Fully automatic rice precision mapping method, system and storage medium in cloudy and rainy areas
By combining optical imaging, remote sensing technology and model construction methods, the problem of low monitoring accuracy of rice in cloudy and rainy areas is solved, and the refined and accurate monitoring of rice is achieved, meeting the needs of accurate mapping.
Patent Information
- Application Number
- CN202411339381.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-25
- Publication Date
- 2024-11-29
- Estimated Expiration
- 2044-09-25
AI Technical Summary
The prior art is difficult to achieve refined and precise monitoring of rice crops in cloudy and rainy areas, especially in the absence of medium resolution remote sensing to express farming space and extreme climate conditions.
Using a combination of optical imaging, remote sensing technology and model construction, a scanning image of the target area is obtained, a contour model is produced, analyses the geomorphological segmentation image, analyzes the remote sensing feature information, determines the distribution information of crops and rice, and generates rice distribution images.
It has achieved refined monitoring of rice in cloudy and rainy areas, met the needs of accurate mapping, and overcomes the monitoring problems of medium resolution remote sensing and extreme climatic conditions.
Smart Images

Figure CN118864752B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of rice crop monitoring, and in particular to a method, system and storage medium for fully automatic rice precision mapping in cloudy and rainy areas. Background Art
[0002] Traditionally, rice distribution information is mainly obtained through field sampling, but it is difficult to adapt to large-scale monitoring needs and relies on a large survey team. The acquisition of information has the disadvantages of high cost, poor timeliness, and the results are greatly affected by subjectivity. In the face of the need for automatic and accurate extraction of rice by remote sensing in cloudy and rainy areas, the processing and mapping results of low- and medium-resolution remote sensing pixels are often unable to correspond to the actual agricultural farming management units, and there are serious problems of mixed pixels and mixed spectral information, which usually cannot meet the accuracy requirements of refined agricultural planting activity monitoring. At the same time, the complex climate environment of cloudy and rainy weather leads to the failure of optical remote sensing, making it difficult to conduct continuous observations and accurately understand the spatial distribution information of rice.
[0003] In response to the above problems, there is an existing Chinese invention patent CN117690024A, which uses SAR images to identify rice fields with multiple planting patterns, although it avoids the impact of weather conditions on image quality to a certain extent. However, for the monitoring of crop production processes, SAR remote sensing data images are subject to some limitations in agricultural remote sensing monitoring applications. The SAR backscatter coefficient is affected by multiple factors such as the structure of crops and surface properties (such as roughness and water content, etc.), and it is difficult to separate agricultural planting activity information from the surface properties from the backscatter coefficient. At the same time, SAR data itself has a large amount of noise, the visual features of the image are not obvious enough, and the ground imaging mechanism is not clear enough. In addition, factors such as geometric distortion caused by terrain undulations or unstable position caused by slant range imaging cause obvious deficiencies in its presentation of the spatial form of ground objects. Therefore, in the extraction of rice in cloudy and rainy areas, when only SAR remote sensing time series data is used to carry out precision agriculture remote sensing application research, it is difficult to achieve the interpretability of optical remote sensing indicators (such as NDVI) time series data for crop phenological response monitoring, resulting in low accuracy in rice crop identification and growth monitoring due to unclear mechanisms. Rice monitoring often requires more refined plot data that can carry agricultural activities. Especially when conducting precision rice extraction application research on the cloudy and rainy surface review, the main problems faced are that the medium-resolution remote sensing caused by the fragmentation of the plot cannot accurately express the farming space and the extreme climatic conditions make it difficult to establish optical long-term series remote sensing data for continuous observation of the crop growth process, which in turn cannot meet the needs of refined and accurate monitoring of rice crops in cloudy and rainy areas. Summary of the invention
[0004] The purpose of the present invention is to solve the shortcomings of the prior art that it cannot meet the needs of refined and accurate monitoring of rice crops in cloudy and rainy areas, and to propose a fully automatic rice precision mapping method, system and storage medium for cloudy and rainy areas.
