of land use characteristics in different geographical areas to generate detection groups and upload them to the detection log.

[0049] Please see Figure 1 In this second embodiment: a land use change detection method based on remote sensing imagery is provided, which includes the following steps:

[0050] S1. Add label attributes to the acquired multi-temporal remote sensing images, the label attributes including time label attributes and ground area label attributes;

[0051] For example, multi-temporal remote sensing images of different geographical areas are collected by a remote sensing platform mounted on a drone. The multi-temporal remote sensing images have time label attributes and ground area label attributes. The time label is the time node of the acquisition of the multi-temporal remote sensing images, and the ground area label is the geographical area displayed by the multi-temporal remote sensing images. The geographical area is set by pre-dividing the surface boundary area, and one surface boundary area corresponds to one geographical area.

[0052] S2. Based on the label attributes, add image feature labels to the multi-temporal remote sensing images, select the ground area label attributes as image flow guides, sort and organize the various image features to obtain image flow clusters;

[0053] For example, the sequence number of the multi-temporal remote sensing images is compiled based on the time tag number order, and the i-th multi-temporal remote sensing image is denoted as RS. i When the a-th ground region label G a The crop type planted at the i-th time label is C. e At that time, for multi-temporal remote sensing images RS iAdditional image feature markers, denoted as RS i (G a C e ), where e represents the crop type number;

[0054] Based on image features, using ground area labels G a To guide the image stream, the various image features are sorted and organized to obtain an image stream cluster, denoted as FL(G). a )={RS i (G a C e )|i∈[1,I],e∈[1,E]}; The sorting and sorting method includes initial sorting and sorting and optimized sorting and sorting, and the optimized sorting and sorting responds to the completion of the initial sorting and sorting; The initial sorting and sorting is: to initialize and arrange each image feature in ascending order according to the crop type number; The optimized sorting and sorting is: to optimize and arrange each image feature in ascending order according to the time tag number based on the initial sorting and sorting result;

[0055] S3. Based on image features, generate feature states and characterize the feature states using a feature capture function;

[0056] For example, regarding the image features RS i (G a C e Feature capture is performed and feature states are generated, denoted as i:e. Based on the feature states, a feature capture function is constructed, denoted as f(y:x) = y, where x is the independent variable and the value of x corresponds to the crop type number, y is the dependent variable and the value of y corresponds to the time tag number. When the feature state is i:e, then f(y:x) = f(i:e) = i.

[0057] S4. Based on image stream clusters and feature capture functions, analyze and calculate the land use change status values;

[0058] For example, based on image stream clusters and feature capture functions, the change state values ​​of land use are analyzed and calculated.

[0059] SV(G a ) indicates the ground area label G a The land use change state value, j:(e+1) represents the image feature RS. j (G a C e+1 The corresponding generated feature state, RS j Let C represent the j-th multi-temporal remote sensing image. e+1 Represents the (e+1)th crop type, NUM[FL(G a)] represents the image stream cluster FL(G a The total number of image features contained in );

[0060] S5. Based on the land use change status value, analyze and detect the correlation of land use characteristics in different geographical areas to generate detection groups and upload them to the detection log;

[0061] For example, based on land use change status values, the correlation of land use characteristics in different geographical areas can be analyzed and detected. In the formula, FC(G) a →G b ) indicates the ground area label G a and ground area label G b Correlation of land use characteristics between them, SV(G b ) indicates the ground area label G b The land use change state value is given by , where b is the ground area label number, and a ≠ b, and μ represents the average value of the land use change state value. A represents the total number of ground areas, σ 2 The variance of the state values ​​representing changes in land use is given, and

[0062] A preset relevance threshold is set. If the relevance of land use characteristics is greater than or equal to the threshold, then the land area is labeled G. a and ground area label G b The samples were grouped together for testing and uploaded to the testing log. Staff then conducted on-site inspections according to the testing groups.

[0063] For example, if the average value is 5, and the state value changes by SV(G) a ) and SV(G b If 10 and 3 are respectively, then SV(G) a )+SV(G b )-2μ equals 3, if the changing state value SV(G) a ) and SV(G a If ) are 7 and 6 respectively, then SV(G a )+SV(G b The value of 3 for )-2μ indicates that the land use behavior of the two land areas is complementary or similar.

[0064] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, for causing the processor to execute the computer program via computer-executable instructions to implement a land use change detection method based on remote sensing imagery according to the present invention.

[0065] For example, a computer storage medium may take the form of any combination of one or more computer-readable media. A computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in connection with an instruction execution system, apparatus, or device.

[0066] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, capable of sending, propagating, or transmitting programs for use by or in connection with an instruction execution system, apparatus, or device.

