CN117151477A

CN117151477A - Crop anomaly identification method, device, electronic equipment and storage medium

Info

Publication number: CN117151477A
Application number: CN202311423854.6A
Authority: CN
Inventors: 郭朝贺; 王宏斌; 秦志珩; 杨子龙
Original assignee: Sinochem Agriculture Holdings
Current assignee: Sinochem Agriculture Holdings
Priority date: 2023-10-31
Filing date: 2023-10-31
Publication date: 2023-12-01
Anticipated expiration: 2043-10-31
Also published as: CN117151477B

Abstract

The invention provides a crop anomaly identification method, a crop anomaly identification device, electronic equipment and a storage medium, and relates to the technical field of agricultural monitoring. The method comprises the following steps: acquiring an actual growth trend monitoring curve of the crop to be identified at the current time, and acquiring a standard growth trend monitoring curve corresponding to the crop to be identified, wherein the actual growth trend monitoring curve and the standard growth trend monitoring curve are the same monitoring plot growth trend monitoring curve; determining a growth trend difference index based on the comparison result of the actual growth trend monitoring curve and the standard growth trend monitoring curve, wherein the growth trend difference index represents the difference degree of the change rate of the actual growth trend monitoring curve and the change rate of the standard growth trend monitoring curve; and determining an abnormal identification result of the crop to be identified on the monitored land parcel based on the comparison result of the growth trend difference index and the at least one difference index threshold. The invention can improve the accuracy of identifying the crop abnormality.

Description

Crop anomaly identification method, device, electronic equipment and storage medium

Technical Field

The present invention relates to the field of agricultural monitoring technologies, and in particular, to a crop anomaly identification method, a device, an electronic apparatus, and a storage medium.

Background

With the rapid development of technology, higher requirements are placed on the yield and quality of crops. In order to increase the yield of crops and improve the quality of crops, the crops need to be monitored abnormally to identify whether the crops are abnormal, so that the influence of disaster and other abnormal reasons on the crops is reduced, and the production safety of the crops is ensured.

Currently, crop anomaly identification is mostly performed for a specific disaster category to identify whether the specific disaster category occurs to the crop or for the severity of the specific disaster category to identify the severity of the crop with respect to the specific disaster category. However, only one abnormality occurrence cause can be identified for a single disaster category, and the abnormality occurrence causes of crops are various, so that identification for only one abnormality occurrence cause is not accurate. Even if the crop growth vigor of which area in the space is high or low is monitored to determine the poor area as the area where the crop abnormality occurs, however, the field crop state is quite different, the situation that whether the abnormality occurs or not cannot be accurately determined by simply judging the growth vigor cannot be simply judged, and the accuracy of identifying the crop abnormality is reduced.

Disclosure of Invention

The invention provides a crop anomaly identification method, a crop anomaly identification device, electronic equipment and a storage medium, which are used for solving the defect of low accuracy of crop anomaly identification in the prior art and realizing high-accuracy crop anomaly identification.

The invention provides a crop anomaly identification method, which comprises the following steps:

acquiring an actual growth trend monitoring curve of a crop to be identified at a current time, and acquiring a standard growth trend monitoring curve corresponding to the crop to be identified, wherein the actual growth trend monitoring curve and the standard growth trend monitoring curve are the same growth trend monitoring curve of a monitoring land block, and the growth trend monitoring curve is used for representing the growth variation trend of the crop to be identified;

determining a growth trend difference index based on the comparison result of the actual growth trend monitoring curve and the standard growth trend monitoring curve, wherein the growth trend difference index represents the degree of difference between the change rate of the actual growth trend monitoring curve and the change rate of the standard growth trend monitoring curve;

and determining an abnormal identification result of the crop to be identified in the monitored plot based on a comparison result of the growth trend difference index and at least one difference index threshold.

According to the crop anomaly identification method provided by the invention, the acquisition of the actual growth trend monitoring curve of the crop to be identified at the current moment comprises the following steps:

determining a vegetation index monitoring curve of the monitored land block based on remote sensing monitoring data of the monitored land block in a preset time before the current moment;

and determining the actual growth trend monitoring curve corresponding to the growth period of the crop to be identified from the vegetation index monitoring curve based on the weather period information of the crop to be identified corresponding to the monitoring land, wherein the weather period information comprises mapping relations of a plurality of time periods and a plurality of weather periods.

According to the method for identifying crop anomalies provided by the invention, the actual growth trend monitoring curve corresponding to the growth period of the crop to be identified is determined from the vegetation index monitoring curve based on the weather period information corresponding to the crop to be identified in the monitoring land, and the method comprises the following steps:

determining a target vegetation index monitoring curve in a growth period of the crop to be identified from the vegetation index monitoring curve based on the last growth lifting point in the vegetation index monitoring curve, wherein the starting point of the target vegetation index monitoring curve is the last growth lifting point, and the target vegetation index monitoring curve is the tail end curve of the vegetation index monitoring curve;

Determining the actual growth trend monitoring curve corresponding to the growth period of the crop to be identified from the target vegetation index monitoring curve based on the weather period information of the crop to be identified corresponding to the monitoring land block;

the growth vigor corresponding to the vegetation index before the growth promoting point in the vegetation index monitoring curve is a descending growth vigor, or the vegetation index before the growth promoting point in the vegetation index monitoring curve does not exist, or the vegetation index before the growth promoting point in the vegetation index monitoring curve does not change; and the growth vigor corresponding to the vegetation index after the growth vigor lifting point in the vegetation index monitoring curve is rising growth vigor.

According to the crop anomaly identification method provided by the invention, the acquisition of the standard growth trend monitoring curve corresponding to the crop to be identified comprises the following steps:

acquiring a template growth trend monitoring curve corresponding to the crop to be identified, wherein the template growth trend monitoring curve and the actual growth trend monitoring curve are the same growth trend monitoring curve of a monitoring land block;

based on the statistical characteristics of the actual growth trend monitoring curve, adjusting the template growth trend monitoring curve to obtain the standard growth trend monitoring curve;

Wherein the statistical characteristic includes at least one of a monitored value of a start point of the growth period, a duration of the plurality of growth phases, and a maximum monitored value during the growth period.

According to the crop anomaly identification method provided by the invention, the anomaly identification result of the crop to be identified in the monitored land is determined based on the comparison result of the growth trend difference index and at least one difference index threshold value, and then the method further comprises the following steps:

under the condition that the abnormal recognition result is a preset recognition result, determining an abnormal condition corresponding to the abnormal recognition result based on a comparison result of the actual growth trend monitoring curve and the standard growth trend monitoring curve;

the preset identification result characterizes that the crop to be identified in the monitored land parcels has abnormal growth at the current moment; the abnormal condition includes a positive abnormal condition that characterizes an actual growth trend greater than a standard growth trend or a negative abnormal condition that characterizes an actual growth trend less than the standard growth trend.

According to the crop anomaly identification method provided by the invention, the abnormal condition corresponding to the anomaly identification result is determined based on the comparison result of the actual growth trend monitoring curve and the standard growth trend monitoring curve, and then the method further comprises the following steps:

Determining an occurrence reason corresponding to the abnormality identification result based on related data of the monitored land parcel, wherein the related data comprises at least one of meteorological data of the monitored land parcel, internet of things equipment monitoring data of the monitored land parcel, ground patrol data of the monitored land parcel and weather period information corresponding to the monitored land parcel, and the weather period information comprises mapping relations of a plurality of time periods and a plurality of weather periods;

and when the abnormal condition is a forward abnormal condition, determining that the occurrence reason corresponding to the abnormal recognition result is weed harm.

According to the method for identifying crop anomalies provided by the invention, the related data comprise meteorological data of the monitored land, monitoring data of internet of things equipment of the monitored land, ground inspection data of the monitored land and weather period information corresponding to the monitored land for the crop to be identified, and the determining of occurrence reasons corresponding to the anomaly identification results based on the related data of the monitored land comprises the following steps:

determining first risk degree data of a lodging reason based on the current moment, the weather period information and the meteorological data;

Determining second risk degree data of waterlogging reasons based on the meteorological data;

determining third risk degree data of a cause of the high-temperature heat injury based on the current time, the weather period information and the meteorological data;

determining fourth risk degree data of pest causes based on the internet of things equipment monitoring data and the ground inspection data;

and determining the occurrence reason corresponding to the abnormal recognition result based on the first risk degree data, the second risk degree data, the third risk degree data and the fourth risk degree data.