[0005] In order to achieve the above object, the present invention adopts the following technical solutions:
[0006] In a first aspect, the present invention provides a method for fully automatic rice precision mapping in cloudy and rainy areas, comprising:
[0007] Obtain a scanned image of the target area, make a contour model of the target area, and determine a landform segmentation image of the target area in combination with the landform features of the target area;
[0008] According to the landform segmentation image of the target area, remote sensing feature information of the target area is obtained respectively, and the remote sensing feature information of the target area is analyzed to determine the distribution information of the crop objects and non-crop objects in the target area;
[0009] Obtain the growth characteristic information of rice in the target area, make the remote sensing characteristic transformation curve of rice in the target area, and determine the distribution information of rice crops and other crops in the target area;
[0010] A rice distribution image of the target area is produced based on the distribution information of the crop objects and non-crop objects in the target area and / or the distribution information of the rice crops and other crops in the target area, and in combination with the landform segmentation image of the target area.
[0011] In some feasible solutions, the method for determining the distribution information of crop objects and non-crop objects in the target area includes:
[0012] Determine the identification objects existing in the target area according to the geomorphic features of the target area and the geomorphic segmentation image of the target area;
[0013] According to the remote sensing feature information of the target area, remote sensing feature information of each identified object in the target area is obtained respectively;
[0014] Acquire remote sensing features of rice objects at different periods, and combine remote sensing feature information of each identified object in the target area to produce first screening information of the rice objects;
[0015] According to the first screening information of the rice object, for the remote sensing feature information of each identified object in the target area, at least one first target object that meets the first screening information is determined in the target area.
[0016] In some feasible solutions, the method of preparing the first screening information of the rice object includes:
[0017] According to the remote sensing characteristics of the rice object at different periods, the minimum value of the remote sensing characteristics of the rice object is determined;
[0018] Acquire remote sensing feature information of each identified object in the target area in at least two different time series, and determine an average value of the remote sensing feature of each identified object in the target area;
[0019] The first screening information of the rice object is generated according to the minimum value of the remote sensing feature of the rice object and the average value of the remote sensing feature of each identified object in the target area.
[0020] In some feasible solutions, the method for determining the distribution information of rice crops and other crops in the target area includes:
[0021] According to the growth characteristic information of rice in the target area and in combination with the remote sensing characteristic information of the target area, a remote sensing characteristic change graph of rice in the target area is determined;
[0022] Producing second screening information of rice objects according to the rice remote sensing characteristic change graph in the target area;
[0023] According to the second screening information of the rice object, the first target object in the target area that meets the first screening information is screened for the second time to determine the rice crops in the target area.
[0024] In some feasible solutions, the method for preparing the second screening information of the rice object includes:
[0025] Determining the growth characteristics of at least two intervals of rice according to the growth characteristic information of rice in the target area;
[0026] According to the growth characteristics of at least two intervals of rice, respectively determine the remote sensing characteristic change graph of each interval of rice;
[0027] According to the remote sensing characteristic change graph of each section of rice, respectively determine the remote sensing characteristic change graph characteristics of each section of rice;
[0028] The second screening information of the rice object is produced according to the graphic features of the remote sensing feature changes of each interval section of the rice.
[0029] In some feasible solutions, the method for determining the distribution information of rice crops and other crops in the target area further includes:
[0030] The development stage of rice is divided into five different stages: bare soil stage, sowing stage, transplanting stage, growth stage, and maturity stage;
[0031] Among them, rice is the only crop that needs a lot of water during sowing and transplanting. The remote sensing characteristics of rice in the sowing-transplanting period and the remote sensing characteristics of rice in the transplanting-growing period are obtained respectively;
[0032] According to the remote sensing characteristics of rice in the sowing-transplanting period and the remote sensing characteristics of rice in the transplanting-growing period, the remote sensing characteristic graphic characteristics of rice in the sowing-transplanting period and the remote sensing characteristic graphic characteristics of rice in the transplanting-growing period are determined respectively;
[0033] According to the remote sensing characteristic graphic features of rice in the sowing-transplanting period and the remote sensing characteristic graphic features of rice in the transplanting-growing period, the first target objects in the target area that meet the first screening information are respectively subjected to secondary screening to determine the rice crops in the target area.
