CN115082803B

CN115082803B - Cultivated land abandoned land monitoring method and device based on vegetation season change and storage medium

Info

Publication number: CN115082803B
Application number: CN202210995735.7A
Authority: CN
Inventors: 杨颖频; 肖文菊; 吴志峰
Original assignee: Guangzhou University
Current assignee: Guangzhou University
Priority date: 2022-08-19
Filing date: 2022-08-19
Publication date: 2023-02-03
Anticipated expiration: 2042-08-19
Also published as: CN115082803A

Abstract

Embodiments of the present disclosure provide a method, an apparatus, and a storage medium for monitoring abandoned cultivated land based on vegetation season changes, the method including: aiming at an area to be detected, acquiring a abandoned land sample feature set and a non-abandoned land sample feature set by acquiring a normalized vegetation index NDVI time sequence curve; determining a abandoned land criterion of the area to be detected based on the abandoned land sample feature set and the non-abandoned land sample feature set; and identifying the abandoned land in the area to be detected according to the abandoned land criterion. The technical scheme provided by the application is used for solving the problem that the quaternary phase change characteristics of a abandoned land are not sufficiently considered in the prior art.

Description

Cultivated land abandoned land monitoring method and device based on vegetation season change and storage medium

Technical Field

The document relates to the field of terrestrial information science, in particular to a cultivated land abandoned monitoring method and device based on vegetation season phase change and a storage medium.

Background

The prior cultivated land abandoned land monitoring based on remote sensing technology can be summarized into the following three major methods: 1) A multi-temporal cultivated land range comparison method, which identifies abandoned lands by comparing the distribution range of cultivated lands between two temporal phases, wherein the data of the cultivated land distribution range is usually derived from land cover remote sensing classification results or homeland survey data; 2) A supervised classification method, which is to collect samples of abandoned land and main crops, extract spectral characteristics, texture characteristics and the like of abandoned land and non-abandoned land by using remote sensing images, train a machine learning model and distinguish the abandoned land and the non-abandoned land on the images; 3) A time sequence analysis method comprises the steps of constructing a normalized vegetation index (NDVI) time sequence curve by utilizing optical images of a plurality of time phases, establishing a abandoned land identification rule by analyzing the similarity of main crops and the NDVI time sequence curve of a abandoned land or comparing the characteristics of the two time sequence curves, such as peak value, value range and the like, and monitoring the abandoned land condition by a threshold segmentation method.

However, the previous research on the characteristics of the change of the secondary phase of a abandoned land is not sufficient, and the method is difficult to be applied to abandoned land areas with complex crop planting structures, such as southwest and south China mountain areas and the like.

Disclosure of Invention

In view of the above analysis, the present application is directed to a abandoned land monitoring method, device and storage medium based on vegetation season changes, so as to solve the problem of insufficient consideration of the season change characteristics of abandoned lands in the prior art.

In a first aspect, one or more embodiments of the present specification provide a cultivated land abandoned monitoring method based on vegetation season change, including:

aiming at an area to be detected, acquiring a abandoned land sample characteristic set and a non-abandoned land sample characteristic set by acquiring a normalized vegetation index NDVI time sequence curve;

determining a abandoned land criterion of the area to be detected based on the abandoned land sample feature set and the non-abandoned land sample feature set;

and identifying the abandoned land in the area to be detected according to the abandoned land criterion.

Further, aiming at the area to be detected, a abandoned land sample feature set and a non-abandoned land sample feature set are obtained by collecting a normalized vegetation index NDVI time sequence curve, and the method comprises the following steps:

determining a plurality of abandoned lands and non-abandoned lands in the area to be detected;

respectively collecting the NDVI time sequence curve of each abandoned land and the NDVI time sequence curve of each non-abandoned land;

determining an NDVI change value corresponding to each abandoned land and an NDVI change value corresponding to each non-abandoned land according to the NDVI time sequence curve;

determining the non-abandoned land sample feature set according to the NDVI change value corresponding to each non-abandoned land;

determining the abandoned land sample feature set according to the NDVI change value corresponding to each abandoned land. Further, determining an NDVI change value for each abandoned land and an NDVI change value for each non-abandoned land based on the NDVI timing curve comprises:

determining a maximum NDVI and a minimum NDVI of the abandoned land's NDVI timing curve;

a difference between a maximum NDVI and a minimum NDVI of the abandoned land's NDVI timing curve is an NDVI change value corresponding to the abandoned land;

determining a maximum NDVI and a minimum NDVI of the non-abandoned land NDVI timing curve;

the difference between the maximum NDVI and the minimum NDVI of the NDVI timing curve of the non-abandoned land is the NDVI variation value corresponding to the non-abandoned land.