According to the crop anomaly identification method provided by the invention, the determining of the occurrence cause corresponding to the anomaly identification result based on the first risk level data, the second risk level data, the third risk level data and the fourth risk level data comprises the following steps:

determining that the occurrence reason corresponding to the abnormal recognition result is other reasons when the first risk degree data, the second risk degree data, the third risk degree data and the fourth risk degree data are all smaller than or equal to a preset risk degree threshold value;

And determining maximum risk degree data from the first risk degree data, the second risk degree data, the third risk degree data and the fourth risk degree data under the condition that at least one of the first risk degree data, the second risk degree data, the third risk degree data and the fourth risk degree data is larger than a preset risk degree threshold value, and determining a reason corresponding to the maximum risk degree data as an occurrence reason corresponding to the abnormal recognition result.

The invention also provides a crop abnormality recognition device, which comprises:

the system comprises a curve acquisition module, a control module and a control module, wherein the curve acquisition module is used for acquiring an actual growth trend monitoring curve of a crop to be identified at the current time and acquiring a standard growth trend monitoring curve corresponding to the crop to be identified, the actual growth trend monitoring curve and the standard growth trend monitoring curve are the same growth trend monitoring curve of a monitoring land, and the growth trend monitoring curve is used for representing the growth change trend of the crop to be identified;

the index determining module is used for determining a growth trend difference index based on the comparison result of the actual growth trend monitoring curve and the standard growth trend monitoring curve, wherein the growth trend difference index represents the degree of difference between the change rate of the actual growth trend monitoring curve and the change rate of the standard growth trend monitoring curve;

And the result determining module is used for determining an abnormal recognition result of the crop to be recognized in the monitored land block based on the comparison result of the growth trend difference index and at least one difference index threshold value.

The invention also provides an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the crop anomaly identification method according to any one of the above when executing the program.

The present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements a crop anomaly identification method as described in any one of the above.

According to the crop anomaly identification method, the device, the electronic equipment and the storage medium, an actual growth trend monitoring curve of crops to be identified at the current moment is obtained, a standard growth trend monitoring curve corresponding to the crops to be identified is obtained, so that a growth trend difference index is determined based on a comparison result of the actual growth trend monitoring curve and the standard growth trend monitoring curve, and an anomaly identification result of the crops to be identified in a monitored area is determined based on a comparison result of the growth trend difference index and at least one difference index threshold. By the method, the abnormal recognition result of the crop to be recognized can be determined, and whether the crop to be recognized is abnormal or not and the degree of the abnormal occurrence can be obtained by recognition instead of recognition of one abnormal occurrence reason, so that various abnormal occurrence reasons are considered, and the accuracy of crop abnormal recognition is improved; meanwhile, the actual growth trend monitoring curve and the standard growth trend monitoring curve are the same growth trend monitoring curve of the monitoring land block, so that the standard growth trend monitoring curve corresponding to the monitoring land block corresponding to the actual growth trend monitoring curve is compared with the actual growth trend monitoring curve, and the situation that different land blocks are different in condition is considered, namely, the situation that different land blocks correspond to different standard growth trend monitoring curves is considered, and the accuracy of identifying abnormal crops is further improved; most importantly, based on the comparison result of the actual growth trend monitoring curve and the standard growth trend monitoring curve, the growth trend difference index is determined, so that whether the growth condition of the crops deviates from the standard growth trend is considered, rather than which area is considered to be more or worse in crop growth condition, namely misjudgment caused by the crop growth condition difference existing in the field is eliminated, and the accuracy of identifying the crop abnormality is further improved.

Drawings

In order to more clearly illustrate the invention or the technical solutions of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.

FIG. 1 is a schematic flow chart of a crop anomaly identification method according to the present invention;

FIG. 2 is a second flow chart of the crop anomaly identification method according to the present invention;

FIG. 3 is a schematic diagram of a growth trend monitoring curve provided by the present invention;

fig. 4 is a schematic structural diagram of a crop anomaly identification apparatus provided by the present invention;

fig. 5 is a schematic structural diagram of an electronic device provided by the present invention.

Detailed Description

For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

Currently, crop anomaly identification is mostly performed for a specific disaster category to identify whether the specific disaster category occurs to the crop or for the severity of the specific disaster category to identify the severity of the crop with respect to the specific disaster category; for example, a ground sample point is used for collecting, multiple vegetation indexes with potential high relevance degrees are calculated by matching with multispectral/hyperspectral data collection and satellite remote sensing data, or novel vegetation indexes are constructed to form a vegetation index set to be selected together, and finally, one vegetation index with the highest relevance degree or a combination of a plurality of vegetation indexes is selected in the vegetation index set to be selected by using a relevance analysis method to construct a regression relation. However, only one abnormality occurrence cause can be identified for a single disaster category, and the abnormality occurrence causes of crops are various, so that identification for only one abnormality occurrence cause is not accurate. And at present, monitoring is mostly carried out aiming at specific agricultural condition indexes, for example, soil moisture content, canopy moisture content, seedling condition, biomass, chlorophyll content, leaf area index, nitrogen content and the like; however, monitoring is performed based on a single index, and the obtained abnormal recognition result and severity of a specific disaster category are inaccurate due to various field conditions, namely, the monitoring of the single index is difficult to be effectively applied to agricultural production practice. Even if the crop growth vigor of which area in the space is high or low is monitored to determine the poor area as the area where the crop abnormality occurs, however, the field crop state is quite different, the situation that whether the abnormality occurs or not cannot be accurately determined by simply judging the growth vigor cannot be simply judged, and the accuracy of identifying the crop abnormality is reduced.

In addition, even if the machine learning model is used for identifying the crop anomalies, because the types and varieties of field crops are various, great differences exist among different regions, and the crop anomaly identification model trained on one or a plurality of varieties of crops in a specific region has great limitation in expandability, namely, the crop anomaly identification model needs to be respectively constructed and trained on the crops of different types, varieties and regions, so that the efficiency is low and the cost is high.

In view of the above problems, the present invention proposes the following embodiments. Fig. 1 is a schematic flow chart of a crop anomaly identification method provided by the present invention, as shown in fig. 1, the crop anomaly identification method includes:

step 110, obtaining an actual growth trend monitoring curve of the crop to be identified at the current time, and obtaining a standard growth trend monitoring curve corresponding to the crop to be identified.

The actual growth trend monitoring curve and the standard growth trend monitoring curve are the same growth trend monitoring curve of the monitoring land block, and the growth trend monitoring curve is used for representing the growth variation trend of the crop to be identified.

Here, the crop to be identified is a crop of an abnormal condition to be identified, and the crop to be identified may be a field crop of one season, for example, wheat, corn, rice, and the like.

Here, the actual growth trend monitoring curve is an actual growth trend monitoring curve of the crop to be identified on the monitored plot, and the actual growth trend monitoring curve is used for representing an actual growth variation trend (growth trend variation condition) of the crop to be identified. The standard growth trend monitoring curve is a growth trend monitoring curve of the crop to be identified for reference in a monitoring land, and is used for representing the standard growth variation trend (growth trend variation condition) of the crop to be identified. The growth trend monitoring curve is a curve of the change of the growth trend index with time. In one embodiment, the growth index may be a vegetation index, for example, the growth index is NDVI (Normalized Difference Vegetation Index, normalized vegetation index).

In some embodiments, the vegetation index monitoring curve of the monitored land is determined based on remote sensing monitoring data of the monitored land within a preset time before the current moment, the vegetation index monitoring curve can be directly determined as an actual growth trend monitoring curve, and the vegetation index monitoring curve can be further processed to obtain the actual growth trend monitoring curve.

The preset time length can be set according to actual needs, for example, according to the growth period length of crops to be identified, for example, the preset time length is 5-6 months. The vegetation index monitoring curve is calculated based on remote sensing monitoring data, for example, the vegetation indexes of each pixel on different images are calculated by using the observation values of original remote sensing images to form a vegetation index set of a time sequence, and then the vegetation index monitoring curve is constructed based on the vegetation index set. The vegetation index monitoring curve is used for representing the growth change trend (growth situation) of the monitored land block; it is a plot of vegetation index over time, i.e., it is a continuous vegetation index variation plot.