[0034] In some feasible solutions, the mapping method further includes:
[0035] According to the growth characteristic information of rice in the target area and in combination with the remote sensing characteristics of rice objects at different periods, at least a remote sensing characteristic set of rice in the target area at one period is determined;
[0036] Determine at least one segmentation unit plot in the target area according to the geomorphic segmentation image of the target area;
[0037] According to at least one period remote sensing feature set of rice in the target area, combined with the landform segmentation image of the target area, the probability of rice existence in each segmentation unit plot in the target area is determined, and the probability of rice in each segmentation unit plot in the target area is obtained respectively;
[0038] According to the probability of rice in each segmented unit plot in the target area, the image of the rice plot in the target area is output;
[0039] According to the rice plot image in the target area and combined with the landform segmentation image of the target area, a rice image model of the target area is produced.
[0040] In some feasible solutions, the method for obtaining the probability of rice in each segmented unit plot in the target area includes:
[0041] Assume that there is rice in the target area The remote sensing feature set of the period, at this time, the output probability image of rice plots in the target area should be , then:
[0042] ,
[0043] in, It refers to the probability that the segmentation unit plot i is classified as rice; Refers to the number of probability images of rice plots in the target area; It refers to the possibility that segmentation unit plot i is classified as rice in rice plot probability image j.
[0044] In a second aspect, the present invention provides a fully automatic rice precision mapping system for cloudy and rainy areas, which adopts a fully automatic rice precision mapping method for cloudy and rainy areas as described in any one of the first aspects.
[0045] In a third aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein when the program is executed by a processor, the program implements a fully automatic rice precision mapping method for cloudy and rainy areas as described in any one of the first aspects.
[0046] The beneficial effects of the present invention are:
[0047] The present invention adopts a method combining optical imaging, remote sensing technology and model construction, which can, on the one hand, detect the overall distribution of rice planning in cloudy and rainy areas, and on the other hand, build a model for the cloudy and rainy areas in a refined manner, effectively meeting the needs of refined and accurate monitoring. That is, it effectively solves the shortcomings of the prior art that cannot meet the needs of refined and accurate monitoring of rice crops in cloudy and rainy areas. BRIEF DESCRIPTION OF THE DRAWINGS
[0048] Figure 1 It is a schematic diagram of the overall process of a fully automatic rice precision mapping method in cloudy and rainy areas provided in an embodiment of the present invention;
[0049] Figure 2 A schematic diagram of a landform segmentation image of a fully automatic rice precision mapping method in a cloudy and rainy area provided in an embodiment of the present invention;
[0050] Figure 3 This is a partial flow chart of a method for fully automatic rice precision mapping in cloudy and rainy areas provided in an embodiment of the present invention. DETAILED DESCRIPTION
[0051] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0052] It should be noted that all directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present invention are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.
[0053] In the present invention, unless otherwise clearly specified and limited, the terms "connection", "fixation", etc. should be understood in a broad sense. For example, "fixation" can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements, unless otherwise clearly defined. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.
[0054] In addition, if there are descriptions involving "first", "second", etc. in the embodiments of the present invention, the descriptions of "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the meaning of "and / or" appearing in the full text includes three parallel schemes. Taking "A and / or B" as an example, it includes scheme A, or scheme B, or a scheme that satisfies both A and B. In addition, the technical solutions between the various embodiments can be combined with each other, but it must be based on the ability of ordinary technicians in the field to implement. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.
[0055] Example: Refer to Figures 1 to 3 In order to solve the shortcomings of the prior art that cannot meet the demand for refined and accurate monitoring of rice crops in cloudy and rainy areas, the present invention provides a method for fully automatic rice precision mapping in cloudy and rainy areas. The mapping method adopts a method that combines optical imaging, remote sensing technology and model construction. On the one hand, it can detect the overall distribution of rice planning in cloudy and rainy areas, and on the other hand, it can construct a model for the cloudy and rainy areas in a refined manner, effectively meeting the demand for refined and accurate monitoring. That is, it effectively solves the shortcomings of the prior art that cannot meet the demand for refined and accurate monitoring of rice crops in cloudy and rainy areas.