Further, determining a abandoned land criterion of the area to be detected based on the abandoned land sample feature set and the non-abandoned land sample feature set includes:

determining a maximum value of a abandoned land NDVI (dimensional change in the abandoned land NDVI) as an initial value according to the abandoned land sample feature set;

determining an accuracy rate and a recall rate for identifying the abandoned land based on the abandoned land quantity, the abandoned land sample feature set and the non-abandoned land sample feature set, wherein the accuracy rate is used for representing a proportion of a practical abandoned land in the sample judged as a abandoned land, and the recall rate is used for representing a proportion of a correct abandoned land judged as a abandoned land;

determining F1 according to the precision rate and the recall rate corresponding to the initial value, wherein the F1 is a numerical value used for representing the incidence relation between the precision rate and the recall rate;

and iterating the initial value and the F1 to obtain the abandoned land criterion.

Further, constructing a abandoned land criterion based on the precision rate and recall rate corresponding to the initial value comprises:

reducing the initial value according to a preset step length, and obtaining a new accuracy rate and a new recall rate;

determining a new F1 according to the new precision rate and the new recall rate;

repeating the process until more than half of the abandoned land sample with the concentrated characteristics can be correctly identified as the abandoned land;

determining a abandoned land NDVI change value corresponding to the maximum value of the F1 as a abandoned land judgment value;

and constructing a abandoned land criterion according to the abandoned land judgment value.

Further, the correlation between the precision rate and the recall rate is determined using the following formula:

wherein precision represents accuracy, call represents recall, and F1 represents the incidence relation between the accuracy and the recall.

Further, the abandoned land criterion is specifically as follows:

determining the abandoned land to be identified as a abandoned land when the change value of the abandoned land NDVI of the abandoned land to be identified is less than the abandoned land determination value.

Further, when the change value of the abandoned land NDVI of the land to be identified is not less than the abandoned land determination value, determining that the land to be identified is a non-abandoned land.

Further, identifying a abandoned land in the area to be detected according to the abandoned land criterion comprises:

determining each plot to be identified in the areas to be detected;

obtaining the NDVI curve of each land to be identified;

determining a change value of the NDVI curve of each to-be-identified plot according to the NDVI curve of each to-be-identified plot;

and respectively determining whether each land to be identified is abandoned land according to the abandoned land criterion and the change value of the NDVI curve of each land to be identified.

In a second aspect, an embodiment of the present application provides a abandoned land monitoring device based on vegetation season change, including: the device comprises an acquisition module, a criterion creation module and a data processing module;

the collection module is used for collecting a normalized vegetation index NDVI time sequence curve aiming at an area to be detected to obtain a abandoned land sample characteristic set and a non-abandoned land sample characteristic set;

the criterion creating module is used for determining a abandoned land criterion of the area to be detected based on the abandoned land sample feature set and the non-abandoned land sample feature set;

the data processing module is used for identifying the abandoned land in the area to be detected according to the abandoned land criterion.

In a second aspect, an embodiment of the present application provides a storage medium, including:

for storing computer-executable instructions that, when executed, implement the following flow:

determining a abandoned land criterion of the area to be detected by constructing an NDVI time sequence curve and extracting NDVI change value characteristics based on the abandoned land sample characteristic set and the non-abandoned land sample characteristic set;

Compared with the prior art, the application can at least realize the following technical effects:

the NDVI time sequence curve reflects the change of the vegetation growth state and the vegetation coverage degree along with time, so that the NDVI time sequence curve is used for extracting the NDVI change value characteristics, the abandoned land and non-abandoned land sample characteristic set of the area to be detected is obtained, a abandoned land criterion is constructed, and the abandoned land and the non-abandoned land can be well distinguished. By the method, the vegetation season change characteristics of the abandoned land and non-abandoned land are digitalized, so that the applicability of abandoned land detection is improved.