In an embodiment, the remote sensing monitoring data is input to the cloud layer identification model to obtain a cloud layer area output by the cloud layer identification model, the remote sensing monitoring data is screened based on the cloud layer area to obtain screened remote sensing monitoring data, so that the screened remote sensing monitoring data does not comprise remote sensing images corresponding to the cloud layer area, and a vegetation index monitoring curve of the monitored land is determined based on the screened remote sensing monitoring data. The cloud layer identification model is obtained based on sample remote sensing monitoring data and corresponding sample cloud layer area labels. In other words, cloud identification is performed on each image in the remote sensing monitoring data by using a remote sensing image cloud detection algorithm, and a cloud mask is generated, so that the vegetation index is not calculated in the area covered by the cloud, the monitoring value in the vegetation index monitoring curve is not influenced by the cloud, and the accuracy of crop anomaly identification is improved.

In an embodiment, based on the weather period information corresponding to the crop to be identified in the monitored land, an actual growth trend monitoring curve corresponding to the growth period of the crop to be identified is determined from the vegetation index monitoring curves.

The weathered period information comprises mapping relations between a plurality of time periods and a plurality of weathered periods. The weather period information is information stored in advance in a database. The weather period information comprises a plurality of mapping relations, wherein any mapping relation is a mapping relation between a time period and a weather period, so that the weather period corresponding to the time can be determined based on the time, and the weather period of the crop is judged based on the weather period information.

Specifically, based on the weather period information corresponding to the crop to be identified in the monitored land, determining a curve which does not belong to the crop to be identified and corresponds to the growth period in the vegetation index monitoring curve, so as to remove the curve which does not belong to the crop to be identified and corresponds to the growth period in the vegetation index monitoring curve, and obtain an actual growth trend monitoring curve which corresponds to the growth period of the crop to be identified, thereby removing the vegetation index in the non-growth period, ensuring that an accurate actual growth trend monitoring curve is obtained, and further improving the accuracy of abnormal identification of the crop.

In an embodiment, based on the last growth promoting point in the vegetation index monitoring curve, determining a target vegetation index monitoring curve in a growth period of the crop to be identified from the vegetation index monitoring curve to extract a curve conforming to the crop development rule, i.e. the extracted curve is in the growth period of the crop to be identified, thereby removing the vegetation index in the non-growth period, ensuring that an accurate actual growth trend monitoring curve is obtained, and further improving the accuracy of identifying the abnormal crop. The starting point of the target vegetation index monitoring curve is the last growth lifting point, and the target vegetation index monitoring curve is the tail end curve of the vegetation index monitoring curve.

The last growth point is the last growth point in the vegetation index monitoring curve, the growth corresponding to the vegetation index before the growth point in the vegetation index monitoring curve is the descending growth point, and the growth corresponding to the vegetation index after the growth point in the vegetation index monitoring curve is the ascending growth point; alternatively, there is no vegetation index before the growth promotion point in the vegetation index monitoring curve.

In some embodiments, the standard growth trend monitoring curve is preset. For example, the method is determined according to a historical growth trend monitoring curve of the crop to be identified in the monitored land block; further, the historical growth trend monitoring curve is a curve corresponding to the normal growth trend.

Step 120, determining a growth trend difference index based on the comparison result of the actual growth trend monitoring curve and the standard growth trend monitoring curve.

The growth trend difference index is used for representing the degree of difference between the actual growth trend of the crop to be identified and the standard growth trend. Further, the growth trend difference index is used for representing the magnitude of the difference degree between the actual growth trend of the crop to be identified in a period of time and the standard growth trend in a corresponding period of time, wherein the smaller the difference degree is used for representing that the actual growth trend is more consistent with the standard growth trend, the larger the difference degree is used for representing that the actual growth trend is less consistent with the standard growth trend, namely the larger the difference degree is used for representing that the actual growth trend is more deviated from the standard growth trend. It will be appreciated that the more the actual growth trend deviates from the standard growth trend, the corresponding monitored plot will need to be brought to the attention of the grower.

In one embodiment, the rate of change of the growth trend monitoring curve is determined based on the ratio of the monitored values at two times in the growth trend monitoring curve to the time interval between the two times. Illustratively, the formula for this rate of change is as follows:

；

wherein K represents the rate of change,representing the monitored value at time t1, +.>The monitored value at time t2 is indicated,representing the time interval between the two moments.

In some embodiments, the growth trend difference index is determined based on a comparison of a rate of change of the actual growth trend monitoring curve to a rate of change of the standard growth trend monitoring curve. Wherein the rate of change of the two curves is the rate of change of the same time period.

In an embodiment, the change rate is a change rate from a first time to a current time, the first time is a time before the current time, and a time interval between the first time and the current time is a preset time interval. Further, if the actual growth trend monitoring curve is determined by the remote sensing monitoring data, the preset time interval may be an acquisition time interval of the remote sensing image in the remote sensing monitoring data. For example, the difference in the rates of change of the two curves may be determined as a growth trend difference index.

In other embodiments, the growth trend difference index is determined based on a comparison of a plurality of rates of change of the actual growth trend monitoring curve with a plurality of rates of change of the standard growth trend monitoring curve. And the change rates corresponding to the target time period corresponding to any change rate of the actual growth trend monitoring curve exist in the change rates of the standard growth trend monitoring curve.

In an embodiment, the change rate from the first time to the current time exists in the plurality of change rates, the first time is a time before the current time, and a time interval between the first time and the current time is a preset time interval. Further, if the actual growth trend monitoring curve is determined by the remote sensing monitoring data, the preset time interval may be an acquisition time interval of the remote sensing image in the remote sensing monitoring data.

In an embodiment, the conversion rate difference values of a plurality of same time periods are determined based on the comparison result of a plurality of change rates of the actual growth trend monitoring curve and a plurality of change rates of the standard growth trend monitoring curve; a growth trend difference index is determined based on a sum of the plurality of transformation ratio differences.

In another embodiment, the plurality of rates of change includes a first rate of change from a first time to a current time, and a second rate of change from a second time to the first time, the second time being a time before the first time, and a time interval between the second time and the first time being a preset time interval. Further, if the actual growth trend monitoring curve is determined by the remote sensing monitoring data, the preset time interval may be an acquisition time interval of the remote sensing image in the remote sensing monitoring data.

Illustratively, a transformation rate difference is determined based on the difference between the first rate of change and the second rate of change; and determining a growth trend difference index based on the difference between the transformation rate difference of the actual growth trend monitoring curve and the transformation rate difference of the standard growth trend monitoring curve.

Of course, the growth trend difference index may also be determined by other methods, which will not be described in detail herein.

It will be appreciated that the growth trend difference index is determined based on the comparison of the actual growth trend monitoring curve and the standard growth trend monitoring curve, so that whether the growth condition of the crop itself deviates from the standard growth trend is considered, rather than which area the growth condition of the crop is vigorous or worse, in other words, the high value abnormality that the grower needs to pay attention to is considered only when the growth trend of the crop itself is abnormal, and measures are taken absolutely.

And step 130, determining an abnormal recognition result of the crop to be recognized in the monitored land parcel based on the comparison result of the growth trend difference index and at least one difference index threshold value.

Here, the number of result types of the abnormality recognition result corresponds to the number of difference index thresholds; for example, if the number of at least one difference index threshold is 1, the abnormality recognition result includes 2 results; the number of at least one difference index threshold is 2, the abnormality recognition result includes 3 results. The at least one difference index threshold is preset according to actual needs.

Illustratively, the at least one differential index threshold comprises a first differential index threshold and a second differential index threshold, the first differential index threshold being less than the second differential index threshold. When the growth trend difference index is greater than or equal to 0 and the growth trend difference index is smaller than the first difference index threshold, the abnormal recognition result is that the abnormal grade is 1, namely, the growth change trend of the crop to be recognized is determined to be normal; when the growth trend difference index is greater than or equal to the first difference index threshold and the growth trend difference index is less than the second difference index threshold, the anomaly identification result is "anomaly grade is 2", that is, the trend that the growth variation trend of the crop to be identified deviates from the standard growth variation trend (normal track) is determined, and it can be understood that the monitoring land block corresponding to the anomaly identification result should be paid attention early; when the growth trend difference index is greater than or equal to the second difference index threshold, the anomaly identification result is "anomaly level is 3", that is, it is determined that the growth variation trend of the crop to be identified has significantly deviated from the standard growth variation trend (normal track), it can be understood that the crop of the monitored plot corresponding to the anomaly identification result has a strong anomaly, and remedial measures need to be taken, that is, the crop to be identified has grown anomalies at the current moment.

