Paeonia lactiflora planting structure identification method and device based on remote sensing cloud data platform

Abstract

The invention provides a Chinese herbaceous peony planting structure identification method and device based on a remote sensing cloud data platform and an algorithm, relates to the field of remote sensing images, and solves the technical problem of insufficient accuracy of Chinese herbaceous peony planting structure identification. The method comprises the following steps: collecting a remote sensing image of Chinese herbaceous peony in a to-be-identified area, and correcting the remote sensing image of the Chinese herbaceous peony; spectral features, vegetation indexes and leaf texture contrast of the peony in the to-be-recognized area are obtained and input into a recognition module together with the corrected remote sensing image, and a planting structure of the peony in the to-be-recognized area is obtained; the method is used in a Chinese herbaceous peony planting structure identification process, spectral features, vegetation indexes, leaf texture contrast and other multi-modal data are combined, the Chinese herbaceous peony feature capture capability of the model is enhanced, a remote sensing cloud data platform can be effectively utilized to accurately identify a Chinese herbaceous peony planting structure, and a dynamic threshold and terrain grouping strategy is introduced to improve the identification accuracy of the Chinese herbaceous peony planting structure. Abnormal image elimination and sample classification are optimized, and the risk of single feature failure is reduced through multi-feature fusion.

Description

Technical Field

[0001] The present application relates to the field of remote sensing imaging, and in particular to a method and device for identifying peony planting structures based on a remote sensing cloud data platform and algorithm. Background Art

[0002] Peony is a key Chinese medicinal material and has ornamental value. The accurate identification of peony planting structure is of great significance to planting management, yield prediction and resource optimization. For example, accurate identification of the spatial distribution, density and boundaries of peony planting can grasp the actual utilization of the field in real time. By comparing the growth of plants under different planting densities or terrains (such as slope and elevation), the planting pattern can be optimized.

[0003] At present, the identification of peony planting structure mainly relies on manual surveys. Although remote sensing technology has gradually replaced some manual work, it relies on single features in data processing and lacks efficient algorithms, resulting in insufficient accuracy in the identification of peony planting structure. In addition, the current multi-source data integration capabilities are insufficient, making it difficult to fully utilize remote sensing data. Summary of the Invention

[0004] The present application provides a method and device for identifying peony planting structures based on a remote sensing cloud data platform and algorithm, which solves the technical problem of insufficient accuracy in identifying peony planting structures.

[0005] To achieve the above objectives, this application adopts the following technical solutions: First, a method for identifying peony planting structure based on a remote sensing cloud data platform and algorithm is provided, comprising: S1: Obtain the spectral characteristics of peony in the planting area through a multispectral sensor; S2: Calculate the vegetation index of peony in the area to be identified based on the spectral characteristics; S3: obtaining a plurality of first remote sensing images of peonies according to the vegetation index of peonies, and extracting the texture contrast of the peonies leaves in the first remote sensing images; S4: removing abnormal images from the first remote sensing images of the plurality of peonies to obtain a second remote sensing image; S5: classifying the second remote sensing image according to the sample distribution of the second remote sensing image to obtain a plurality of image groups; S6: training the recognition model to be trained in batches using the spectral features, texture features, and planting structures corresponding to the plurality of image groups and the second remote sensing images in the plurality of image groups to obtain a recognition model; S7: Collect remote sensing images of peony in the area to be identified, correct the remote sensing images, and obtain the spectral characteristics, vegetation index, leaf texture contrast of peony in the area to be identified and input the corrected remote sensing images into the recognition module to obtain the planting structure of peony in the area to be identified.

[0006] In combination with the first aspect above, in a possible implementation, the spectral characteristics include: reflectivity NIR in a near-infrared band, reflectivity RED in a red light band, and reflectivity BLUE in a blue light band.

[0007] In combination with the first aspect above, in a possible implementation, the vegetation index of the peony includes the normalized vegetation index NDVI and the enhanced vegetation index EVI; The calculation method of the normalized difference vegetation index NDVI is: ; The calculation method of the enhanced vegetation index EVI is: ; in, and are the proportional coefficients of the red light band reflectivity RED and the blue light band reflectivity BLUE, and L is the soil adjustment coefficient.

[0008] In conjunction with the first aspect above, in a possible implementation, obtaining first remote sensing images of a plurality of peonies according to the vegetation index of peonies includes the following steps: When the normalized difference vegetation index NDVI is within the range threshold, the normalized vegetation index NDVI is linearly fitted according to the time period of obtaining the spectral characteristics to obtain the first fitting curve ; For the first fitting curve Perform the derivative operation and get ;Will Compared with the fluctuation threshold, when If it is less than the fluctuation threshold, jump to S13; otherwise, do nothing; Determine whether the enhanced vegetation index EVI meets the constraints ; Yes, obtain the first remote sensing image through the remote sensing imaging device; otherwise, do nothing; where t is the number of times the spectral characteristics of peony are collected.