[0056] In this embodiment, in order to facilitate understanding of how to use optical imaging, remote sensing technology and model construction to combine with each other to carry out fine mapping of rice distribution in cloudy and rainy areas, the following description is made here. Specifically, the mapping method includes: obtaining a scanning image of the target area through an unmanned aerial vehicle or a radar, and making a contour model of the target area according to the obtained scanning image. In addition, in order to ensure the accuracy of the acquisition of the target area, the geomorphic features of the target area can also be combined, and then the geomorphic segmentation image of the target area is determined (such as subdividing the target area into a flat area, a hillside area, and a forest and grass area, etc.);
[0057] In a preferred embodiment, the present invention can use the SAM (Segment Anything Model) model to obtain the farmland plot boundaries of the target area as the minimum basic unit for automatic generation and mapping of rice samples. However, the key feature of SAM is the prompt-based visual Transformer (ViT) model, which is trained on a visual dataset SA-1B containing more than 1 billion masks from 11 million images, and can segment any target on a given image, including plot targets. This capability makes SAM a basic model in the field of vision and can also generate application value in fields beyond vision. However, since the ViT-H image encoder in SAM has 632M parameters (the prompt-based decoder only needs 387M parameters), the actual computational and memory costs of using SAM to perform any segmentation task are very high, which is challenging for real-time applications. At the same time, SAM is not designed for farmland plots, and high-resolution remote sensing data needs to be used to fine-tune SAM. In response to this problem, SAM is optimized, and the main steps are as follows:
[0058] 1. Data preprocessing. Since the image encoder module of SAM has been well trained, this module does not need to be trained. We scale the image to 1024 by the long side, and pad it to 1024 if the short side is not enough. Then convert it into a torch tensor and call the model image encode to extract image features. 2. Model training. We fine-tune the SAM data based on the cultivated land data. First, load the model sam_vit_b_01ec64.pth, create an optimizer, redefine the binary classification loss (Formula 1), and fine-tune the model using our cultivated land sample data. The optimized reconstruction loss function is as follows:
[0059] (Formula 1),
[0060] Where L is the loss function, are the weight parameters of the loss function, N is the total number of samples, Indicates the label of the sample, the positive class is 1 and the negative class is 0. Represents the probability that sample i is predicted as a positive class. 3. Results show that after fine-tuning, SAM has a good effect on the segmentation of high-resolution remote sensing images. Figure 2 As shown. That is, the fine and complete boundary lines of cultivated land plots are finally obtained and independent cultivated land plot surface vector objects are constructed, providing a "fine" spatial base for the next step of accurate mapping of rice in cloudy and rainy areas. That is, the image can be segmented according to the topography of the target area, and the remote sensing feature information of the target area can be obtained respectively by using SAR high remote sensing technology, and the remote sensing feature information of the target area can be analyzed to determine the distribution information of crop objects and non-crop objects in the target area; that is, the distribution information of crop objects and non-crop objects in the target area can be determined by initial screening based on the special values and change trends in the remote sensing feature information of the target area. Then, the growth feature information of rice in the target area can be obtained, the remote sensing feature transformation curve of rice in the target area can be made, and the distribution information of rice crops and other crops in the target area can be determined; that is, according to the characteristics of the growth cycle of rice in the target area, the targets after the initial screening are screened for the second time to screen out the objects that best meet the characteristics of rice. Then, according to the distribution information of the crop objects and non-crop objects in the target area and / or the distribution information of the rice crops and other crops in the target area, and in combination with the landform segmentation image of the target area, a rice distribution image of the target area is produced.
