Crop yield loss assessment method and device, electronic equipment and storage medium

Abstract

The invention relates to the technical field of deep learning and the field of financial science and technology, and discloses a crop yield loss assessment method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining a regional cluster image and a single cluster image of a to-be-assessed crop; processing the regional cluster image by using the trained cluster number prediction model to obtain the total cluster number per unit area; processing a single ear image by using the trained ear grain number prediction model to obtain the ear grain number; calculating a unit area yield prediction value according to the unit area cluster number and the unit area grain number; and obtaining a yield loss assessment result according to deviation degree data between the unit area yield prediction value and a preset historical yield value. The cluster number prediction model and the cluster number prediction model are utilized to calculate the accurate cluster number per unit area and the cluster number of a single cluster, the yield prediction value per unit area is further calculated, yield loss assessment is carried out based on the preset historical yield value, and an accurate and reliable yield loss assessment result is obtained.

Description

Technical Field

[0001] The present disclosure relates to the fields of deep learning technology and financial technology, and in particular to a method, device, electronic device and storage medium for assessing crop yield losses. Background Art

[0002] In the modern agricultural market environment, whether it is the agricultural insurance industry, agricultural product processing enterprises, or government agricultural management departments, there is an urgent need for crop yield loss assessment. Specifically, agricultural insurance companies need to formulate reasonable insurance compensation amounts based on accurate loss assessment data; agricultural product processing enterprises need to arrange procurement plans based on loss assessment; government departments need loss assessment information to formulate agricultural policies and allocate resources. In this context, studying a scientific method for crop yield loss assessment has important practical significance.

[0003] The traditional method of assessing crop yield loss mainly relies on the subjective cognition and past experience of the assessor. This method has many disadvantages: on the one hand, the efficiency of manual experience-based loss assessment is extremely low, and the assessor needs to spend a lot of time observing and estimating in the fields. On the other hand, it is difficult to guarantee a high-accuracy loss assessment result every time by relying solely on naked eye observation and experience judgment. Summary of the invention

[0004] In view of the above situation, the embodiments of the present disclosure provide a method, device, electronic device and storage medium for assessing crop yield loss, aiming to solve the technical problems of low efficiency and low accuracy in assessing crop yield loss in the prior art.

[0005] In a first aspect, an embodiment of the present disclosure provides a method for assessing crop yield loss, the method comprising:

[0006] Acquire regional ear images and single ear images of crops to be assessed for damage;

[0007] The trained ear number prediction model is used to process the regional ear image to obtain the total number of ears; and the number of ears per unit area is calculated according to the total number of ears and the shooting area of ​​the regional ear image;

[0008] Using the trained grain number per ear prediction model, the single ear image is processed to obtain the grain number per ear;

[0009] Calculating the predicted yield per unit area of ​​the crop to be determined for damage according to the number of ears per unit area and the number of grains per ear;

[0010] According to the deviation degree data between the predicted yield per unit area and the preset historical yield value, the yield loss of the crops to be assessed is assessed to obtain a yield loss assessment result.

[0011] In a second aspect, the present disclosure also provides a crop yield loss assessment device, the device comprising:

[0012] An acquisition module, used for acquiring regional ear images and single ear images of crops to be assessed for damage;

[0013] The ear number prediction module is used to process the regional ear image using the trained ear number prediction model to obtain the total number of ears; and calculate the number of ears per unit area according to the total number of ears and the shooting area of ​​the regional ear image;

[0014] The grain number per ear prediction module is used to process the single ear image using the trained grain number per ear prediction model to obtain the grain number per ear;

[0015] A prediction module, used for calculating the predicted yield per unit area of ​​the crop to be determined for damage according to the number of ears per unit area and the number of grains per ear;

[0016] The damage assessment module is used to assess the yield damage of the crops to be assessed based on the deviation degree data between the predicted yield per unit area and the preset historical yield value, and obtain the yield damage assessment result.

[0017] In a third aspect, an embodiment of the present disclosure further provides an electronic device, comprising: a processor; and a memory arranged to store computer executable instructions, wherein when the executable instructions are executed, the processor executes the steps of the above-mentioned crop yield loss assessment method.