[0009] In combination with the first aspect above, in a possible implementation, the leaf texture contrast acquisition method includes: Convert the first remote sensing image into a single-channel grayscale image using Open CV; The grayscale value of the single-channel grayscale image is discretized into 8 levels, and a grayscale co-occurrence matrix is ​​constructed. The number of times each grayscale level (x, y) appears at a preset distance and in a preset direction is counted, and then the grayscale co-occurrence matrix is ​​normalized to obtain a probability matrix p(x, y); where x is the grayscale level of the reference pixel and y is the grayscale level of the adjacent pixel; each interval discretized into 8 levels is defined as , z∈[0,7]; From the gray-level co-occurrence moment, we can use the formula Calculate leaf texture contrast.

[0010] It should be noted that the larger the value of the leaf texture contrast, the clearer the texture of the peony leaf.

[0011] In conjunction with the first aspect above, in one possible implementation, the abnormal image determination method includes: Extract the leaf texture contrast of peony in the first remote sensing image and compare it with the dynamic threshold range; when the leaf texture contrast is within the dynamic threshold range, the corresponding first remote sensing image is marked as a normal image; otherwise, the corresponding first remote sensing image is marked as an abnormal image; wherein the dynamic threshold range is [ ],in, is the historical mean value of the leaf texture contrast of several first remote sensing images, k is the adjustment coefficient, is the 25% quantile leaf texture contrast in several first remote sensing images, It is the 75% quantile of leaf texture contrast in several first remote sensing images.

[0012] It should be noted that the adjustment coefficient is dynamically adjusted according to the growth period of the peony. For example, when the peony is in the flowering period, k=1.05.

[0013] In conjunction with the first aspect above, in a possible implementation, classifying the second remote sensing image according to the sample distribution of the second remote sensing image includes the following steps: The terrain slope of the peony planting area is obtained through the terrain analysis tool in the remote sensing cloud service platform; Matching the location information of the second remote sensing image with the terrain slope of the peony planting area; Grouping the terrain slopes corresponding to the second remote sensing images in the plurality of time phase groups to obtain a plurality of time phase groups; The multiple phase groups are divided into multiple image groups according to the growth period of the peony in the second remote sensing image.

[0014] In combination with the first aspect above, in one possible implementation, the loss of the batch training process of the recognition model to be trained is: , in, 、 、 are weight parameters; N is the number of batch samples, is the true category label of the i-th sample, is the probability that the i-th sample belongs to the c-th class predicted by the model, N is the number of batch samples, C is the total number of image groups, B is the batch size, and M is the number of positive samples for each sample. is the feature similarity between the bth sample and the i-th positive sample, , is the multimodal feature vector of the b-th sample, is the feature vector of the positive sample, , is the feature vector of the negative sample, m is the contrast margin, is the regularization weight used to control the strength of terrain constraints, K is the number of groups of terrain slope, is the number of samples in the kth group, is the index set of the k-th group of samples, is the k-th group feature mean vector, is the norm.

[0015] In combination with the first aspect above, in a possible implementation, the remote sensing image is corrected, including: geometric correction, radiation correction, and atmospheric correction.

[0016] In a second aspect, an electronic device is provided, comprising: a communication unit and a processing unit; the communication unit is used for remote sensing images of peonies in an area to be identified; the processing unit is used to correct the remote sensing images of peonies; obtain vegetation index and leaf texture contrast, and input the spectral characteristics, vegetation index, and leaf texture contrast of the peonies in the area to be identified into an identification module together with the corrected remote sensing image to obtain the planting structure of the peonies in the area to be identified.

[0017] In a third aspect, the present application provides a processing device comprising: a processor and a storage medium; the storage medium comprising instructions, the processor configured to execute the instructions to implement the method described in the first aspect and any possible implementation of the first aspect. The processing device may be an electronic device or a chip within the electronic device.

[0018] In a fourth aspect, the present application provides a peony planting structure identification system based on a remote sensing cloud data platform and algorithm, comprising: an information acquisition module and a processing module; wherein the information acquisition module is used to obtain remote sensing images and spectral characteristics of peony in the area to be identified; The processing module is used to correct the remote sensing image of peony; obtain the vegetation index and leaf texture contrast, and input the spectral characteristics, vegetation index, and leaf texture contrast of the peony in the area to be identified into the recognition module together with the corrected remote sensing image to obtain the planting structure of the peony in the area to be identified.