[0061] In this embodiment, in order to facilitate understanding of how to use the remote sensing feature information of the target area to process and distinguish the distribution information of the crop objects and non-crop objects in the target area, the following description is given. Specifically, the method for determining the distribution information of the crop objects and non-crop objects in the target area includes:
[0062] According to the geomorphic features of the target area and the geomorphic segmentation image of the target area, the identification objects existing in the target area are determined; that is, the identification objects that may exist in the target area are pre-determined according to the scanned image, and then the remote sensing feature information of each identification object in the target area is obtained respectively according to the remote sensing feature information of the target area; then the remote sensing features of the rice object at different periods are obtained, and the first screening information of the rice object is prepared in combination with the remote sensing feature information of each identification object in the target area; then, according to the first screening information of the rice object, the remote sensing feature information of each identification object in the target area is used to determine at least one first target object that meets the first screening information in the target area.
[0063] Specifically, the method for preparing the first screening information of the rice object includes:
[0064] According to the remote sensing characteristics of the rice object at different periods, the minimum value of the remote sensing characteristics of the rice object is determined; the remote sensing characteristic information of each identified object in the target area under at least two different time segments is obtained to determine the average value of the remote sensing characteristics of each identified object in the target area; according to the minimum value of the remote sensing characteristics of the rice object and the average value of the remote sensing characteristics of each identified object in the target area, the first screening information of the rice object is produced. That is, in this embodiment, since the remote sensing characteristics of rice (such as the VH backscattering degree) are the minimum in the flooding period and the maximum in the heading period, at the same time, the VH backscattering intensity of water areas and aquaculture is low throughout the year, which is significantly lower than the minimum backscattering intensity of rice. On the contrary, the VH backscattering intensity of urban areas and natural vegetation is always high, generally exceeding the backscattering intensity observed throughout the rice growing season. Using these differences, we developed simple rules to distinguish crops from non-crop objects. Specifically, water bodies and aquaculture with low VH backscatter intensities throughout the year were excluded by the average VH backscatter value; urban areas and natural vegetation characterized by high backscatter intensities were excluded by the minimum VH backscatter value during the rice growing season, calculated as follows:
[0065] (Formula 2),
[0066] in, and They represent the median value of VH throughout the year and the minimum value of VH during the entire rice growing period, respectively.
[0067] The determination of T1 and T2 values follows a simple statistical method based on the backscatter intensity of natural vegetation and temporary water bodies. First, the rule (NDVImax > 0.3) is used to detect natural vegetation objects, while the rule (NDVImax > 0.4, NDWImax > 0.3) is used to detect temporary water bodies; then the T1 value is calculated as the 50th percentile of the annual average backscatter intensity of all temporary water bodies, while the T2 value is determined as the 25th percentile of the annual minimum backscatter intensity of all natural vegetation objects.
[0068] In this embodiment, in order to facilitate understanding of how to distinguish rice crops from other crops in the target area, the following description is given here. Specifically, the method for determining the distribution information of rice crops and other crops in the target area includes: determining the remote sensing characteristic change graph of rice in the target area according to the growth characteristic information of rice in the target area and in combination with the remote sensing characteristic information of the target area; then, according to the remote sensing characteristic change graph of rice in the target area, making the second screening information of rice objects; then, according to the second screening information of rice objects, performing a second screening on the first target objects in the target area that meet the first screening information, and determining the rice crops in the target area. Specifically, the method for making the second screening information of rice objects includes: determining the growth characteristics of at least two intervals of rice according to the growth characteristic information of rice in the target area; determining the remote sensing characteristic change graph of each interval of rice according to the growth characteristics of at least two intervals of rice; determining the remote sensing characteristic change graph characteristics of each interval of rice according to the remote sensing characteristic change graph of each interval of rice; making the second screening information of rice objects according to the remote sensing characteristic change graph characteristics of each interval of rice. More specifically, assume that the development stage of rice is divided into five different stages: bare soil stage, sowing stage, transplanting stage, growth stage, and maturity stage; among them, rice is the only crop that requires a lot of water during sowing and transplanting, and the remote sensing characteristics of rice in the sowing-transplanting stage and the remote sensing characteristics of rice in the transplanting-growth stage are obtained respectively; according to the remote sensing characteristics of rice in the sowing-transplanting stage and the remote sensing characteristics of rice in the transplanting-growth stage, the remote sensing characteristic graphic characteristics of rice in the sowing-transplanting stage and the remote sensing characteristic graphic characteristics of rice in the transplanting-growth stage are determined respectively; then, according to the remote sensing characteristic graphic characteristics of rice in the sowing-transplanting stage and the remote sensing characteristic graphic characteristics of rice in the transplanting-growth stage, the first target object in the target area that meets the first screening information is screened for the second time to determine the rice crop in the target area. That is, as the only crop that requires a lot of water during transplanting, rice shows obvious "soil-water" and "water-plant" characteristic signals before and after transplanting, which is shown as a "V" shape (i.e., first lagging and then rising) in the VH backscattering time series. Therefore, the present invention establishes the rule of "V"-shaped characteristic phenological constraint in Sentinel-1 VH backscattering time series, and the calculation formula is as follows:
[0069] (Formula 3)
[0070] in, refers to the slope of the VH backscatter time series during the sowing-transplanting period; Refers to the slope of the VH backscatter time series during the transplanting-growth period.