Drawings

In order to more clearly illustrate one or more embodiments or prior art solutions of the present specification, the drawings that are needed in the description of the embodiments or prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments described in the present specification, and that other drawings can be obtained by those skilled in the art without inventive exercise.

Fig. 1 is a flowchart of a cultivated land abandoned monitoring method based on vegetation season change according to one or more embodiments of the present disclosure;

FIG. 2 is a graphical illustration of the change in NDVI curves provided in one or more embodiments of the present description;

fig. 3 is a timing diagram of a abandoned and non-abandoned NDVI timing diagram provided by one or more embodiments of the present disclosure;

FIG. 4 is a graph illustrating an amplitude histogram of an NDVI timing curve provided in one or more embodiments of the present disclosure;

fig. 5 is a schematic diagram of a calculation result of F1 provided in one or more embodiments of the present disclosure.

Detailed Description

In order to make those skilled in the art better understand the technical solutions in one or more embodiments of the present disclosure, the technical solutions in one or more embodiments of the present disclosure will be clearly and completely described below with reference to the drawings in one or more embodiments of the present disclosure, and it is obvious that the described embodiments are only a part of the embodiments of the present disclosure, and not all embodiments. All other embodiments that can be derived by a person skilled in the art from one or more of the embodiments described herein without making any inventive step shall fall within the scope of protection of this document.

In southwest and south China, crops are various in planting type and complex in cooking structure, the characteristic of different crops usually has great difference, abandoned years of different fields are not the same, the NDVI level of a single-year or seasonal abandoned land is lower, and a abandoned land for many years is covered by weeds at a high level, the NDVI reaches a higher level, the image is equivalent to the NDVI value of a crop in a vigorous growth period, great difficulty is brought to abandoned land identification, and a abandoned land identification method based on the NDVI time sequence curve peak value is difficult to apply.

Based on the above scenario, the embodiment of the application provides a abandoned land monitoring method based on vegetation season change, which includes the following steps:

step 1, aiming at an area to be detected, a abandoned land sample characteristic set and a non-abandoned land sample characteristic set are obtained by collecting a normalized vegetation index NDVI time sequence curve.

In the embodiment of the present application, generally, one area to be detected is an administrative unit, for example, province, city, county. Therefore, a plurality of plots to be identified exist in the area to be detected. Selecting a plurality of confirmed abandoned land and non-abandoned land from a plurality of land parcels to be identified. And then collecting the NDVI curves of the abandoned land and the non-abandoned land, and preparing a abandoned land sample characteristic set and a non-abandoned land sample characteristic set.

Specifically, step 1 comprises:

determining a plurality of abandoned lands and non-abandoned lands in the area to be detected, and respectively collecting the NDVI time sequence curves of the abandoned lands and the non-abandoned lands.

And determining the NDVI change value corresponding to each abandoned land and the NDVI change value corresponding to each non-abandoned land according to the NDVI time sequence curve. The NDVI variation determination method is shown in fig. 2, and determines the maximum NDVI and the minimum NDVI of the abandoned NDVI timing curve according to the NDVI timing curve. The difference between the maximum and minimum NDVI of a abandoned NDVI timing curve is the corresponding NDVI variation value of the abandoned land. Similarly, the maximum NDVI and the minimum NDVI of the non-abandoned NDVI timing curve are determined according to the NDVI timing curve. The difference between the maximum and minimum NDVI of an NDVI timing curve of an unprovoked area is the change in the corresponding NDVI of the unprovoked area.

It should be noted that because of the processes of seeding, growing, harvesting and re-planting of crops, the annual NDVI varies significantly from weeds in abandoned land. In a abandoned land, although weeds wither at all seasons, the change in NDVI is not significant compared to a non-abandoned land because no manual intervention is involved and the plants in the abandoned land are not completely removed. In addition, for land abandoned for many years, the NDVI may reach the peak value of the NDVI of non-abandoned land, but the change value of the NDVI is still not obvious. Specifically, as shown in fig. 3, waste land 1 is a multi-year waste land, and waste land 2 is a less than one year old waste land. In fig. 3, either the NDVI change level of abandoned land 1 or abandoned land 2 is less than that of a non-abandoned land. This is seen. The method can effectively distinguish abandoned land from non-abandoned land by using the change value of NDVI.