[0019] In a fifth aspect, the present application provides a computer-readable storage medium, which stores instructions. When the instructions are executed on a processing device, the processing device executes the method described in the first aspect and any possible implementation of the first aspect.

[0020] In a sixth aspect, the present application provides a computer program product comprising instructions, which, when executed on a processing device, enables the processing device to execute the method as described in the first aspect and any possible implementation of the first aspect.

[0021] The present application provides a method and device for identifying peony planting structures based on a remote sensing cloud data platform and algorithm. By combining multimodal data such as spectral characteristics, vegetation index, and leaf texture contrast, the model's ability to capture peony features is enhanced, and the remote sensing cloud data platform can be effectively utilized to accurately identify peony planting structures. Dynamic thresholds and terrain grouping strategies are introduced to optimize abnormal image removal and sample classification, reduce misjudgment, and multi-feature fusion reduces the risk of single feature failure.

[0022] It should be understood that the description of technical features, technical solutions, beneficial effects or similar language in this application does not imply that all features and advantages can be realized in any single embodiment. On the contrary, it is understood that the description of a feature or beneficial effect means that a specific technical feature, technical solution or beneficial effect is included in at least one embodiment. Therefore, the description of a technical feature, technical solution or beneficial effect in this specification does not necessarily refer to the same embodiment. Furthermore, the technical features, technical solutions and beneficial effects described in the present embodiment can also be combined in any appropriate manner. Those skilled in the art will understand that the embodiment can be implemented without one or more specific technical features, technical solutions or beneficial effects of a specific embodiment. In other embodiments, additional technical features and beneficial effects can also be identified in specific embodiments that do not embody all embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 A system architecture diagram of an identification system provided in an embodiment of the present application; Figure 2 A flowchart of a method for identifying peony planting structure based on a remote sensing cloud data platform and algorithm provided in an embodiment of the present application; Figure 3A schematic diagram of the process of training a recognition model provided in an embodiment of the present application; Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application; Figure 5 A schematic diagram of the hardware structure of a processing device provided in an embodiment of the present application; DETAILED DESCRIPTION

[0024] In the description of this application, unless otherwise specified, " / " means "or", for example, A / B can mean A or B. "And / or" in this article is merely a description of the association relationship of associated objects, indicating that three relationships can exist. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, "at least one" means one or more, and "a plurality" means two or more. Words such as "first" and "second" do not limit the quantity and execution order, and words such as "first" and "second" do not limit them to be necessarily different.

[0025] It should be noted that, in this application, words such as "exemplary" or "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design described in this application as "exemplary" or "for example" should not be construed as being preferred or advantageous over other embodiments or designs. Rather, the use of words such as "exemplary" or "for example" is intended to present the relevant concepts in a concrete manner.

[0026] The embodiment of the present application provides a method for identifying peony planting structures based on a remote sensing cloud data platform and an algorithm, which can be applied to Figure 1 In the identification system 100 shown, Figure 1 As shown, the communication system includes: a remote sensing device 101, a multispectral sensor 102 and an edge computing node 103.

[0027] The remote sensing device 101 is used to collect remote sensing images of peony in the area to be identified.

[0028] The multispectral sensor 102 is used to obtain the spectral characteristics of the peony in the area to be identified.

[0029] The edge computing node 103 is used to correct the remote sensing image of peony, obtain the vegetation index based on the spectral characteristics, obtain the leaf texture contrast based on the corrected remote sensing image, and input the spectral characteristics, vegetation index, and leaf texture contrast of the peony in the area to be identified into the recognition module together with the corrected remote sensing image to obtain the planting structure of the peony in the area to be identified.

[0030] To solve the technical problem of insufficient accuracy in identifying peony planting structures in the prior art, the present invention provides a method for identifying peony planting structures based on a remote sensing cloud data platform and an algorithm. The method comprises: collecting remote sensing images of peony in the area to be identified, and correcting the remote sensing images of peony; The spectral characteristics, vegetation index, and leaf texture contrast of the peony in the area to be identified are obtained and input into the recognition module together with the corrected remote sensing image to obtain the planting structure of the peony in the area to be identified; based on this, the model's ability to capture peony features is enhanced by combining multimodal data such as spectral characteristics, vegetation index, and leaf texture contrast. The remote sensing cloud data platform can be effectively used to accurately identify the peony planting structure, and dynamic thresholds and terrain grouping strategies are introduced to optimize abnormal image removal and sample classification, reduce misjudgment, and reduce the risk of single feature failure through multi-feature fusion.