[0018] In a fourth aspect, an embodiment of the present disclosure further provides a computer-readable storage medium, which stores one or more programs. When the one or more programs are executed by an electronic device including multiple application programs, the electronic device executes the steps of the above-mentioned crop yield loss assessment method.

[0019] By means of the above technical scheme, the crop yield loss assessment method, device, electronic device and storage medium provided by the embodiments of the present disclosure can accurately calculate the number of ears per unit area and the number of grains per ear of a single ear by acquiring the regional ear image and the single ear image of the crop to be assessed, and using the trained ear number prediction model and the ear grain number prediction model to process them respectively. On this basis, the yield prediction value per unit area is further calculated. Then, the yield loss is assessed based on the deviation degree data between the yield prediction value and the preset historical yield value, and an accurate and reliable yield loss assessment result is obtained. It can be seen that the embodiments of the present disclosure provide a scientific loss assessment method, which can effectively reduce the error of human estimation, significantly improve the efficiency and accuracy of loss assessment, and provide scientific, objective and valuable data support for relevant decisions such as claim settlement and loss assessment.

[0020] The above description is only an overview of the technical solution of the present disclosure. In order to more clearly understand the technical means of the present disclosure, it can be implemented according to the contents of the specification. In order to make the above and other purposes, features and advantages of the present disclosure more obvious and easy to understand, the specific implementation methods of the present disclosure are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The drawings described herein are used to provide a further understanding of the present disclosure and constitute a part of the present disclosure. The illustrative embodiments of the present disclosure and their descriptions are used to explain the present disclosure and do not constitute an improper limitation on the present disclosure. In the drawings:

[0022] Figure 1 A schematic diagram of an application environment of the crop yield loss assessment method provided by an embodiment of the present disclosure is shown;

[0023] Figure 2 A schematic diagram of the process of the crop yield loss assessment method provided by an embodiment of the present disclosure is shown;

[0024] Figure 3 The target area ear image provided by the embodiment of the present disclosure is shown;

[0025] Figure 4 A single ear image provided by an embodiment of the present disclosure is shown;

[0026] Figure 5 A schematic diagram of the structure of a crop yield loss assessment device provided by an embodiment of the present disclosure is shown;

[0027] Figure 6 A schematic structural diagram of an electronic device provided by an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical solutions and advantages of the present disclosure clearer, the technical solutions of the present disclosure will be clearly and completely described below in combination with the specific embodiments of the present disclosure and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of the present disclosure, not all of the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present disclosure.

[0029] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0030] It should be noted that the terms "first", "second", etc. in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that such usage is interchangeable where appropriate, so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. In addition, the term "including" and its variants are to be interpreted as open-ended terms meaning "including but not limited to".

[0031] As mentioned above, the traditional method of assessing crop yield damage mainly relies on the subjective cognition and past experience of the assessor. This method has many disadvantages: on the one hand, the efficiency of manual experience-based damage assessment is extremely low, and the assessors need to spend a lot of time observing and estimating in the fields. On the other hand, it is difficult to guarantee a high-accuracy damage assessment result every time by relying solely on naked eye observation and experience. Based on this, the present invention proposes a method for assessing crop yield damage, and the present invention is described in detail below through specific embodiments.