[0071] In this embodiment, in order to ensure the accuracy of rice mapping in the target area, the mapping method further includes:
[0072] According to the growth characteristic information of rice in the target area and in combination with the remote sensing characteristics of rice objects at different periods, at least one period remote sensing feature set of rice in the target area is determined (such as three vegetation indices including normalized difference vegetation index (NDVI), enhanced vegetation index (EVI) and land surface water index (LSWI) and two spectral bands of red edge 2 (RE2) and shortwave infrared 2 (SWIR2), and the radar feature is one or more combinations of VH); then, according to the landform segmentation image of the target area, at least one segmentation unit plot in the target area is determined; then, according to at least one period remote sensing feature set of rice in the target area and in combination with the landform segmentation image of the target area, the probability of rice in each segmentation unit plot in the target area is determined, and the probability of rice in each segmentation unit plot in the target area is obtained respectively; then, according to the rice probability of each segmentation unit plot in the target area, the rice plot image in the target area is output; then, according to the rice plot image in the target area and in combination with the landform segmentation image of the target area, a rice image model of the target area is prepared.
[0073] In some specific feasible means, an optical feature set and a radar feature set can be produced every month during the rice growing season, and a total of 10 feature sets are obtained. They are respectively input into the recurrent neural network artificial intelligence classification model for training, and rice mapping is performed with plot objects as units. The model outputs 10 rice probability maps. By calculating the average probability of each plot being identified as rice, plots with an average probability of being identified as rice greater than 0.5 are considered to be rice, and the final rice map is obtained. That is, the method for obtaining the probability of rice in each segmented unit plot in the target area includes:
[0074] Assume that there is rice in the target area The remote sensing feature set of the period, at this time, the output probability image of rice plots in the target area should be , then:
[0075] (Formula 4),
[0076] in, It refers to the probability that the segmentation unit plot i is classified as rice; Refers to the number of probability images of rice plots in the target area; It refers to the possibility that segmentation unit plot i is classified as rice in rice plot probability image j.
[0077] Finally, the accuracy of the classification results can be evaluated using validation samples independent of the training samples. The accuracy indicators include overall accuracy, user accuracy, mapping accuracy, and F1 score.
[0078] (Formula 5),
[0079] (Formula 6),
[0080] (Formula 7),
[0081] (Formula 8),
[0082] Among them, OA is used to evaluate the overall accuracy of the classification results. Indicates the number of samples that are correctly classified (i.e., the predicted category is consistent with the true category). Indicates the number of all samples involved in the validation. For a specific crop type, the accuracy is evaluated by PA, UA and F1. Represents the number of samples correctly identified as this type. This means that the type is incorrectly identified as another type (missed). Indicates that other types are mistakenly identified as this type (misclassification).
[0083] In a second aspect, the present invention provides a fully automatic rice precision mapping system for cloudy and rainy areas, which adopts a fully automatic rice precision mapping method for cloudy and rainy areas as described in any one of the first aspects.
[0084] In some embodiments, the mapping system can communicate using any currently known or future developed network protocol such as HTTP (Hyper Text Transfer Protocol), and can be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.