According to the crop anomaly identification method provided by the embodiment of the invention, the actual growth trend monitoring curve of the crop to be identified at the current moment is obtained, the standard growth trend monitoring curve corresponding to the crop to be identified is obtained, the growth trend difference index is determined based on the comparison result of the actual growth trend monitoring curve and the standard growth trend monitoring curve, and the anomaly identification result of the crop to be identified in the monitored area is determined based on the comparison result of the growth trend difference index and at least one difference index threshold. By the method, the abnormal recognition result of the crop to be recognized can be determined, and whether the crop to be recognized is abnormal or not and the degree of the abnormal occurrence can be obtained by recognition instead of recognition of one abnormal occurrence reason, so that various abnormal occurrence reasons are considered, and the accuracy of crop abnormal recognition is improved; meanwhile, the actual growth trend monitoring curve and the standard growth trend monitoring curve are the same growth trend monitoring curve of the monitoring land block, so that the standard growth trend monitoring curve corresponding to the monitoring land block corresponding to the actual growth trend monitoring curve is compared with the actual growth trend monitoring curve, and the situation that different land blocks are different in condition is considered, namely, the situation that different land blocks correspond to different standard growth trend monitoring curves is considered, and the accuracy of identifying abnormal crops is further improved; most importantly, based on the comparison result of the actual growth trend monitoring curve and the standard growth trend monitoring curve, the growth trend difference index is determined, so that whether the growth condition of the crops deviates from the standard growth trend is considered, rather than which area is considered to be more or worse in crop growth condition, namely misjudgment caused by the crop growth condition difference existing in the field is eliminated, and the accuracy of identifying the crop abnormality is further improved.

Based on any of the above embodiments, fig. 2 is a second flowchart of the crop anomaly identification method provided by the present invention, as shown in fig. 2, in the step 110, an actual growth trend monitoring curve of the crop to be identified at the current time is obtained, which includes:

step 111, determining a vegetation index monitoring curve of the monitored land parcel based on remote sensing monitoring data of the monitored land parcel in a preset time before the current time.

Here, the preset time period may be set according to actual needs, for example, according to the growth period length of the crop to be identified, for example, the preset time period is 5-6 months.

The vegetation index monitoring curve is calculated based on remote sensing monitoring data, for example, the vegetation indexes of each pixel on different images are calculated by using the original remote sensing image observation values to form a vegetation index set of a time sequence, and then the vegetation index monitoring curve is constructed based on the vegetation index set. The vegetation index monitoring curve is used for representing the growth change trend (growth situation) of the monitored land block; it is a plot of vegetation index over time, i.e., it is a continuous vegetation index variation plot.

In an embodiment, the remote sensing monitoring data is input to the cloud layer identification model to obtain a cloud layer area output by the cloud layer identification model, the remote sensing monitoring data is screened based on the cloud layer area to obtain screened remote sensing monitoring data, so that the screened remote sensing monitoring data does not comprise remote sensing images corresponding to the cloud layer area, and a vegetation index monitoring curve of the monitored land is determined based on the screened remote sensing monitoring data.

The cloud layer identification model is obtained based on sample remote sensing monitoring data and corresponding sample cloud layer area labels. In other words, cloud identification is performed on each image in the remote sensing monitoring data by using a remote sensing image cloud detection algorithm, and a cloud mask is generated, so that the vegetation index is not calculated in the area covered by the cloud, the monitoring value in the vegetation index monitoring curve is not influenced by the cloud, and the accuracy of crop anomaly identification is improved.

In an embodiment, pixels are used as a processing unit, and a vegetation index monitoring curve of a plurality of pixels corresponding to a monitored land is determined based on remote sensing monitoring data. Correspondingly, the actual growth trend monitoring curve obtained later is also a pixel-level curve, the corresponding standard growth trend monitoring curve is also a pixel-level curve, and the actual growth trend monitoring curve and the standard growth trend monitoring curve are the same pixel growth trend monitoring curve.

Step 112, determining the actual growth trend monitoring curve corresponding to the growth period of the crop to be identified from the vegetation index monitoring curve based on the weather period information corresponding to the crop to be identified in the monitored land, wherein the weather period information comprises mapping relations between a plurality of time periods and a plurality of weather periods.

Here, the weather period information is information stored in advance in a database. The weather period information comprises a plurality of mapping relations, wherein any mapping relation is a mapping relation between a time period and a weather period, so that the weather period corresponding to the time can be determined based on the time, and the weather period of the crop is judged based on the weather period information.

In an embodiment, pre-stored weather period information is acquired, the weather period information is adjusted based on an actual growth trend monitoring curve, and adjusted weather period information is obtained, so that reliability and accuracy of judging the weather period of crops are ensured, and accuracy of identifying abnormal crops is further improved.

According to the crop anomaly identification method provided by the embodiment of the invention, the vegetation index monitoring curve of the monitored land is determined based on the remote sensing monitoring data of the monitored land in the preset time before the current moment, so that the actual growth trend monitoring curve corresponding to the growth period of the crop to be identified is determined from the vegetation index monitoring curve based on the weather period information corresponding to the monitored land of the crop to be identified, and the weather period information comprises the mapping relation between a plurality of time periods and a plurality of weather periods, so that the vegetation index in the non-growth period is removed, the accurate actual growth trend monitoring curve is ensured to be obtained, and the accuracy of crop anomaly identification is further improved.

Based on any of the above embodiments, the method further includes the step 112 of:

And determining the actual growth trend monitoring curve corresponding to the growth period of the crop to be identified from the target vegetation index monitoring curve based on the weather period information of the crop to be identified corresponding to the monitoring land.

It should be noted that, as shown in fig. 3, in the vegetation index monitoring curve, the growth situation corresponding to the vegetation index before the growth situation lifting point is a descending growth situation, that is, the growth situation before the box enclosed by the first vegetation index monitoring curve in fig. 3 is a descending growth situation, or, the vegetation index before the growth situation lifting point is unchanged, that is, the vegetation index before the box enclosed by the second vegetation index monitoring curve in fig. 3 is unchanged, meanwhile, the last growth situation lifting point is the first vegetation index in the box enclosed by the vegetation index monitoring curve, and the growth situation corresponding to the vegetation index after the growth situation lifting point is an ascending growth situation.

If the vegetation index monitoring curve is larger and better in growth vigor, the target vegetation index monitoring curve in one growth period should accord with the rule that the vegetation index is firstly increased and then decreased; if the vegetation index monitoring curve is smaller and better in growth vigor, the target vegetation index monitoring curve in one growth period should accord with the rule that the vegetation index is firstly decreased and then increased. I.e., the target vegetation index monitoring curve during one growth period is a continuous variation curve of a single growth period.

It should be noted that, considering that a vegetation index monitoring curve in other similar growth periods may exist before the current growth period, in order to ensure that the target vegetation index monitoring curve is a curve corresponding to the current time, it is necessary to ensure that the target vegetation index monitoring curve is an end curve of the vegetation index monitoring curve, so as to improve the accuracy of determining the actual growth trend monitoring curve, and further improve the accuracy of identifying crop anomalies.

Specifically, based on the weather period information corresponding to the crop to be identified in the monitored land, determining a curve which does not belong to the crop to be identified and corresponds to the growth period in the target vegetation index monitoring curve, so as to remove the curve which does not belong to the crop to be identified and corresponds to the growth period in the target vegetation index monitoring curve, and obtain an actual growth trend monitoring curve which corresponds to the growth period of the crop to be identified, thereby removing the vegetation index in the non-growth period, ensuring that an accurate actual growth trend monitoring curve is obtained, and further improving the accuracy of abnormal identification of the crop.