[0031] like Figure 2 As shown, the embodiment of the present application provides a method for identifying peony planting structure based on a remote sensing cloud data platform and an algorithm, including: S201 , collecting remote sensing images of peony in the area to be identified, and correcting the remote sensing images of peony.

[0032] Among them, geometric correction, radiation correction and atmospheric correction can be used when correcting remote sensing images.

[0033] In some implementations, when performing geometric correction, a mathematical mapping relationship between image coordinates and geographic coordinates can be established through control points with known geographic coordinates (such as road intersections and landmarks). When performing radiation correction, the digital quantization value (DN) can be converted into radiation brightness through sensor calibration parameters (such as gain and offset). Atmospheric correction can use ground-measured spectral data to establish a linear relationship between image brightness and true reflectivity for correction.

[0034] For example, when performing geometric correction, control points are matched through Google Earth high-precision maps. During radiation correction, DN values ​​are converted into reflectance based on sensor gain (Gain=1.2) and offset (Offset=5). During atmospheric correction, the FLAASH module of ENVI software is used to eliminate the influence of atmospheric scattering.

[0035] It should be noted that the corrected remote sensing images reduce geometric distortion and noise interference, and the combination of multi-dimensional features (spectrum, texture, vegetation index) enhances the model's ability to capture peony characteristics.

[0036] S202: Acquire spectral features of the peony in the area to be identified through a multispectral sensor, and calculate a vegetation index based on the spectral features.

[0037] Among them, spectral characteristics include: near-infrared reflectance NIR, red reflectance RED, blue reflectance BLUE, and vegetation indices, including normalized vegetation index NDVI and enhanced vegetation index EVI; NIR and RED bands are core indicators of vegetation health (such as the subsequent NDVI calculation depends on), and BLUE band helps distinguish between vegetation and non-vegetated areas; The calculation method of normalized vegetation index NDVI is: ; The calculation method of Enhanced Vegetation Index EVI is: ; in, and are the proportional coefficients of the red light band reflectivity RED and the blue light band reflectivity BLUE, and L is the soil adjustment coefficient, which is usually 1.

[0038] S203 , inputting the spectral characteristics, vegetation index, and leaf texture contrast of the peony in the area to be identified and the corrected remote sensing image into a recognition module to obtain the planting structure of the peony in the area to be identified.

[0039] Among them, the recognition model is a model pre-trained based on the first feature data; the first feature data is a historical remote sensing image and feature data of peonies in several areas in multiple growth cycles; the feature data includes spectral feature data and leaf texture contrast; the leaf texture contrast is obtained based on the corrected remote sensing image.

[0040] The leaf texture contrast acquisition method includes: The grayscale value of the single-channel grayscale image is discretized into 8 levels, and a grayscale co-occurrence matrix is ​​constructed. The number of times each grayscale level (x, y) appears at a preset distance and in a preset direction is counted, and then the grayscale co-occurrence matrix is ​​normalized to obtain a probability matrix p(x, y); where x is the grayscale level of the reference pixel and y is the grayscale level of the adjacent pixel; each interval discretized into 8 levels is defined as , z∈[0,7]; From the gray-level co-occurrence moment, we can use the formula Calculate leaf texture contrast.

[0041] See also Figure 3 As shown, in some implementations, the training method of the recognition model includes: The spectral characteristics of peony in the planting area were obtained using the multispectral sensor MicaSense RedEdge-MX carried by the drone; Calculate the vegetation index of peony in the area to be identified based on the spectral characteristics; Acquiring a plurality of first remote sensing images of peonies according to the vegetation index of peonies, and extracting the texture contrast of the peonies leaves in the first remote sensing images; Eliminating abnormal images from the first remote sensing images of a plurality of peonies to obtain a second remote sensing image; Classifying the second remote sensing image according to the sample distribution of the second remote sensing image to obtain a number of image groups; The spectral features, vegetation indexes, texture features and corresponding planting structures of the plurality of image groups and the second remote sensing images in the plurality of image groups are batch-trained for the recognition model to be trained to obtain a recognition model; wherein the recognition model to be trained is constructed based on a deep learning model.

[0042] In a possible implementation of the embodiment of the present application, the loss function of the batch training process of the recognition model to be trained is: , in, 、 、 are all weight parameters; N is the number of batch samples; is the true category label of the i-th sample; is the probability that the i-th sample belongs to the c-th class predicted by the model; N is the number of samples in the image group; C is the total number of image groups; B is the batch size; M is the number of positive samples for each sample; is the feature similarity between the bth sample and the i-th positive sample; ; is the multimodal feature vector of the b-th sample; is the feature vector of the positive sample; ; is the feature vector of the negative sample; m is the contrast margin, which is used to control the degree of distinction between positive and negative samples; is the regularization weight; K is the number of groups of terrain slope; is the number of samples in the kth group; is the index set of the kth group of samples; is the feature mean vector of the kth group; is the norm.