[0085] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.
[0086] In a third aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the program is executed by a processor to implement a method for fully automatic rice precision mapping in a cloudy and rainy area as described in any one of the first aspects. The computer-readable medium in this embodiment can be written in one or more programming languages or a combination thereof to write computer program codes for performing the operations of some embodiments of the present application, wherein the programming languages include object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as "C" language or similar programming languages. The program code can be executed entirely on a user's computer, partially on a user's computer, as an independent software package, partially on a user's computer, partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through 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., using an Internet service provider to connect through the Internet).
[0087] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.
[0088] In particular, according to some embodiments of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, some embodiments of the present application include a computer program product, which includes a computer program carried on a computer readable medium, and the computer program includes a program code for executing the method shown in the flowchart.
[0089] The fourth aspect of the present invention provides an electronic device, comprising: one or more processors; a storage device having one or more programs stored thereon; when the one or more programs are executed by the one or more processors, the one or more processors implement a method for fully automatic precision mapping of rice in cloudy and rainy areas as described in the first aspect. The above-mentioned computer-readable medium may be included in the above-mentioned electronic device; or it may exist separately, that is, it is not assembled into the electronic device. The above-mentioned computer-readable medium carries one or more programs, and when the above-mentioned one or more programs are executed by the electronic device, the electronic device can implement a method for fully automatic precision mapping of rice in cloudy and rainy areas as described in the first aspect.
[0090] A fifth aspect of the present invention provides a computer program product, comprising a computer program, which, when executed by a processor, implements a fully automatic rice precision mapping method for cloudy and rainy areas as described in the first aspect.
[0091] The above description is only some preferred embodiments of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present application is not limited to the technical solutions formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above invention concept. For example, the above features are replaced with (but not limited to) technical features with similar functions disclosed in the embodiments of the present application to form a technical solution.
Claims
1. A fully automatic rice precision mapping method in cloudy and rainy areas, characterized in that: Bag include: Obtain a scanned image of the target area, make a contour model of the target area, and determine a landform segmentation image of the target area in combination with the landform features of the target area; According to the landform segmentation image of the target area, remote sensing feature information of the target area is obtained respectively, and the remote sensing feature information of the target area is analyzed to determine the distribution information of the crop objects and non-crop objects in the target area; Obtain the growth characteristic information of rice in the target area, make the remote sensing characteristic transformation curve of rice in the target area, and determine the distribution information of rice crops and other crops in the target area; Producing a rice distribution image of the target area based on distribution information of crop objects and non-crop objects in the target area and / or distribution information of rice crops and other crops in the target area, and in combination with a landform segmentation image of the target area; Specifically, the method for determining the distribution information of crop objects and non-crop objects in the target area includes: Determine the identification objects existing in the target area according to the geomorphic features of the target area and the geomorphic segmentation image of the target area; According to the remote sensing feature information of the target area, remote sensing feature information of each identified object in the target area is obtained respectively; Acquire remote sensing features of rice objects at different periods, and combine remote sensing feature information of each identified object in the target area to produce first screening information of the rice objects; According to the first screening information of the rice object, for the remote sensing feature information of each identified object in the target area, at least one first target object that meets the first screening information is determined in the target area; Wherein, the first screening information of the rice object includes: the minimum value of the remote sensing feature of the rice object and the average value of the remote sensing feature of each identified object in the target area; Specifically, the method for determining the distribution information of rice crops and other crops in the target area includes: According to the growth characteristic information of rice in the target area and in combination with the remote sensing characteristic information of the target area, a remote sensing characteristic change graph of rice in the target area is determined; Producing second screening information of rice objects according to the rice remote sensing characteristic change graph in the target area; According to the second screening information of the rice object, the first target object in the target area that meets the first screening information is screened for the second time to determine the rice crop in the target area; The method for preparing the second screening information of the rice object includes: Determining the growth characteristics of at least two intervals of rice according to the growth characteristic information of rice in the target area; According to the growth characteristics of at least two intervals of rice, respectively determine the remote sensing characteristic change graph of each interval of rice; According to the remote sensing characteristic change graph of each section of rice, respectively determine the remote sensing characteristic change graph characteristics of each section of rice; The second screening information of the rice object is produced according to the graphic features of the remote sensing feature changes of each interval section of the rice.