It can be understood that a single-period continuous change curve with a growth lifting point as a starting point at the tail end of the curve is extracted from the vegetation index monitoring curve, so as to extract a curve conforming to the development rule of crops, namely, the extracted curve is in the growth period of crops to be identified, thereby removing the vegetation index in a non-growth period, then, based on the weather period information corresponding to the monitored land block of the crops to be identified, removing partial curves in the non-crop growth period possibly remained in the target vegetation index monitoring curve, finally, intercepting an accurate actual growth trend monitoring curve, and further improving the accuracy of identifying the abnormal crops.

According to the crop anomaly identification method provided by the embodiment of the invention, the target vegetation index monitoring curve in one growth period of the crop to be identified is determined from the vegetation index monitoring curve based on the last growth point in the vegetation index monitoring curve, so that the curve conforming to the growth rule of the crop is extracted, namely, the extracted curve is in the growth period of the crop to be identified, so that the vegetation index in the non-growth period is removed, the determination accuracy of the actual growth trend monitoring curve is improved, the accuracy of the crop anomaly identification is further improved, then the actual growth trend monitoring curve corresponding to the growth period of the crop to be identified is determined from the target vegetation index monitoring curve based on the weather period information corresponding to the monitored land, so that part of curves possibly residual non-crop growth period in the target vegetation index monitoring curve are removed, and finally the accurate actual growth trend monitoring curve is intercepted, and the accuracy of the crop anomaly identification is further improved.

Based on any one of the above embodiments, in the method, in the step 110, obtaining a standard growth trend monitoring curve corresponding to the crop to be identified includes:

and adjusting the template growth trend monitoring curve based on the statistical characteristics of the actual growth trend monitoring curve to obtain the standard growth trend monitoring curve.

In some embodiments, the template growth trend monitoring curve is pre-set. For example, the method is determined according to a historical growth trend monitoring curve of the crop to be identified in the monitored land block; further, the historical growth trend monitoring curve is a curve corresponding to the normal growth trend.

Considering that the template growth trend monitoring curve cannot be directly used for judging whether the growth health of crops is good or not, the template growth trend monitoring curve needs to be adjusted based on the statistical characteristics of the actual growth trend monitoring curve to obtain a standard growth trend monitoring curve, namely, the template growth trend monitoring curve is adapted according to the actual statistical characteristics of crops in a monitored plot so as to be completely adapted to the crop growth process of the monitored plot, and finally, the adjusted standard growth trend monitoring curve can be used for judging the health state of the current crops. In other words, the standard growth trend monitoring curve is completely adapted to the crop growth process of the current land by comprehensive analysis of actual monitoring data of crops in the field, so that the standard growth trend monitoring curve can be used as a standard curve for monitoring abnormal conditions of the crop growth process in the field.

In some embodiments, the monitored value of the starting point of the template growth trend monitoring curve is adjusted based on the monitored value of the starting point of the actual growth trend monitoring curve so that the monitored value of the starting point of the template growth trend monitoring curve is adapted to the monitored value of the starting point of the actual growth trend monitoring curve. Further, the monitored value of the starting point of the template growth trend monitoring curve may be made the same as the monitored value of the starting point of the actual growth trend monitoring curve.

In an embodiment, if the monitored plot corresponds to a plurality of actual growth trend monitoring curves, determining a maximum starting point monitoring value based on the monitoring values of the starting points of the plurality of actual growth trend monitoring curves, and adjusting the monitoring value of the starting point of the template growth trend monitoring curve based on the maximum starting point monitoring value, so that the monitoring value of the starting point of the template growth trend monitoring curve is adapted to the maximum starting point monitoring value. Further, the monitoring value of the starting point of the template growth trend monitoring curve may be made the same as the maximum starting point monitoring value.

In some embodiments, the length of the plurality of growth phases of the template growth trend monitoring curve is adjusted based on the length of the plurality of growth phases of the actual growth trend monitoring curve such that the length of the plurality of growth phases of the template growth trend monitoring curve is adapted to the length of the plurality of growth phases of the actual growth trend monitoring curve, i.e., the template growth trend monitoring curve is compressed and stretched on the abscissa. Further, the length of the plurality of growth phases of the template growth trend monitoring curve may be made the same as the length of the plurality of growth phases of the actual growth trend monitoring curve.

In some embodiments, the maximum monitored value of the template growth trend monitoring curve during the growth period is adjusted based on the maximum monitored value of the actual growth trend monitoring curve during the growth period so that the maximum monitored value of the template growth trend monitoring curve during the growth period is adapted to the maximum monitored value of the actual growth trend monitoring curve during the growth period. Further, the maximum monitored value of the template growth trend monitoring curve in the growth period can be the same as the maximum monitored value of the actual growth trend monitoring curve in the growth period.

In an embodiment, if the monitored plot corresponds to a plurality of actual growth trend monitoring curves, determining a maximum monitored value in a maximum growth period based on the maximum monitored value in the growth period of the plurality of actual growth trend monitoring curves, and adjusting the maximum monitored value in the growth period of the template growth trend monitoring curve based on the maximum monitored value in the maximum growth period, so that the maximum monitored value in the growth period of the template growth trend monitoring curve is adapted to the maximum monitored value in the maximum growth period. Further, the maximum monitored value of the template growth trend monitoring curve during the growth period may be made the same as the maximum monitored value during the maximum growth period.

According to the crop anomaly identification method provided by the embodiment of the invention, the template growth trend monitoring curve is adjusted based on the statistical characteristics of the actual growth trend monitoring curve to obtain the standard growth trend monitoring curve, so that the standard growth trend monitoring curve is completely suitable for monitoring the actual crop growth process of the land, the determination accuracy of the standard growth trend monitoring curve is improved, and the accuracy of crop anomaly identification is further improved.

Based on any of the above embodiments, after the step 130, the method further includes:

and under the condition that the abnormal recognition result is a preset recognition result, determining the abnormal condition corresponding to the abnormal recognition result based on the comparison result of the actual growth trend monitoring curve and the standard growth trend monitoring curve.

Here, the forward abnormal condition characterizes that the actual growth trend monitoring curve is vigorous in growth vigor relative to the standard growth trend monitoring curve; the negative abnormal condition represents that the actual growth trend monitoring curve is worse than the standard growth trend monitoring curve.

In some embodiments, the abnormal condition corresponding to the abnormal recognition result is determined based on a comparison of the change rate of the actual growth trend monitoring curve with the change rate of the standard growth trend monitoring curve.

；

In one embodiment, the abnormal condition corresponding to the abnormal recognition result is determined based on a comparison result of a change rate of the actual growth trend monitoring curve and a change rate of the standard growth trend monitoring curve. Wherein the rate of change of the two curves is the rate of change of the same time period.

The change rate is a change rate from a first time to a current time, the first time is a time before the current time, and a time interval between the first time and the current time is a preset time interval. Further, if the actual growth trend monitoring curve is determined by the remote sensing monitoring data, the preset time interval may be an acquisition time interval of the remote sensing image in the remote sensing monitoring data.

For example, in the case where the growth trend is better as the monitored value is larger in the growth trend monitoring curve, if the change rate of the actual growth trend monitoring curve is larger than that of the standard growth trend monitoring curve, the abnormal condition corresponding to the abnormality recognition result is a positive abnormal condition, and otherwise, is a negative abnormal condition.

In another embodiment, the abnormal condition corresponding to the abnormal recognition result is determined based on the comparison result of the plurality of change rates of the actual growth trend monitoring curve and the plurality of change rates of the standard growth trend monitoring curve. And the change rates corresponding to the target time period corresponding to any change rate of the actual growth trend monitoring curve exist in the change rates of the standard growth trend monitoring curve.

In an exemplary embodiment, the change rate from the first time to the current time exists in the plurality of change rates, the first time is a time before the current time, and a time interval between the first time and the current time is a preset time interval. Further, if the actual growth trend monitoring curve is determined by the remote sensing monitoring data, the preset time interval may be an acquisition time interval of the remote sensing image in the remote sensing monitoring data.

For example, the plurality of change rates include a first change rate (not taking an absolute value) from a first time to a current time, and a second change rate (not taking an absolute value) from a second time to the first time, wherein the second time is a time before the first time, and a time interval between the second time and the first time is a preset time interval. Further, if the actual growth trend monitoring curve is determined by the remote sensing monitoring data, the preset time interval may be an acquisition time interval of the remote sensing image in the remote sensing monitoring data. Illustratively, a transformation rate difference is determined based on the difference between the first rate of change and the second rate of change; and determining the abnormal condition corresponding to the abnormal identification result based on the difference value of the transformation rate difference of the actual growth trend monitoring curve and the transformation rate difference of the standard growth trend monitoring curve.