[0032] To facilitate understanding of this embodiment, the crop yield loss assessment method disclosed in the embodiment of the present disclosure is first introduced in detail. The crop yield loss assessment method provided in the embodiment of the present disclosure can be applied to Figure 1In the application environment, the client communicates with the server through the network. The server can receive and obtain the regional ear image and the single ear image of the crop to be determined through the client; use the trained ear number prediction model to process the regional ear image to obtain the total number of ears; and calculate the number of ears per unit area according to the total number of ears and the shooting area of ​​the regional ear image; use the trained ear number prediction model to process the single ear image to obtain the number of grains per ear; calculate the unit area yield prediction value of the crop to be determined according to the number of ears per unit area and the number of grains per ear; according to the deviation degree data between the unit area yield prediction value and the preset historical yield value, the yield of the crop to be determined is determined, and the yield determination result is obtained, and the yield determination result is fed back to the client. In the present invention, by obtaining the regional ear image and the single ear image of the crop to be determined, and using the trained ear number prediction model and the ear number prediction model to process them respectively, the number of ears per unit area and the number of grains per ear of a single ear can be accurately calculated. On this basis, the yield prediction value per unit area is further calculated. Then, the output loss is determined based on the deviation data between the output forecast value and the preset historical output value, and an accurate and reliable output loss determination result is obtained. It can be seen that the embodiment of the present disclosure provides a scientific loss determination method, which can effectively reduce the error of human estimation, significantly improve the efficiency and accuracy of loss determination, and provide scientific, objective and valuable data support for relevant decisions such as claim loss determination. Among them, the client can be but not limited to various personal computers, laptops, smart phones, tablet computers and portable wearable devices. The server is generally a computer device with a certain computing power, and the computer device includes, for example: a terminal device or a server or other processing device, wherein the terminal device can be a user equipment (User Equipment, UE), a mobile device, a user terminal, a terminal, a personal digital assistant (Personal Digital Assistant, PDA), a handheld device, a computing device, etc. The server can be implemented as an independent server or a server cluster composed of multiple servers. In some possible implementations, the method can be implemented by a processor calling a computer-readable instruction stored in a memory. The present invention is described in detail below through specific embodiments.

[0033] Figure 2 The schematic diagram of the process of the crop yield loss assessment method provided by the embodiment of the present disclosure is shown. Figure 2 It can be seen that the embodiment of the present disclosure at least includes steps S201-S205:

[0034] S201: Acquire regional ear images and single ear images of crops to be assessed for damage.

[0035] The crop yield loss assessment method provided by the embodiment of the present disclosure can be applied to the yield loss assessment of various crops that have ears and grains in their structures. The types of crops to be assessed include, but are not limited to, cereal crops, legume crops, etc. Among them, cereal crops include, but are not limited to, wheat, rice, corn, etc. Legume crops include, but are not limited to, soybeans, soy beans, etc.

[0036] The regional ear image shows the distribution of ears of multiple crops within a certain geographical area. The single ear image shows the morphology and detailed features of a single ear. The disclosed embodiments do not limit the means for obtaining the regional ear image and the single ear image. For example, a farmer or damage assessor can use a mobile device (such as a mobile phone or tablet computer) to take an image and then upload the image to an interface specified by the execution subject of the method, such as a server, through a network connection.

[0037] S202: using the trained ear number prediction model, processing the regional ear image to obtain the total number of ears; and calculating the number of ears per unit area according to the total number of ears and the shooting area of ​​the regional ear image.

[0038] The models on which the trained ear number prediction model and the subsequent trained ear grain number prediction model are based can be a YOLO (You Only Look Once) model, a Faster R-CNN (Region-based Convolutional Neural Network) model, etc., which are not limited in this embodiment. Preferably, the pre-trained large model can be retrained to obtain the model.

[0039] S203: using the trained model for predicting the number of grains per ear, the single ear image is processed to obtain the number of grains per ear.

[0040] Specifically, a single ear image can be first input into a trained ear grain number prediction model, and the model outputs the number of grains on the front side of the ear. Then, it can be assumed that the number of grains on the front and back sides of the ear is equal, and the number of grains on a single ear is twice the model output result. Here, the quantitative relationship between the model output result and the number of grains on a single ear can be set according to actual needs.

[0041] S204: Calculate the predicted yield per unit area of ​​the crop to be determined for damage according to the number of ears per unit area and the number of grains per ear.

[0042] Specifically, the number of ears per unit area, the number of grains per ear, the preset thousand-grain weight and the preset water content ratio can be multiplied to obtain the predicted yield per unit area. The preset thousand-grain weight and the preset water content ratio can be set according to actual conditions.

[0043] S205: According to the deviation degree data between the predicted yield per unit area and the preset historical yield value, the yield loss of the crop to be determined is determined to obtain a yield loss determination result.