And determining a non-abandoned land sample feature set according to the corresponding NDVI change value of each non-abandoned land. Determining a abandoned land sample feature set according to the NDVI change value corresponding to each abandoned land. Wherein the non-abandoned land sample feature set is used for representing the data distribution of the change value of non-abandoned land NDVI, and the abandoned land sample feature set is used for representing the data distribution of the change value of abandoned land NDVI.

The abandoned sample feature set preferably preprocesses the NDVI timing curve: and extracting cloud pollution pixels by using a quality control waveband, and removing cloud pollution time phase points on the NDVI time sequence curve. And judging the point on the NDVI time sequence curve where the NDVI reduction amplitude within 14 days exceeds 0.2 or the NDVI reduction amplitude within 21 days exceeds 0.5 as an abnormal value point, and removing the point from the NDVI time sequence curve. And finally, performing linear interpolation on the NDVI time sequence curve to fill missing values caused by abnormal observation.

In the embodiment of the present application, an NDVI curve collected by Sentinel-2 with a resolution of 10 meters can be selected, and thus, a abandoned land and a non-abandoned land can be accurately distinguished.

Preferably, a abandoned land and a non-abandoned land are selected as the sample based on the species of the area to be detected. For example, if 10 crops are frequently grown in the area to be detected, then the species to which the sample is not abandoned needs to contain these 10 crops. At the same time, abandoned land needs to contain weeds frequently appearing in the area to be detected as much as possible. Because the species of the area to be detected cannot be changed in a short time, the change is only limited to a small scale, and the accuracy of the criterion cannot be influenced, so that the abandoned land criterion has better applicability and can be used for a long time. If the species in the area to be detected changes obviously, only new species need to be added into the sample and old species need to be eliminated, so that the abandoned land criterion can be finely adjusted, and the applicability of the abandoned land criterion can be further ensured.

And 2, determining a abandoned land criterion of the area to be detected based on the abandoned land sample characteristic set and the non-abandoned land sample characteristic set.

The non-abandoned land sample feature set is used for representing the data distribution of the change value of non-abandoned land NDVI, and the abandoned land sample feature set is used for representing the data distribution of the change value of abandoned land NDVI. For illustrative purposes, the present application uses data histograms to characterize the data distribution of abandoned NDVI variation values and the data distribution of non-abandoned NDVI variation values, as shown in fig. 4.

As can be seen in fig. 4, the data distribution of the change values of a abandoned land NDVI and the data distribution of the change values of a non-abandoned land NDVI partially intersect each other, and this results in a partial non-abandoned land being identified as a abandoned land or a abandoned land being classified as a non-abandoned land. To solve the above problems, the present application uses NDVI variation, precision, and recall to characterize the degree of overlap between the data distribution of abandoned NDVI variations and the data distribution of non-abandoned NDVI variations. Wherein the precision rate is indicative of a proportion of a sample determined to be a abandoned land that is actually a abandoned land, and the recall rate is indicative of a proportion of a sample determined to be a abandoned land that has had its features concentrated correctly. For example, the change values of NDVI of 50 abandoned lands and 50 non-abandoned lands are collected. Taking an NDVI variation (amplitude in fig. 4) of 0.5 means that an NDVI variation of less than 0.5 is considered a abandoned land, and at this time, 50 abandoned lands are identified as abandoned lands, so that a recall rate of 1 is achieved, and 14 non-abandoned lands are also identified as abandoned lands, so that an accuracy rate of 78% is achieved.

The problem of identifying non-abandoned land and abandoned land is converted into the problem of finding a balance point at the precision rate and the recall rate in the manner described above. In order to find the balance point, the parameter F1 is provided for representing the incidence relation between the accuracy rate and the recall rate, and the balance point is found by solving the maximum value of the F1. The process for distinguishing and identifying non-abandoned land and abandoned land is realized by the mode.

In particular, the amount of the solvent to be used,

wherein precision represents accuracy, and recall represents recall.

The step 2 comprises the following steps: determining a abandoned land quantity according to the initial value, the abandoned land sample feature set and the non-abandoned land sample feature set;

determining an accuracy rate and a recall rate for identifying a abandoned land based on the quantity of abandoned land, the sample characteristic set of abandoned land and the sample characteristic set of non-abandoned land; and determining F1 according to the accuracy and the recall rate corresponding to the initial value.