For example, in the case where the larger the monitored value is in the growth trend monitoring curve, the better the growth trend is, if the change rate difference of the actual growth trend monitoring curve is larger than the change rate difference of the standard growth trend monitoring curve, the abnormal condition corresponding to the abnormal recognition result is a positive abnormal condition, otherwise, a negative abnormal condition is a negative abnormal condition.

Of course, the anomaly identification result may also be determined by other manners, which will not be described in detail herein.

According to the crop anomaly identification method provided by the embodiment of the invention, under the condition that the anomaly identification result is the preset identification result, the anomaly condition corresponding to the anomaly identification result is determined based on the comparison result of the actual growth trend monitoring curve and the standard growth trend monitoring curve, so that whether the anomaly condition corresponding to the anomaly identification result is positive anomaly condition or negative anomaly condition is further determined, and the accuracy of crop anomaly identification is further improved.

Based on any of the foregoing embodiments, after determining the abnormal situation corresponding to the abnormal recognition result based on the comparison result of the actual growth trend monitoring curve and the standard growth trend monitoring curve, the method further includes:

Here, the weather data is weather data during a period in which an abnormal condition occurs in the crop to be identified. In an embodiment, the weather data includes data from a first time to a current time, the first time is a time before the current time, and a time interval between the first time and the current time is a preset time interval. Further, if the actual growth trend monitoring curve is determined by the remote sensing monitoring data, the preset time interval may be an acquisition time interval of the remote sensing image in the remote sensing monitoring data. The meteorological data may include, but is not limited to, at least one of: air temperature, rainfall, wind speed, light duration, etc. The meteorological data can be used for judging the risk of occurrence of abnormality reasons such as lodging, waterlogging, high temperature heat injury and the like.

Here, the internet of things monitoring data is monitoring data during the period when the abnormal condition of the crop to be identified occurs. In an embodiment, the monitoring data of the internet of things device includes data from a first time to a current time, the first time is a time before the current time, and a time interval between the first time and the current time is a preset time interval. Further, if the actual growth trend monitoring curve is determined by the remote sensing monitoring data, the preset time interval may be an acquisition time interval of the remote sensing image in the remote sensing monitoring data.

In one embodiment, the internet of things device monitoring data includes insect pest monitoring data obtained from internet of things devices (insect pest monitoring devices) deployed in the field.

Here, the ground inspection data is inspection record information recorded by offline personnel during field inspection, and the information is difficult to obtain by means of remote sensing, internet of things and the like, especially the occurrence condition of insect diseases.

According to the crop anomaly identification method provided by the embodiment of the invention, under the condition that the anomaly condition is positive anomaly, the occurrence cause corresponding to the anomaly identification result is determined to be weed harm, under the condition that the anomaly condition is negative anomaly, the occurrence cause corresponding to the anomaly identification result is determined based on the related data of the monitored land, wherein the related data comprises at least one of meteorological data of the monitored land, monitoring data of the Internet of things equipment of the monitored land, ground inspection data of the monitored land and weather period information of crops to be identified corresponding to the monitored land, so that the occurrence cause corresponding to the anomaly identification result is accurately determined based on the multivariate data, a grower can not only know the condition of the field crops, but also can primarily grasp the occurrence cause causing the crop anomaly, and the accuracy of crop anomaly identification is further improved; meanwhile, the method and the device for identifying the crop anomalies in the crop comprise the steps of firstly determining the anomaly identification result of the crop to be identified, and then determining the occurrence reason corresponding to the anomaly identification result, so that identification is not carried out on one anomaly occurrence reason, various anomaly occurrence reasons are considered, and finally, the accuracy of identifying the crop anomalies is further improved.

Based on any of the above embodiments, the related data includes meteorological data of the monitored plot, internet of things equipment monitoring data of the monitored plot, ground inspection data of the monitored plot, and weather period information corresponding to the monitored plot by the crop to be identified, and the determining, based on the related data of the monitored plot, an occurrence cause corresponding to the abnormal identification result includes the following steps 210-250.

Step 210, determining first risk level data of a lodging reason based on the current moment, the weather period information and the meteorological data.

In a specific embodiment, determining the weather period of the crop to be identified at the current moment based on the current moment and the weather period information; under the condition that the physical period of the crop to be identified at the current moment is a seedling period, determining first risk degree data of lodging reasons as a first preset risk value; and determining first risk degree data of the lodging reason based on the meteorological data under the condition that the weather period of the crop to be identified at the current moment is the weather period after the seedling period.

In some embodiments, first risk level data of a cause of windy type lodging is determined based on the meteorological data, and/or first risk level data of a cause of windy type lodging is determined based on the meteorological data.

In one embodiment, a maximum wind speed is determined based on the meteorological data; determining the first risk degree data of the cause of the high wind type lodging as a third preset risk value under the condition that the maximum wind speed is greater than or equal to a first preset wind speed; determining the first risk degree data of the high wind type lodging reason as a second preset risk value under the condition that the maximum wind speed is larger than or equal to the second preset wind speed and smaller than the first preset wind speed; and determining first risk degree data of the high wind type lodging reason based on the maximum wind speed under the condition that the maximum wind speed is smaller than the second preset wind speed. Wherein the first preset wind speed is greater than the second preset wind speed, and the first preset wind speed is 16m/s and the second preset wind speed is 8m/s.

In a specific embodiment, when the maximum wind speed is smaller than the second preset wind speed, determining a ratio of the maximum wind speed to the second preset wind speed, and determining the first risk degree data of the high wind lodging reason based on a product value of the ratio and the second preset risk value. The calculation formula of the first risk level data is shown in the following exemplary embodiment:

；

in the method, in the process of the invention,for the first risk level data,/a>For maximum wind speed>For a second preset wind speed,/for >And (5) setting a risk value for the second preset. />

In an embodiment, the first preset risk value is smaller than the second preset risk value, and the second preset risk value is smaller than the third preset risk value. For example, the first preset risk value is 0, the second preset risk value is 0.8, and the third preset risk value is 1.

In one embodiment, a maximum wind speed and an accumulated rain amount are determined based on the meteorological data; determining first risk degree data of a wind and rain type lodging reason as a third preset risk value under the condition that the maximum wind speed is greater than or equal to a third preset wind speed and the accumulated rainfall is greater than or equal to a first preset accumulated rainfall; determining first risk degree data of a wind and rain type lodging reason based on the maximum wind speed under the condition that the maximum wind speed is smaller than a third preset wind speed and the accumulated rainfall is larger than or equal to a first preset accumulated rainfall; determining first risk degree data of a wind-rain type lodging reason based on the accumulated rainfall under the condition that the maximum wind speed is greater than or equal to a third preset wind speed and the accumulated rainfall is less than the first preset accumulated rainfall; in the case where the maximum wind speed is smaller than the third preset wind speed and the accumulated rainfall is smaller than the first preset accumulated rainfall, first risk degree data of a wind and rain type lodging cause is determined based on the accumulated rainfall and the maximum wind speed. The third preset wind speed is, for example, 5m/s, the first preset accumulated rainfall is 15mm, and the third preset risk value is 1.

In a specific embodiment, in a case where the maximum wind speed is smaller than the third preset wind speed and the accumulated rainfall is greater than or equal to the first preset accumulated rainfall, a ratio of the maximum wind speed to the third preset wind speed is determined, and the first risk degree data of the wind and rain type lodging reason is determined based on the ratio. The calculation formula of the first risk level data is shown in the following exemplary embodiment:

；

in the method, in the process of the invention,for the first risk level data,/a>For maximum wind speed>And a third preset wind speed.

In a specific embodiment, in a case where the maximum wind speed is greater than or equal to the third preset wind speed and the accumulated rainfall is less than the first preset accumulated rainfall, a ratio of the accumulated rainfall to the first preset accumulated rainfall is determined, and first risk degree data of the weather type lodging cause is determined based on the ratio. The calculation formula of the first risk level data is shown in the following exemplary embodiment:

；

in the method, in the process of the invention,for the first risk level data,/a>For accumulating rain, ->The accumulated rainfall is a first preset.