[0044] Here, the preset historical yield value can be set according to the actual situation, and the embodiment of the present disclosure does not limit this. For example, if the crop to be determined as damaged seed is winter wheat, the existing national winter wheat historical per-acre yield data set (resolution of 30m*30m) can be used in combination with longitude and latitude positioning technology to accurately obtain the historical yield data of the plot, and determine the preset historical yield value based on each historical yield data.

[0045] from Figure 1 It can be seen from the method shown that the disclosed embodiment obtains the regional ear image and the single ear image of the crop to be assessed for damage, and processes them respectively using the trained ear number prediction model and the ear grain number prediction model, so as to accurately calculate the number of ears per unit area and the number of grains per ear per single ear. On this basis, the yield forecast value per unit area is further calculated. Then, the yield loss is assessed based on the deviation degree data between the yield forecast value and the preset historical yield value, and an accurate and reliable yield loss assessment result is obtained. It can be seen that the disclosed embodiment provides a scientific loss assessment method, which can effectively reduce the error of human estimation, significantly improve the efficiency and accuracy of loss assessment, and provide scientific, objective and valuable data support for relevant decisions such as claim settlement and loss assessment.

[0046] Furthermore, in order to better illustrate the process of the above-mentioned crop yield loss assessment method, as a refinement and extension of the above-mentioned embodiment, the embodiment of the present invention provides several optional embodiments, but is not limited thereto.

[0047] In a possible implementation, the trained ear number prediction model is generated according to the following method: obtaining a data set, the data set including an open source data set and an actual data set; the data set includes multiple ear images of a sample area, the label of each sample is the number of ears, the average resolution and / or average shooting angle height of the sample images in the actual data set is smaller than the average resolution and / or average shooting height of the sample images in the open source data set; using the data set to train an initial ear number prediction model to obtain the trained ear number prediction model.

[0048] In this embodiment, for open source data sets, for example, if the crop to be determined is wheat, the open source data set can be the Global Wheat Head Detection Data Set (Global Wheat Head Detection 2021), which contains wheat ear images from 7 countries, taken by a high-definition camera 1.8-3m above the ground, with a shooting angle ranging from 10 degrees to 45 degrees, and the collection period is from flowering to maturity. For the actual data set, the collection conditions (collection equipment, shooting height, etc.) of the samples in the actual data set are the same as the collection conditions when the user collects the images to be predicted in the field when the trained ear grain number detection model is put into use later. Based on this, the average resolution, average shooting height, etc. of the images in the actual data set will be smaller than the average resolution and average shooting height of the images in the open source data set.

[0049] Then, you can select an initial ear number prediction model, such as the pre-trained large model YOLOv8, input the preprocessed image data into the large model, and adjust the model parameters through the back-propagation algorithm to minimize the prediction error. During implementation, you can use an open source dataset for preliminary training, and then merge the actual dataset with the open source dataset for further training. During the training process, use a portion of the data as a validation set to monitor model performance and avoid overfitting. After training is completed, use an independent test set to evaluate the generalization ability of the model. Based on the results of the validation set and test set, tune the model, which may include adjusting the network structure, learning rate, optimizer, etc.

[0050] This embodiment uses open source data sets to train the initial model, which can enable the model to quickly acquire the initial recognition ability of fruit ears and the ability to extract general features. Combined with the actual data sets collected in the field to train the model, the model can fully adapt to various complex situations in actual scenes, thereby effectively improving the accuracy and robustness of the model for fruit ear detection, thereby constructing a higher performance fruit ear number prediction model.

[0051] In a possible implementation, the shooting area is calculated according to the following method: obtaining a target area ear image having the same shooting area as the regional ear image, wherein the target area ear image includes a first calibration plate; obtaining an actual area of ​​the first calibration plate; calculating a pixel area of ​​the first calibration plate; calculating a proportional factor based on the actual area and the pixel area; and calculating the shooting area based on the proportional factor and the pixel area of ​​the target area ear image.