[0067] The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to wireless, wired, optical fiber, etc., or any suitable combination thereof. The computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages—such as Java, Smalltalk, and C++—and conventional procedural programming languages—such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer can be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0068] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0069] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A land use change detection method based on remote sensing imagery, characterized in that, The method includes the following steps: S1. Add label attributes to the acquired multi-temporal remote sensing images, the label attributes including time label attributes and ground area label attributes; S2. Based on the label attributes, add image feature labels to the multi-temporal remote sensing images, select the ground area label attributes as image flow guides, sort and organize the various image features to obtain image flow clusters; S3. Based on image features, generate feature states and characterize the feature states using a feature capture function; S4. Based on image stream clusters and feature capture functions, analyze and calculate the land use change status values; S5. Based on the land use change status value, analyze and detect the correlation of land use characteristics in different geographical areas to generate detection groups and upload them to the detection log; The image stream clusters are generated in the following way: Based on the time-label sequence, the multi-temporal remote sensing images are numbered, and the i-th multi-temporal remote sensing image is denoted as... When the a-th ground area label The type of crop planted under the i-th time label is Then, for multi-temporal remote sensing images Additional image feature markers, denoted as 'e' represents the crop type number; Based on image features, using ground area labels To guide the image stream, the various image features are sorted and organized to obtain an image stream cluster, denoted as... The sorting and sorting methods include initial sorting and sorting and optimized sorting and sorting, and the optimized sorting and sorting responds to the completion of the initial sorting and sorting. The initial sorting and sorting is: to initially arrange each image feature in ascending order according to the crop type number; the optimized sorting and sorting is: to optimize the arrangement of each image feature in ascending order according to the time tag number based on the initial sorting and sorting results. The specific implementation process of characterizing the feature state through the feature capture function includes: For the image features Perform feature capture and generate feature states, denoted as i:e. Based on the feature states, construct a feature capture function, denoted as f(y:x)=y, where x is the independent variable and the value of x corresponds to the crop type number, y is the dependent variable and the value of y corresponds to the time tag number. When the feature state is i:e, then f(y:x)=f(i:e)=i. The specific implementation process for analyzing and calculating the state values ​​of land use change includes: Based on image stream clusters and feature capture functions, the change state values ​​of land use are analyzed and calculated. In the formula, Ground area label The land use change state value, j:(e+1) represents the image feature. Corresponding to the generated feature states, This represents the j-th multi-temporal remote sensing image. This represents the (e+1)th crop type. Represents image stream clusters The total number of image features contained therein; The specific implementation process for generating the detection group includes: Based on land use change status values, the correlation of land use characteristics across different geographical regions was analyzed and detected. In the formula, Ground area label and ground area labels Correlation of land use characteristics between them Ground area label The land use change status value, b is the ground area label number, and a≠b. This represents the average value of the state of land use change, and A represents the total number of ground areas. The variance of the state values ​​representing changes in land use is given, and ; A preset relevance threshold is set. If the relevance of land use characteristics is greater than or equal to the threshold, then the ground area is labeled. and ground area labels The samples were grouped together for testing and uploaded to the testing log. Staff then conducted on-site inspections according to the testing groups.

2. The land use change detection method based on remote sensing imagery according to claim 1, characterized in that, The specific implementation process of the additional tag attribute includes: Multi-temporal remote sensing images of different geographical areas are collected by a remote sensing platform mounted on a drone. The multi-temporal remote sensing images have time label attributes and ground area label attributes. The time label is the time node of the acquisition of the multi-temporal remote sensing images, and the ground area label is the geographical area displayed by the multi-temporal remote sensing images. The geographical area is set by pre-dividing the surface boundary area, and one surface boundary area corresponds to one geographical area.

3. A land use change detection system based on remote sensing imagery, executing the land use change detection method based on remote sensing imagery as described in any one of claims 1-2, characterized in that, The system includes: a remote sensing module, an image data management module, and a cloud data analysis module; The remote sensing module is used to remotely control the drone and acquire multi-temporal remote sensing images of the area. The image data management module is used to attach label attributes to the acquired multi-temporal remote sensing images and perform image feature processing to generate image stream clusters. It is also used to characterize the feature state corresponding to the image features through feature capture functions. The cloud-based data analysis module is used to analyze and calculate the change status value of land use, and also to analyze and detect the correlation of land use characteristics in different geographical areas to generate detection groups and upload them to the detection log.

4. A land use change detection system based on remote sensing imagery according to claim 3, characterized in that: The remote sensing module is used to acquire the multi-temporal remote sensing images by acquiring multi-temporal remote sensing images of different geographical areas through a remote sensing platform mounted on a drone.

5. A land use change detection system based on remote sensing imagery according to claim 3, characterized in that: The tag attributes include time tag attributes and ground area tag attributes. The time tag is the acquisition time node of the multi-temporal remote sensing image, and the ground area tag is the geographical range displayed by the multi-temporal remote sensing image. The geographical range is set by pre-dividing the surface boundary area, and one surface boundary area corresponds to one geographical range. The image feature processing includes time tag dimension processing, ground area tag dimension processing and crop type dimension processing, which respectively generate time tag serial number, ground area tag serial number and crop type number; The initial arrangement of each image feature is performed based on the crop type number in ascending order. In response to the completion of the initial arrangement, the result of the initial arrangement is optimized based on the time tag number in ascending order. The image stream cluster is generated through the two methods of initial arrangement and optimization arrangement.

6. An electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, By means of computer-executable instructions, the processor executes the computer program to implement the land use change detection method based on remote sensing imagery as described in any one of claims 1-2.

Images