In a specific embodiment, when the maximum wind speed is smaller than the third preset wind speed and the accumulated rainfall is smaller than the first preset accumulated rainfall, a first ratio of the maximum wind speed to the third preset wind speed is determined, a second ratio of the accumulated rainfall to the first preset accumulated rainfall is determined, and based on a product value of the first ratio and the second ratio, first risk degree data of a weather type lodging reason is determined. The calculation formula of the first risk level data is shown in the following exemplary embodiment:

；

In the method, in the process of the invention,for the first risk level data,/a>For maximum wind speed>For a third preset wind speed,/>For accumulating rain, ->The accumulated rainfall is a first preset.

Step 220, determining second risk level data of the waterlogging cause based on the meteorological data.

In some embodiments, an accumulated rain amount is determined based on the meteorological data; and determining second risk level data of the waterlogging cause based on the accumulated rainfall.

In an embodiment, when the accumulated rainfall is greater than or equal to a second preset accumulated rainfall, determining second risk degree data of the waterlogging cause as a third preset risk value; under the condition that the accumulated rainfall is smaller than the second preset accumulated rainfall and is larger than or equal to the third preset accumulated rainfall, determining second risk degree data of waterlogging reasons as a fourth preset risk value; under the condition that the accumulated rainfall is smaller than the third preset accumulated rainfall and is larger than or equal to the fourth preset accumulated rainfall, determining second risk degree data of waterlogging reasons as a fifth preset risk value; and under the condition that the accumulated rainfall is smaller than the fourth preset accumulated rainfall, determining the second risk degree data of the waterlogging cause as a sixth preset risk value.

Wherein the second preset accumulated rainfall is greater than the third preset accumulated rainfall, the third preset accumulated rainfall is greater than the fourth preset accumulated rainfall, and the second preset accumulated rainfall is 50mm, the third preset accumulated rainfall is 33mm and the fourth preset accumulated rainfall is 25mm. The third preset risk value is greater than the fourth preset risk value, the fourth preset risk value is greater than the fifth preset risk value, the fifth preset risk value is greater than the sixth preset risk value, and illustratively, the third preset risk value is 1, the fourth preset risk value is 0.9, the fifth preset risk value is 0.8, and the sixth preset risk value is 0.

Step 230, determining third risk level data of the cause of the high temperature heat injury based on the current time, the weather period information and the meteorological data.

In a specific embodiment, determining the weather period of the crop to be identified at the current moment based on the current moment and the weather period information; determining a temperature threshold corresponding to a physical period from a physical period-temperature threshold mapping relation based on the physical period of the crop to be identified at the current moment; determining a daily maximum temperature for each day based on the meteorological data and a daily cumulative rainfall; and determining third risk degree data of the cause of the high-temperature heat damage based on the comparison result of the daily highest temperature and the temperature threshold value and the comparison result of the daily accumulated rainfall and the preset rainfall threshold value.

The mapping relationship between the physical period and the temperature threshold comprises a mapping relationship between a plurality of physical periods and a plurality of temperature thresholds. For example, when the waiting period of the crop to be identified at the current moment is a seedling period, the temperature threshold is 36 ℃; when the weather period of the crop to be identified at the current moment is the period from jointing to heading, the temperature threshold value is 32 ℃; when the weather period of the crop to be identified at the current moment is the mature period, the temperature threshold value is 28 ℃.

It is understood that if the weather data is weather data in an abnormality occurrence period corresponding to the abnormality identification result, the daily maximum temperature is determined based on the weather data, that is, the daily maximum temperature in the abnormality occurrence period, and the daily cumulative rainfall is determined based on the weather data, that is, the daily cumulative rainfall in the abnormality occurrence period.

In an embodiment, when the daily maximum temperature of the first preset day or more is greater than or equal to the temperature threshold value and the daily accumulated rainfall of the second preset day or more is less than the preset rainfall threshold value, determining the third risk degree data of the cause of the thermal injury as a third preset risk value; when there is no continuous first preset day or more in the daily maximum temperature, or there is no continuous second preset day or more in the daily cumulative rainfall, the daily cumulative rainfall is smaller than the preset rainfall threshold, and there is a third preset day or more in the daily cumulative rainfall, the first day number of the daily maximum temperature is larger than or equal to the temperature threshold, the second day number of the daily cumulative rainfall is larger than or equal to the preset rainfall threshold, the second day number is self-added, the product value of the second day after self-addition and the first preset day is determined, and the ratio of the first day to the product value is determined as the third risk degree data of the cause of high temperature injury; and determining a first day of the daily maximum temperatures being greater than or equal to a temperature threshold value, or determining a ratio of the first day to the first preset day as third risk degree data of a cause of high temperature injury, wherein if the ratio of the first day to the first preset day is greater than 1, the third risk degree data is 1, when there is no continuous first preset day or more in the daily maximum temperatures, or no continuous second preset day or more in the daily cumulative rainfall, less than the preset rainfall threshold value, and no third preset day or more in the daily cumulative rainfall is greater than or equal to the preset rainfall threshold value.

It is understood that when the daily accumulated rainfall is greater than or equal to the preset rainfall threshold, it is considered as one effective rainfall in consideration of the situation that the high temperature heat damage needs to evaluate the effective rainfall.

The threshold for the preset rainfall is 10mm, the first preset number of days is 5, the second preset number of days is 8, and the third preset number of days is 1. Assuming that the weather data is the weather data within 10 days, acquiring the air temperature and rainfall data of the last 10 days, and making the following statistics: when the highest temperature is more than or equal to Thresh (temperature threshold) for more than 5 continuous days and no effective rainfall exists for more than 8 continuous days, risk (third Risk degree data) =1; otherwise, when the number of effective rainfall is not less than 1, risk=the number of days at which the highest temperature of the days reaches Thresh/(5 (effective rainfall days+1)); otherwise, risk=the number of days/5 of which the highest temperature of the day reaches Thresh, and when Risk obtained in this way is greater than 1, risk is set to 1.

And step 240, determining fourth risk degree data of the pest cause based on the internet of things equipment monitoring data and the ground inspection data.

Specifically, based on the internet of things equipment monitoring data and the ground inspection data, insect pest data are determined, and based on the insect pest data, fourth risk degree data of pest causes are determined.

Here, the ground inspection data includes positional information (latitude and longitude) of the inspection site, and the occurrence of insect pests at each position. The internet of things equipment monitoring data comprises information of each position and insect condition data of each position. It can be understood that if the ground patrol data has no data at certain positions, the internet of things equipment monitoring data is utilized to make up, namely, the internet of things equipment monitoring data and the ground patrol data are integrated, and insect condition data of all positions in the monitored land are comprehensively determined. Further, the ground inspection data is considered to be more accurate, so that the ground inspection data is taken as the main part, namely, the insect condition data of a certain position is determined based on the ground inspection data under the condition that the ground inspection data and the internet of things monitoring data exist at the certain position.

In a specific embodiment, an interpolation map is generated based on pest situation data of each position, and fourth risk degree data of pest causes is determined based on the interpolation map. By way of example, generating an interpolation map using insect pest data at each location may obtain an interpolation map of fourth risk level data. The fourth risk degree data of the place where the insect pest occurs is confirmed to be 1 by the insect pest data, the risk value of other places where the fourth risk degree data is obtained through interpolation is determined by using a spatial distance nearest neighbor method, and the risk value is higher when the place where the risk value is 1 is closer, and otherwise, the risk value is lower.

Further, after fourth risk degree data corresponding to different pest and disease types in the monitored land are determined, the largest risk degree data is determined from the fourth risk degree data corresponding to different pest and disease types and used as the fourth risk degree data of the monitored land.

Step 250, determining a cause of occurrence corresponding to the abnormal recognition result based on the first risk level data, the second risk level data, the third risk level data and the fourth risk level data.

In an embodiment, the maximum risk degree data is determined from the first risk degree data, the second risk degree data, the third risk degree data and the fourth risk degree data, and a reason corresponding to the maximum risk degree data is determined as an occurrence reason corresponding to the abnormal recognition result. It is understood that, if there are a plurality of maximum risk level data, the causes corresponding to the plurality of maximum risk level data are all determined as the occurrence causes corresponding to the abnormality recognition result.