[0052] This embodiment does not limit the shape, color and other attributes of the first calibration plate. The shape of the first calibration plate can be square, circular, rectangular, irregular, etc. As for the color of the first calibration plate, preferably, it can be white or other colors that are very different from the color of crops. The actual area of ​​the first calibration plate can be directly input by the user through the front-end interface. The parameters input by the user can also be calculated. For example, if the first calibration plate is a square calibration plate of 14cm*14cm, the actual area of ​​the calibration plate can be calculated as 14cm*14cm=196cm 2 . Then the pixel area of ​​the first calibration plate can be calculated. In specific implementation, for example, an image recognition algorithm and an edge detection algorithm can be used to determine the outline of the first calibration plate, thereby calculating the pixel area. After obtaining the actual area and pixel area of ​​the first calibration plate, the proportional factor is calculated based on the actual area and pixel area. Finally, the shooting area is calculated based on the proportional factor and the pixel area of ​​the ear image of the target area. For example, the actual area / pixel area of ​​the first calibration plate is used as the proportional factor, and then the proportional factor is multiplied by the pixel area of ​​the ear image of the target area to obtain the shooting area.

[0053] Figure 3 FIG. 2 shows an image of a fruit ear in a target area provided by an embodiment of the present disclosure. For example, the first calibration plate is a 14 cm*14 cm QR code calibration plate. Figure 3 The actual area of ​​the QR code calibration plate is 196cm 2 The image recognition algorithm can be used to process the acquired ear image of the target area, extract the first calibration plate area, and then determine its outline through edge detection, and calculate that the first calibration plate occupies 60 pixels in the image. Scale factor = 196cm 2 / 60 pixels = 3.3 square centimeters / pixel. Through image processing technology, it is found that the pixel area of ​​the whole target area ear image is 900 pixels. Then the shooting area = 3.3 square centimeters / pixel × 900 pixels = 2970 square centimeters = 0.297 square meters.

[0054] This embodiment does not rely on professional equipment, such as area measuring instruments, levels and other tools, but only uses a calibration plate. By obtaining the fruit ear image of the target area including the calibration plate, and then obtaining the actual area of ​​the calibration plate, the pixel area is calculated, and then the proportional factor is obtained. Finally, the proportional factor and the pixel area of ​​the fruit ear image of the target area can be used to conveniently, accurately and efficiently calculate the shooting area.

[0055] In a possible implementation manner, the regional ear image is obtained according to the following method: receiving the ear image of the area to be verified and the ear image of the area to be processed sent by a user; the difference between the ear image of the area to be processed and the ear image of the area to be verified is that the ear image of the area to be processed also includes a second calibration plate, and the second calibration plate has multiple logos; using a preset logo extraction algorithm, processing the ear image of the area to be processed to obtain the position data of the multiple logos; based on the position data of the multiple logos and preset judgment conditions, judging whether the shooting angle of the ear image of the area to be processed is horizontal; if the shooting angle of the ear image of the area to be processed is horizontal, using the ear image of the area to be verified as the regional ear image; if the shooting angle of the ear image of the area to be processed is not horizontal, generating and displaying a prompt that the image does not meet the requirements, and returning to the step of receiving the ear image of the area to be verified and the ear image of the area to be processed sent by the user.

[0056] In the present embodiment, the fruit ear image of the area to be verified and the fruit ear image of the area to be processed sent by the user through the terminal can be received first. The shooting area, shooting angle, shooting height, etc. of the two images are all the same, and the only difference is that the fruit ear image of the area to be processed also includes a second calibration plate. Based on this, it is judged whether the shooting angle of the fruit ear image of the area to be processed is horizontal, and it can be indirectly judged whether the shooting angle of the fruit ear image of the area to be verified is horizontal.

[0057] The preset identification extraction algorithm can be used to process the ear image of the processing area to obtain the position data of each identification. Here, the preset identification extraction algorithm can be Harris corner detection algorithm, target detection algorithm based on deep learning, Hough transform, etc., which is not limited in the embodiment of the present disclosure.