[0043] It should be noted that the loss function includes three weighted terms: classification error, feature alignment, and distribution consistency constraint, to avoid the model being biased towards a single indicator. 、 、 Different values ​​are adopted according to different growth periods. ∈[0.4,0.6], ∈[0.3,0.5], ∈[0.01,0.1], the contrast margin m is used to control the degree of distinction between positive and negative samples. When the growth period of peony is the flowering period, m=0.8, and when the growth period of peony is the non-flowering period, m=0.5.

[0044] In a possible implementation of the embodiment of the present application, obtaining first remote sensing images of several peonies according to the vegetation index of the peonies includes the following steps: When the normalized difference vegetation index NDVI is within the range threshold, the normalized vegetation index NDVI is linearly fitted according to the time period of obtaining the spectral characteristics to obtain the first fitting curve ; For the first fitting curve Perform the derivative operation and get ;Will Compared with the fluctuation threshold, when If it is less than the fluctuation threshold, jump to S13; otherwise, do nothing; Determine whether the enhanced vegetation index EVI meets the constraints ; Yes, obtain the first remote sensing image through the remote sensing imaging device; otherwise, do nothing; where t is the number of times the spectral characteristics of peony are collected.

[0045] In a possible implementation of the embodiment of the present application, a method for determining abnormal images includes: Extract the leaf texture contrast of peony in the first remote sensing image and compare it with the dynamic threshold range; when the leaf texture contrast is within the dynamic threshold range, the corresponding first remote sensing image is marked as a normal image; otherwise, the corresponding first remote sensing image is marked as an abnormal image; wherein the dynamic threshold range is [ ],in, is the historical mean value of the leaf texture contrast of several first remote sensing images, k is the adjustment coefficient, is the 25% quantile leaf texture contrast in several first remote sensing images, It is the 75% quantile of leaf texture contrast in several first remote sensing images.

[0046] It should be pointed out that the adjustment coefficient is dynamically adjusted according to the growth period of the peony. For example, when the peony is in the flowering period, k=1.05.

[0047] In a possible implementation of the embodiment of the present application, classifying the second remote sensing image according to the sample distribution of the second remote sensing image includes the following steps: The terrain slope of the peony planting area is obtained through the terrain analysis tool in the remote sensing cloud service platform; Matching the location information of the second remote sensing image with the terrain slope of the peony planting area; Grouping the terrain slopes corresponding to the second remote sensing images in the plurality of time phase groups to obtain a plurality of time phase groups; The multiple phase groups are divided into multiple image groups according to the growth period of the peony in the second remote sensing image.

[0048] It should be noted that when grouping according to terrain slope, those with a slope less than 5° are grouped together, those with a slope between 5° and 10° are grouped together, and those with a slope greater than 10° are grouped together.

[0049] The above mainly introduces the scheme of the embodiment of the present application from the perspective of device implementation. It is understandable that each device, for example, an electronic device, in order to implement the above functions, includes at least one of the hardware structure and software modules corresponding to the execution of each function. Those skilled in the art should easily appreciate that, in combination with the units and algorithm steps of each example described in the embodiments disclosed herein, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in a hardware or computer software driven hardware manner depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0050] The embodiment of the present application can divide the functional units of the electronic device according to the above method example. For example, each functional unit can be divided according to each function, or two or more functions can be integrated into one processing unit. The above integrated unit can be implemented in the form of hardware or in the form of software functional units. It should be noted that the division of units in the embodiment of the present application is schematic and is only a logical function division. There may be other division methods in actual implementation.

[0051] In the case of an integrated unit, Figure 5 A possible structural diagram of the electronic device (denoted as processing device 30 ) involved in the above embodiment is shown. The electronic device 30 includes a processing unit 301 and a communication unit 302 , and may further include a storage unit 303 . Figure 5 The structural schematic diagram shown can be used to illustrate the structure of the electronic device involved in the above embodiments.

[0052] when Figure 5 The structural diagram shown is used to illustrate the structure of the electronic device involved in the above embodiment. The processing unit 301 is used to control and manage the actions of the electronic device, the communication unit 302 is used for the electronic device to communicate with other devices, and the storage unit 303 is used to store program codes and data of the electronic device.