2. The method for fully automatic rice precision mapping in cloudy and rainy areas according to claim 1, characterized in that: The method for preparing the first screening information of the rice object comprises: According to the remote sensing characteristics of the rice object at different periods, the minimum value of the remote sensing characteristics of the rice object is determined; Acquire remote sensing feature information of each identified object in the target area in at least two different time series, and determine an average value of the remote sensing feature of each identified object in the target area; The first screening information of the rice object is generated according to the minimum value of the remote sensing feature of the rice object and the average value of the remote sensing feature of each identified object in the target area.
3. The method for fully automatic rice precision mapping in cloudy and rainy areas according to claim 2 is characterized in that: The method for determining the distribution information of rice crops and other crops in the target area also includes: The development stage of rice is divided into five different stages: bare soil stage, sowing stage, transplanting stage, growth stage, and maturity stage; Among them, rice is the only crop that needs a lot of water during sowing and transplanting. The remote sensing characteristics of rice in the sowing-transplanting period and the remote sensing characteristics of rice in the transplanting-growing period are obtained respectively; According to the remote sensing characteristics of rice in the sowing-transplanting period and the remote sensing characteristics of rice in the transplanting-growing period, the remote sensing characteristic graphic characteristics of rice in the sowing-transplanting period and the remote sensing characteristic graphic characteristics of rice in the transplanting-growing period are determined respectively; According to the remote sensing characteristic graphic features of rice in the sowing-transplanting period and the remote sensing characteristic graphic features of rice in the transplanting-growing period, the first target objects in the target area that meet the first screening information are respectively subjected to secondary screening to determine the rice crops in the target area.
4. The method for fully automatic rice precision mapping in cloudy and rainy areas according to claim 3, characterized in that: The mapping method further comprises: According to the growth characteristic information of rice in the target area and in combination with the remote sensing characteristics of rice objects at different periods, at least a remote sensing characteristic set of rice in the target area is determined; Determine at least one segmentation unit plot in the target area according to the geomorphic segmentation image of the target area; According to at least one period remote sensing feature set of rice in the target area, combined with the landform segmentation image of the target area, the probability of rice existence in each segmentation unit plot in the target area is determined, and the probability of rice in each segmentation unit plot in the target area is obtained respectively; According to the probability of rice in each segmented unit plot in the target area, a probability image of rice plots in the target area is output; According to the probability image of rice plots in the target area and combined with the landform segmentation image of the target area, a rice image model of the target area is produced.
5. The method for fully automatic rice precision mapping in cloudy and rainy areas according to claim 4, characterized in that: The method for obtaining the probability of rice in each segmentation unit plot in the target area comprises: Assume that there is rice in the target area The remote sensing feature set of the period, at this time, the output probability image of rice plots in the target area should be , then: , in, It refers to the probability that the segmentation unit plot i is classified as rice; Refers to the number of probability images of rice plots in the target area; It refers to the possibility that segmentation unit plot i is classified as rice in rice plot probability image j.
6. A fully automatic rice precision mapping system for cloudy and rainy areas, characterized in that: A fully automatic rice precision mapping method for cloudy and rainy areas as described in any one of claims 1 to 5 is adopted.
7. A computer-readable storage medium having a computer program stored thereon, characterized in that: in, When the program is executed by the processor, a fully automatic rice precision mapping method for cloudy and rainy areas is implemented as described in any one of claims 1 to 5.
Citation Information
Patent Citations
Integrated remote sensing identification method for multi-planting-mode paddy field
CN117690024A
Remote sensing assessment system of disaster areas of crops caused by hail
CN106126920A
Rice recognition method based on multi-temporal multi-source remote sensing data
CN109345555A
Phonological knowledge fused remote sensing rice mapping method and application thereof
CN112215090A
Rice yield prediction method, device and equipment based on remote sensing image
CN114612896A