Thus, the initial value of the NDVI change value and the initial value of the F1 are obtained, and then an iterative algorithm is utilized to find a balance point. The initial value of the NDVI change value is usually set to ensure a recall of 1.

The specific process of iteration is as follows:

and A1, reducing the initial threshold according to a preset step length, and obtaining a new accuracy rate and a new recall rate.

And A2, determining a new F1 according to the new accuracy rate and the new recall rate.

And A3, repeating the process until more than half of the abandoned land sample features are collected to be correctly identified as abandoned land.

And A4, determining that the abandoned NDVI change value corresponding to the maximum value of F1 is a abandoned judgment value.

And A5, constructing a abandoned land criterion according to the abandoned land judgment value.

For example, as shown in fig. 4 and 5, the initial value of the NDVI change is 0.5, the step size is 0.01, and the NDVI change is recurred to a series of NDVI change values, which now result in F1, and the NDVI change value of 0.415 corresponding to the maximum value of F1 is selected as the abandoned decision value, i.e., the NDVI change value is the equilibrium point as described above.

Thus, the abandoned land criterion can be obtained: determining the abandoned land to be identified as a abandoned land when the NDVI change value of the abandoned land to be identified is less than 0.415;

determining the land to be identified as a non-abandoned land when the change value of the abandoned land NDVI of the land to be identified is not less than 0.415.

And 3, identifying the abandoned land in the area to be detected according to abandoned land criteria.

In the embodiment of the present application, step 3 includes:

determining each plot to be identified in the areas to be detected;

obtaining an NDVI curve of each land parcel to be identified;

determining the change value of the NDVI curve of each parcel to be identified according to the NDVI curve of each parcel to be identified;

and respectively determining whether each land to be identified is abandoned according to abandoned land criteria and the change value of the NDVI curve of each land to be identified.

The embodiment of the application provides a cultivated land abandoned land monitoring devices based on vegetation season changes, includes: the device comprises an acquisition module, a criterion creation module and a data processing module;

the collecting module is used for collecting a normalized vegetation index NDVI time sequence curve aiming at the area to be detected to obtain a abandoned land sample feature set and a non-abandoned land sample feature set;

the criterion creating module is used for determining a abandoned land criterion of the area to be detected based on the abandoned land sample characteristic set and the non-abandoned land sample characteristic set;

the data processing module is used for identifying the abandoned land in the area to be detected according to abandoned land criteria.

An embodiment of the present application provides a storage medium, including:

aiming at an area to be detected, acquiring a abandoned land sample characteristic set and a non-abandoned land sample characteristic set by acquiring a normalized vegetation index NDVI time sequence curve; determining a abandoned land criterion of the area to be detected based on the abandoned land sample characteristic set and the non-abandoned land sample characteristic set;

To illustrate the correctness of the above scheme, the following examples are given:

when the method is applied to abandoned land monitoring in the first grade area of the Zhanjiang city of 2020, the NDVI timing amplitude threshold for identifying abandoned land and non-abandoned land is determined to be 0.415 according to the method, 90 abandoned land sample feature sets and 90 non-abandoned land sample feature sets are respectively selected for precision verification, and the result shows that the identification precision of the abandoned land is 93.3% and the identification precision of the non-abandoned land is 90.0%.

The foregoing description of specific embodiments has been presented for purposes of illustration and description. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims can be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.

In the 30 s of the 20 th century, improvements in a technology could clearly distinguish between improvements in hardware (e.g., improvements in circuit structures such as diodes, transistors, switches, etc.) and improvements in software (improvements in process flow). However, as technology advances, many of today's process flow improvements have been seen as direct improvements in hardware circuit architecture. Designers almost always obtain the corresponding hardware circuit structure by programming an improved method flow into the hardware circuit. Thus, it cannot be said that an improvement in the process flow cannot be realized by hardware physical modules. For example, a Programmable Logic Device (PLD), such as a Field Programmable Gate Array (FPGA), is an integrated circuit whose Logic functions are determined by programming the Device by a user. A digital system is "integrated" on a PLD by the designer's own programming without requiring the chip manufacturer to design and fabricate application-specific integrated circuit chips. Furthermore, nowadays, instead of manually manufacturing an Integrated Circuit chip, such Programming is often implemented by "logic compiler" software, which is similar to a software compiler used in program development, but the original code before compiling is also written in a specific Programming Language, which is called Hardware Description Language (HDL), and the HDL is not only one kind but many kinds, such as abll (Advanced boot Expression Language), AHDL (alternate hard Description Language), traffic, CUPL (computer universal Programming Language), HDCal (Java hard Description Language), lava, lola, HDL, PALASM, software, rhydl (Hardware Description Language), and vhul-Language (vhyg-Language), which is currently used in the field. It will also be apparent to those skilled in the art that hardware circuitry that implements the logical method flows can be readily obtained by merely slightly programming the method flows into an integrated circuit using the hardware description languages described above.