[0053] For example, the communication unit 302 is used to obtain remote sensing images and spectral characteristics of peony in the area to be identified; The processing unit 302 is used to correct the remote sensing image of the peony; obtain the vegetation index and leaf texture contrast, and input the spectral characteristics, vegetation index, and leaf texture contrast of the peony in the area to be identified into the recognition module together with the corrected remote sensing image to obtain the planting structure of the peony in the area to be identified; wherein the recognition model is a model pre-trained based on the first feature data; the first feature data is a pile of historical remote sensing images and feature data of peonies in several areas in multiple growth cycles; the feature data includes spectral feature data and leaf texture contrast; the leaf texture contrast is obtained based on the corrected remote sensing image.

[0054] The processing unit 301 may be a processor or a controller, and the communication unit 302 may be a communication interface, a transceiver, a transceiver, a transceiver circuit, a transceiver device, etc. The communication interface is a general term and may include one or more interfaces. The storage unit 303 may be a memory. When the electronic device 30 is a chip, the processing unit 301 may be a processor or a controller, and the communication unit 302 may be an input interface and / or output interface, a pin or a circuit, etc. The storage unit 303 may be a storage unit within the chip (for example, a register, a cache, etc.), or a storage unit located outside the chip (for example, a read-only memory (ROM), a random access memory (RAM), etc.).

[0055] Among them, the communication unit can also be called a transceiver unit. The antenna and control circuit with transceiver functions in the electronic device 30 can be regarded as the communication unit 302 of the electronic device 30, and the processor with processing function can be regarded as the processing unit 301 of the electronic device 30. Optionally, the device used to implement the receiving function in the communication unit 302 can be regarded as a communication unit, and the communication unit is used to perform the receiving steps in the embodiment of the present application. The communication unit can be a receiver, a receiver, a receiving circuit, etc. The device used to implement the sending function in the communication unit 302 can be regarded as a sending unit, and the sending unit is used to perform the sending steps in the embodiment of the present application. The sending unit can be a transmitter, a transmitter, a sending circuit, etc.

[0056] Figure 5If the integrated units are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) or a processor to execute all or part of the steps of the method described in each embodiment of the present application. The storage medium for storing computer software products includes various media that can store program codes, such as USB flash drives, mobile hard drives, read-only memories, random access memories, magnetic disks or optical disks.

[0057] Figure 5 A unit in a can also be called a module, for example, a processing unit can be called a processing module.

[0058] The present application also provides a hardware structure diagram of a processing device (denoted as processing device 40), see Figure 5 The processing device 40 includes a processor 401 and, optionally, a memory 402 connected to the processor 401 .

[0059] In the first possible implementation, see Figure 5 The processing device 40 further includes a transceiver 403. The processor 401, the memory 402, and the transceiver 403 are connected via a bus. The transceiver 403 is used to communicate with other devices or a communication network. Optionally, the transceiver 403 may include a transmitter and a receiver. The device used to implement the receiving function in the transceiver 403 can be considered a receiver, and the receiver is used to perform the receiving step in the embodiment of the present application. The device used to implement the transmitting function in the transceiver 403 can be considered a transmitter, and the transmitter is used to perform the transmitting step in the embodiment of the present application.

[0060] Based on the first possible implementation, Figure 5 The structural schematic diagram shown can be used to illustrate the structure of the processing device involved in the above embodiments.

[0061] in, Figure 5 It can also represent a system chip in the processing device. In this case, the actions performed by the processing device can be implemented by the system chip. The specific actions performed can be found above and will not be described in detail here.

[0062] During implementation, each step of the method provided in this embodiment can be completed by hardware integrated logic circuits in a processor or by software instructions. The steps of the method disclosed in the embodiments of this application can be directly implemented as execution by a hardware processor, or as a combination of hardware and software modules in a processor.

[0063] The processor in this application may include, but is not limited to, at least one of the following: a central processing unit (CPU), a microprocessor, a digital signal processor (DSP), a microcontroller unit (MCU), or an artificial intelligence processor, among other types of computing devices that run software. Each computing device may include one or more cores for executing software instructions to perform operations or processing. The processor may be a standalone semiconductor chip or integrated into a semiconductor chip with other circuits. For example, it may form a system-on-chip (SoC) with other circuits (such as a codec circuit, a hardware acceleration circuit, or various bus and interface circuits). Alternatively, it may be integrated into an ASIC as a built-in processor. The ASIC with the integrated processor may be packaged separately or together with other circuits. In addition to the core for executing software instructions to perform operations or processing, the processor may further include necessary hardware accelerators, such as a field programmable gate array (FPGA), a PLD (programmable logic device), or logic circuits that implement specialized logic operations.