According to the crop anomaly identification method provided by the embodiment of the invention, the occurrence reason corresponding to the anomaly identification result can be accurately determined based on the multivariate data, so that a grower can not only know the condition of a field crop, but also primarily grasp the occurrence reason causing crop anomaly, thereby further improving the accuracy of crop anomaly identification.

Based on any one of the above embodiments, the method further includes the step 250:

Here, the preset risk level threshold may be set according to the actual situation, for example, 0.6.

It is understood that, if there are a plurality of maximum risk level data, the causes corresponding to the plurality of maximum risk level data are all determined as the occurrence causes corresponding to the abnormality recognition result.

For example, if the first risk level data is 0.8, the second risk level data is 1, the third risk level data is 0.6, and the fourth risk level data is 0.6, determining that the occurrence cause corresponding to the abnormal recognition result is the waterlogging cause.

According to the crop anomaly identification method provided by the embodiment of the invention, the reason for occurrence corresponding to the anomaly identification result is considered to be not the four known reasons, and based on the reason, under the condition that the first risk degree data, the second risk degree data, the third risk degree data and the fourth risk degree data are all smaller than or equal to the preset risk degree threshold value, the reason for occurrence corresponding to the anomaly identification result is determined to be other reasons, namely the reason for occurrence corresponding to the anomaly identification result is considered to be other than the four known reasons, so that the determination accuracy of the reason for occurrence corresponding to the anomaly identification result is improved, and the accuracy of crop anomaly identification is further improved; meanwhile, the reason corresponding to the maximum risk degree data is determined to be the occurrence reason corresponding to the abnormal recognition result, and the determination accuracy of the occurrence reason corresponding to the abnormal recognition result can be ensured.

Based on the embodiments, the invention can help a grower dynamically analyze the real-time condition of field crops and further provide analysis and guidance about problems occurring in the field under the condition of lacking priori knowledge. The invention also synthesizes the remote sensing data, the meteorological data, the weather period information, the offline patrol records and other multivariate data, comprehensively researches and judges the reasons causing the abnormality of crops, and finally provides guidance and suggestion for the farmers to land on production practice. Specifically, around the dynamic monitoring of the whole growth period of field crops, a reference curve (standard growth trend monitoring curve) of the growth process of the field crops is constructed through space-time two-dimensional analysis. The reference curve is completely adapted to the crop growth process of the current land by comprehensively analyzing the actual monitoring data of the crops in the field, so that the reference curve can be used as a standard curve for detecting the abnormal condition of the crop growth process in the field. The abnormal conditions extracted according to the reference curve are compared based on the change trend, so that misjudgment caused by the difference of crop growth vigor existing in the field is eliminated, and the abnormal conditions of crop growth which the farmer really should pay attention to are eliminated. Besides, on the basis of extracting high-value abnormal conditions, the invention also combines the meteorological data, the physical data, the offline personnel inspection record data and other multivariate data in the abnormal occurrence period to attribute the reasons for causing the abnormal conditions, so that a grower can not only know the conditions of field crops, but also primarily grasp the reasons for causing different conditions of the crops.

The crop anomaly recognition device provided by the invention is described below, and the crop anomaly recognition device described below and the crop anomaly recognition method described above can be referred to correspondingly to each other.

Fig. 4 is a schematic structural diagram of a crop anomaly identification apparatus according to the present invention, as shown in fig. 4, the crop anomaly identification apparatus includes:

the curve acquisition module 410 is configured to acquire an actual growth trend monitoring curve of a crop to be identified at a current time, and acquire a standard growth trend monitoring curve corresponding to the crop to be identified, where the actual growth trend monitoring curve and the standard growth trend monitoring curve are the same growth trend monitoring curve of a monitored plot, and the growth trend monitoring curve is used for representing a growth variation trend of the crop to be identified;

an index determining module 420, configured to determine a growth trend difference index based on a comparison result of the actual growth trend monitoring curve and the standard growth trend monitoring curve, where the growth trend difference index characterizes a degree of difference between a change rate of the actual growth trend monitoring curve and a change rate of the standard growth trend monitoring curve;

the result determining module 430 is configured to determine an abnormal recognition result of the crop to be recognized on the monitored plot based on a comparison result of the growth trend difference index and at least one difference index threshold.

Fig. 5 illustrates a physical schematic diagram of an electronic device, as shown in fig. 5, which may include: processor 510, communication interface (Communications Interface) 520, memory 530, and communication bus 540, wherein processor 510, communication interface 520, memory 530 complete communication with each other through communication bus 540. Processor 510 may invoke logic instructions in memory 530 to perform a crop anomaly identification method comprising: acquiring an actual growth trend monitoring curve of a crop to be identified at a current time, and acquiring a standard growth trend monitoring curve corresponding to the crop to be identified, wherein the actual growth trend monitoring curve and the standard growth trend monitoring curve are the same growth trend monitoring curve of a monitoring land block, and the growth trend monitoring curve is used for representing the growth variation trend of the crop to be identified; determining a growth trend difference index based on the comparison result of the actual growth trend monitoring curve and the standard growth trend monitoring curve, wherein the growth trend difference index represents the degree of difference between the change rate of the actual growth trend monitoring curve and the change rate of the standard growth trend monitoring curve; and determining an abnormal identification result of the crop to be identified in the monitored plot based on a comparison result of the growth trend difference index and at least one difference index threshold.

Further, the logic instructions in the memory 530 described above may be implemented in the form of software functional units and may be stored in a computer-readable storage medium when sold or used as a stand-alone product. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.

In yet another aspect, the present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, is implemented to perform the crop anomaly identification method provided by the above methods, the method comprising: acquiring an actual growth trend monitoring curve of a crop to be identified at a current time, and acquiring a standard growth trend monitoring curve corresponding to the crop to be identified, wherein the actual growth trend monitoring curve and the standard growth trend monitoring curve are the same growth trend monitoring curve of a monitoring land block, and the growth trend monitoring curve is used for representing the growth variation trend of the crop to be identified; determining a growth trend difference index based on the comparison result of the actual growth trend monitoring curve and the standard growth trend monitoring curve, wherein the growth trend difference index represents the degree of difference between the change rate of the actual growth trend monitoring curve and the change rate of the standard growth trend monitoring curve; and determining an abnormal identification result of the crop to be identified in the monitored plot based on a comparison result of the growth trend difference index and at least one difference index threshold.

The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.

From the above description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course may be implemented by means of hardware. Based on this understanding, the foregoing technical solution may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the respective embodiments or some parts of the embodiments.

Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for identifying crop anomalies, comprising:

2. The method for identifying crop anomalies according to claim 1, wherein the acquiring an actual growth trend monitoring curve of the crop to be identified at the current time, includes:

3. The method for identifying abnormal crop according to claim 2, wherein the determining the actual growth trend monitoring curve corresponding to the growth period of the crop to be identified from the vegetation index monitoring curve based on the weather period information corresponding to the crop to be identified in the monitored plot comprises:

4. The method for identifying crop anomalies according to claim 1, wherein the obtaining a standard growth trend monitoring curve corresponding to the crop to be identified includes:

5. The crop anomaly identification method of any one of claims 1 to 4, wherein the determining the anomaly identification result for the crop to be identified at the monitored site based on the comparison of the growth trend difference index to at least one difference index threshold value is followed by:

6. The method for identifying crop anomalies according to claim 5, wherein the determining an anomaly corresponding to the anomaly identification result based on the comparison of the actual growth trend monitoring curve with the standard growth trend monitoring curve, further comprises:

7. The method for identifying abnormal crop according to claim 6, wherein the related data includes meteorological data of the monitored plot, internet of things equipment monitoring data of the monitored plot, ground patrol data of the monitored plot, and weather period information of the crop to be identified corresponding to the monitored plot, and the determining, based on the related data of the monitored plot, an occurrence cause corresponding to the abnormal identification result includes:

8. The crop anomaly identification method of claim 7, wherein the determining the cause of the anomaly identification result based on the first risk level data, the second risk level data, the third risk level data, and the fourth risk level data comprises:

9. A crop anomaly identification apparatus, comprising:

10. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the crop anomaly identification method of any one of claims 1 to 8 when the program is executed by the processor.

11. A non-transitory computer readable storage medium having stored thereon a computer program, wherein the computer program when executed by a processor implements the crop anomaly identification method of any one of claims 1 to 8.