The controller may be implemented in any suitable manner, for example, the controller may take the form of, for example, a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro) processor, logic gates, switches, an Application Specific Integrated Circuit (ASIC), a programmable logic controller, and an embedded microcontroller, examples of which include, but are not limited to, the following microcontrollers: ARC 625D, atmel AT91SAM, microchip PIC18F26K20, and Silicone Labs C8051F320, the memory controller may also be implemented as part of the control logic for the memory. Those skilled in the art will also appreciate that, in addition to implementing the controller in purely computer readable program code means, the same functionality can be implemented by logically programming method steps such that the controller is in the form of logic gates, switches, application specific integrated circuits, programmable logic controllers, embedded microcontrollers and the like. Such a controller may thus be regarded as a hardware component and the means for performing the various functions included therein may also be regarded as structures within the hardware component. Or even means for performing the functions may be regarded as being both a software module for performing the method and a structure within a hardware component.

The systems, devices, modules or units illustrated in the above embodiments may be implemented by a computer chip or an entity, or by a product with certain functions. One typical implementation device is a computer. In particular, the computer may be, for example, a personal computer, a laptop computer, a cellular telephone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

For convenience of description, the above devices are described as being divided into various units by function, respectively. Of course, the functions of the various elements may be implemented in the same one or more pieces of software and/or hardware in practicing embodiments of the present description.

One skilled in the art will recognize that one or more embodiments of the present description may be provided as a method, system, or computer program product. Accordingly, one or more embodiments of the present description may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the description may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

The description has been described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the description. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

In a typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.

The memory may include forms of volatile memory in a computer readable medium, random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.

Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, a computer readable medium does not include a transitory computer readable medium such as a modulated data signal and a carrier wave.

It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrases "comprising a," "8230," "8230," or "comprising" does not exclude the presence of other like elements in a process, method, article, or apparatus comprising the element.

One or more embodiments of the present description may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. One or more embodiments of the specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.

The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, as for the system embodiment, since it is substantially similar to the method embodiment, the description is relatively simple, and reference may be made to the partial description of the method embodiment for relevant points.

The above description is only an example of the present document and is not intended to limit the present document. Various modifications and changes may occur to those skilled in the art from this document. Any modification, equivalent replacement, improvement or the like made within the spirit and principle of this document shall be included in the scope of the claims of this document.

Claims

1. A abandoned cultivated land monitoring method based on vegetation season change is characterized by comprising the following steps: aiming at an area to be detected, acquiring a abandoned land sample characteristic set and a non-abandoned land sample characteristic set by acquiring a normalized vegetation index NDVI time sequence curve based on the species of the area to be detected; the abandoned land sample feature set comprises: the abandoned land's NDVI timing curve has a difference between a maximum NDVI and a minimum NDVI, and the non-abandoned land sample feature set comprises: a difference between a maximum NDVI and a minimum NDVI of the non-abandoned NDVI timing curve;

identifying a abandoned land in the area to be detected according to the abandoned land criterion;

determining a abandoned land criterion of the area to be detected based on the abandoned land sample feature set and the non-abandoned land sample feature set, wherein the abandoned land criterion comprises:

the abandoned land sample feature set comprises at least one abandoned land NDVI variation value;

determining the maximum value of the abandoned land NDVI (new landform NDVI) change value as an initial value according to the abandoned land sample feature set;

determining a abandoned land quantity according to the initial value, the abandoned land sample feature set and the non-abandoned land sample feature set;

determining F1 according to the precision rate and the recall rate corresponding to the initial value, wherein the F1 is a numerical value and is used for representing the incidence relation between the precision rate and the recall rate;