[0064] The memory in the embodiments of the present application may include at least one of the following types: read-only memory (ROM) or other types of static storage devices that can store static information and instructions, random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, and electrically erasable programmable read-only memory (EEPROM). In some scenarios, the memory may also be a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto.

[0065] An embodiment of the present application also provides a computer-readable storage medium, comprising instructions, which, when executed on a computer, enables the computer to execute any of the above methods.

[0066] An embodiment of the present application also provides a computer program product comprising instructions, which, when executed on a computer, enables the computer to execute any of the above methods.

[0067] An embodiment of the present application also provides a chip, which includes a processor and an interface circuit, the interface circuit is coupled to the processor, the processor is used to run a computer program or instruction to implement the above method, and the interface circuit is used to communicate with other modules outside the chip.

[0068] In the above embodiments, all or part of the embodiments can be implemented using software, hardware, firmware, or any combination thereof. When implemented using a software program, all or part of the embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that includes one or more available media. Available media may be magnetic media (eg, floppy disks, hard disks, magnetic tapes), optical media (eg, DVDs), or semiconductor media (eg, solid state disks (SSDs)).

[0069] Although the present application is described herein in conjunction with various embodiments, in the process of implementing the claimed application, those skilled in the art may understand and implement other variations of the disclosed embodiments by reviewing the drawings, the disclosure, and the appended claims. In the claims, the word "comprising" does not exclude other components or steps, and "a" or "an" does not exclude multiple situations. A single processor or other unit may implement several functions listed in the claims. The fact that certain measures are recorded in different dependent claims does not mean that these measures cannot be combined to produce good results.

[0070] Although the present application has been described with reference to specific features and embodiments thereof, it is apparent that various modifications and combinations may be made thereto without departing from the spirit and scope of the present application. Accordingly, this specification and the drawings are merely illustrative of the present application as defined by the appended claims and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, those skilled in the art may make various modifications and variations to the present application without departing from the spirit and scope of the present application. Thus, the present application is intended to encompass such modifications and variations as fall within the scope of the claims of the present application and their equivalents.

Claims

1. A method for identifying peony planting structure based on remote sensing cloud data platform and algorithm, characterized in that: include: Collect remote sensing images of peony in the area to be identified, and correct the remote sensing images of peony; The spectral characteristics, vegetation index, and leaf texture contrast of the peony in the area to be identified are obtained and input into the recognition module together with the corrected remote sensing image to obtain the planting structure of the peony in the area to be identified; wherein, the recognition model is a model pre-trained based on the first feature data; the first feature data is a pile of remote sensing images and feature data of peonies in several areas in multiple growth cycles; the feature data includes spectral feature data and leaf texture contrast; the leaf texture contrast is obtained based on the corrected remote sensing image, and the vegetation index is obtained based on the spectral characteristics.

2. The method for identifying peony planting structure based on a remote sensing cloud data platform and algorithm according to claim 1, characterized in that: The training method of the recognition model includes: The spectral characteristics of peony in the planting area are obtained through multispectral sensors; Calculate the vegetation index of peony in the area to be identified based on the spectral characteristics; Acquiring a plurality of first remote sensing images of peonies according to the vegetation index of peonies, and extracting the texture contrast of the peonies leaves in the first remote sensing images; Eliminating abnormal images from the first remote sensing images of a plurality of peonies to obtain a second remote sensing image; Classifying the second remote sensing image according to the sample distribution of the second remote sensing image to obtain a number of image groups; The spectral features, vegetation indexes, texture features and corresponding planting structures of the plurality of image groups and the second remote sensing images in the plurality of image groups are batch-trained for the recognition model to be trained to obtain a recognition model; wherein the recognition model to be trained is constructed based on a deep learning model.

3. The method for identifying peony planting structure based on remote sensing cloud data platform and algorithm according to claim 2 is characterized in that: The loss function of the batch training process for the recognition model to be trained is: Among them, α, β, and γ are weight parameters; N is the number of batch samples; y ic is the true category label of the i-th sample; p ic is the probability that the i-th sample belongs to the c-th class predicted by the model; N is the number of samples in the image group; C is the total number of image groups; B is the batch size; M is the number of positive samples for each sample; is the feature similarity between the bth sample and the i-th positive sample; f b is the multimodal feature vector of the b-th sample; is the feature vector of the positive sample; is the feature vector of the negative sample; m is the contrast margin, which is used to control the degree of distinction between positive and negative samples; λ is the regularization weight; K is the number of groups of terrain slope; N k is the number of samples in the kth group; S k is the index set of the kth group of samples; μ k is the mean vector of the kth group of features; ‖‖ is the norm.

4. The method for identifying peony planting structure based on a remote sensing cloud data platform and algorithm according to claim 2, characterized in that: The spectral characteristics include: reflectivity NIR of the near infrared band, reflectivity RED of the red light band, and reflectivity BLUE of the blue light band.

5. The method for identifying peony planting structure based on remote sensing cloud data platform and algorithm according to claim 2, characterized in that: The vegetation index of the peony includes the normalized vegetation index NDVI and the enhanced vegetation index EVI; The calculation method of the normalized difference vegetation index NDVI is: The calculation method of the enhanced vegetation index EVI is: Among them, C1 and C2 are the proportional coefficients of the reflectivity RED of the red light band and the reflectivity BLUE of the blue light band, respectively, and L is the soil adjustment coefficient.

6. The method for identifying peony planting structure based on remote sensing cloud data platform and algorithm according to claim 5, characterized in that: The method of obtaining a plurality of first remote sensing images of peonies according to the vegetation index of peonies includes the following steps: When the normalized difference vegetation index NDVI is within the range threshold, the normalized vegetation index NDVI is linearly fitted according to the time period of obtaining the spectral characteristics to obtain the first fitting curve F NDVI ; For the first fitting curve F NDVI Perform the derivative operation and get F' NDVI ; F' NDVI (t) is compared with the fluctuation threshold, when F' NDVI If (t) is less than the fluctuation threshold, jump to S13; otherwise, do nothing; Determine whether the enhanced vegetation index EVI meets the constraints If yes, obtain the first remote sensing image through the remote sensing imaging device; otherwise, do nothing; where t is the number of times the spectral characteristics of the peony are collected.

7. The method for identifying peony planting structure based on remote sensing cloud data platform and algorithm according to claim 2, characterized in that: The leaf texture contrast acquisition method comprises: Convert the first remote sensing image into a single-channel grayscale image using Open CV; Discrete the grayscale values ​​of a single-channel grayscale image to 8 levels, construct a grayscale co-occurrence matrix, count the number of times each grayscale level (x, y) appears at a preset distance and in a preset direction, and then normalize the grayscale co-occurrence matrix to obtain a probability matrix p(x, y); where x is the grayscale level of the reference pixel and y is the grayscale level of the adjacent pixel; each interval of the discretization to 8 levels is defined as z∈[0,7]; The leaf texture contrast is calculated from the gray-level symbiosis moment using the formula.

8. The method for identifying peony planting structure based on remote sensing cloud data platform and algorithm according to claim 2 is characterized in that: The abnormal image determination method includes: Extract the leaf texture contrast of peony in the first remote sensing image and compare it with the dynamic threshold range; when the leaf texture contrast is within the dynamic threshold range, the corresponding first remote sensing image is marked as a normal image; otherwise, the corresponding first remote sensing image is marked as an abnormal image; wherein the dynamic threshold range is Among them, μ contrast is the historical mean value of the leaf texture contrast of several first remote sensing images, k is the adjustment coefficient, Q 25 is the 25% quantile leaf texture contrast in several first remote sensing images, Q 75 It is the 75% quantile of leaf texture contrast in several first remote sensing images.

9. The method for identifying peony planting structure based on remote sensing cloud data platform and algorithm according to claim 2, characterized in that: The classifying the second remote sensing image according to the sample distribution of the second remote sensing image includes: The terrain slope of the peony planting area is obtained through the terrain analysis tool in the remote sensing cloud service platform; Matching the location information of the second remote sensing image with the terrain slope of the peony planting area; Grouping the terrain slopes corresponding to the second remote sensing images in the plurality of time phase groups to obtain a plurality of time phase groups; The multiple phase groups are divided into multiple image groups according to the growth period of the peony in the second remote sensing image.

10. An electronic device, characterized in that: The device includes: a communication unit and a processing unit; The communication unit is used to obtain remote sensing images and spectral characteristics of peony in the area to be identified; The processing unit is used to correct the remote sensing image of peony; obtain vegetation index and leaf texture contrast, and input the spectral characteristics, vegetation index and leaf texture contrast of peony in the area to be identified into the recognition module together with the corrected remote sensing image to obtain the planting structure of peony in the area to be identified; wherein, the recognition model is a model pre-trained based on first feature data; the first feature data is a pile of historical remote sensing images and feature data of peonies in several areas in multiple growth cycles; the feature data includes spectral feature data and leaf texture contrast; the leaf texture contrast is obtained based on the corrected remote sensing image.

Images