iterating the initial value and the F1 to obtain the abandoned land criterion;

aiming at an area to be detected, a abandoned land sample characteristic set and a non-abandoned land sample characteristic set are obtained by collecting a normalized vegetation index NDVI time sequence curve, and the method comprises the following steps:

determining a abandoned land sample feature set according to the NDVI change value corresponding to each abandoned land;

determining an NDVI variation value corresponding to each abandoned land and an NDVI variation value corresponding to each non-abandoned land according to the NDVI timing curve, comprising:

a difference between a maximum NDVI and a minimum NDVI of an NDVI timing curve of the abandoned land is a NDVI variation value corresponding to the abandoned land;

determining a maximum NDVI and a minimum NDVI of the non-abandoned land's NDVI timing curve;

the difference between the maximum NDVI and the minimum NDVI of the NDVI timing curve of the non-abandoned land is the corresponding NDVI change value of the non-abandoned land;

the method further comprises the following steps: and determining whether the species in the area to be detected changes, and if so, updating the non-abandoned land sample feature set and the abandoned land sample feature set according to the species change.

2. The method of claim 1,

iterating the initial value and the F1 to obtain the abandoned land criterion, comprising:

A1. reducing the initial value according to a preset step length, and obtaining a new precision rate and a new recall rate;

A2. determining a new F1 according to the new precision rate and the new recall rate;

repeating A1-A2 until more than half of the samples in the abandoned land sample can be correctly identified as abandoned land;

constructing the abandoned land criterion according to the abandoned land criterion value.

3. The method of claim 2,

determining the correlation between the accuracy rate and the recall rate by using the following formula:

4. The method of claim 2,

the abandoned land criterion is specifically as follows:

determining the abandoned land to be identified as a abandoned land when the change value of the abandoned land NDVI of the land to be identified is smaller than the abandoned land determination value;

or the like, or, alternatively,

determining the land to be identified as a non-abandoned land when the abandoned land NDVI change value of the land to be identified is not less than the abandoned land determination value.

5. The method of claim 4,

identifying a abandoned land in the area to be detected according to the abandoned land criterion comprises:

determining each plot to be identified in the areas to be detected;

obtaining an NDVI curve of each land parcel to be identified;

and respectively determining whether each land to be identified is abandoned according to the abandoned land criterion and the change value of the NDVI curve of each land to be identified.

6. A cultivated land abandoned monitoring devices based on vegetation season change, which is characterized in that, includes: the device comprises an acquisition module, a criterion creation module and a data processing module;

the collection module is used for collecting a normalized vegetation index NDVI time sequence curve to obtain a abandoned land sample characteristic set and a non-abandoned land sample characteristic set aiming at the area to be detected based on the species of the area to be detected; the abandoned land sample feature set comprises: the abandoned land's NDVI timing curve has a difference between a maximum NDVI and a minimum NDVI, and the non-abandoned land sample feature set comprises: the difference between the maximum and minimum NDVI of the non-abandoned land NDVI timing curve

the data processing module is used for identifying a abandoned land in the area to be detected according to the abandoned land criterion;

said abandoned sample feature set comprises at least one said abandoned NDVI variation value;

determining an accuracy rating for identifying the abandoned land based on the quantity of the abandoned land, the sample characteristic set of the abandoned land and the sample characteristic set of the non-abandoned land, the accuracy rating being indicative of a proportion of the sample determined to be a abandoned land that is actually a abandoned land, and a recall rating being indicative of a proportion of the sample characteristic set of the abandoned land that is correctly determined to be a abandoned land;

7. A storage medium, comprising:

for storing computer-executable instructions that, when executed, implement the following flow: aiming at a to-be-detected area, acquiring a normalized vegetation index NDVI time sequence curve based on species of the to-be-detected area to obtain a abandoned land sample feature set and a non-abandoned land sample feature set; the abandoned land sample feature set comprises: the difference between a maximum NDVI and a minimum NDVI of the abandoned land NDVI timing curve, the non-abandoned land sample feature set comprising: a difference between a maximum NDVI and a minimum NDVI of the non-abandoned NDVI timing curve;

determining the abandoned land quantity according to the initial value, the abandoned land sample feature set and the non-abandoned land sample feature set;