[0058] Next, according to the position data of multiple marks and the preset judgment conditions, it is judged whether the shooting angle of the ear image of the area to be processed is horizontal. Here, the preset judgment condition can be customized according to the actual situation, and the embodiment of the present disclosure is not limited to this. For example, it is assumed that three marking points are set on the second calibration plate to form a triangle. When the shooting angle is horizontal, the internal angle of the triangle is fixed. Then the preset judgment condition at this time can be: if the deviation between any internal angle calculated according to the position data of the three marks and the angle in the horizontal state exceeds the preset threshold, then it is judged that the shooting angle of the ear image of the area to be processed is not horizontal; otherwise, it is judged that the shooting angle of the ear image of the area to be processed is horizontal. For another example, it is assumed that four marks are set on the second calibration plate to form a quadrilateral. In the horizontal state, the ratio of the lengths of the adjacent sides of the quadrilateral is fixed. Then the preset judgment condition at this time can be: if the deviation between the ratio of the lengths of the adjacent sides calculated according to the position data of each mark and the ratio of the lengths of the adjacent sides in the horizontal state exceeds the preset threshold, then it is judged that the shooting angle of the ear image of the area to be processed is not horizontal; otherwise, it is judged that the shooting angle of the ear image of the area to be processed is horizontal.

[0059] Finally, if the shooting angle of the fruit ear image of the area to be processed is in a horizontal state, the fruit ear image of the area to be verified is used as the regional fruit ear image; if the shooting angle of the fruit ear image of the area to be processed is not in a horizontal state, a prompt that the image does not meet the requirements is generated and displayed on the front-end interface. Then, the fruit ear image of the area to be verified and the fruit ear image of the area to be processed sent by the user are continuously received, and whether the shooting angle of the fruit ear image of the area to be processed is horizontal is judged.

[0060] This embodiment first determines whether the shooting angle of the ear image of the area to be verified sent by the user is horizontal. Only when the shooting angle of the ear image of the area to be verified is horizontal, the ear image of the area to be verified is used as the regional ear image for subsequent ear number prediction. The horizontally shot image can provide a more consistent and standard viewing angle, reduce the problems of ear shape distortion, overlap and occlusion caused by shooting angle deviation, thereby ensuring the accuracy and reliability of subsequent ear number prediction.

[0061] In actual application scenarios, there may be noise and other ears in the captured single ear image, which may affect the accuracy of the ear-grain prediction model. Based on this, in a possible implementation of the present disclosure, the single ear image contains a background for highlighting the ears in the single ear image. Here, the background is used to highlight the ears, and the background includes but is not limited to: a background board, a human hand, the sky, etc. Figure 4 The single ear image provided by the embodiment of the present disclosure is shown. It can be seen that the white background board is placed behind the ear of wheat as the background. This embodiment effectively improves the detection accuracy of the model by strengthening the background of the single ear of wheat.

[0062] In a possible implementation, the yield loss of the crops to be assessed is assessed based on the deviation degree data between the predicted yield per unit area and the preset historical yield value to obtain a yield loss assessment result, including: calculating the deviation degree data based on the predicted yield per unit area and the preset historical yield value; and obtaining the yield loss assessment result based on the deviation degree data based on a preset correspondence rule between the deviation degree and the loss level; wherein the yield loss assessment results include: no disaster, light disaster, moderate disaster, severe disaster, and total loss.

[0063] In this embodiment, the deviation degree data can be calculated based on the unit area yield prediction value and the preset historical yield value. Specifically, the unit area yield prediction value can be subtracted from the preset historical yield value to obtain the deviation degree data. Alternatively, the unit area yield prediction value can be subtracted from the preset historical yield value to obtain the difference, and the difference / preset historical yield value×100% is used as the deviation degree data.

[0064] After obtaining the deviation degree data, the output loss assessment result can be determined based on the preset deviation degree and the corresponding rule of the loss assessment result. Here, the preset deviation degree and the corresponding rule of the loss assessment result can be set according to actual needs. For example, if the deviation degree data is in the form of a percentage, the preset deviation degree and the corresponding rule of the loss assessment result can be: if the deviation degree data is greater than -5%, the output loss assessment result is "no disaster"; if the deviation degree data is between -15% and -5%, the output loss assessment result is "minor disaster"; if the deviation degree data is between -30% and -15%, the output loss assessment result is "medium disaster", if the deviation degree data is between -50% and -30%, the output loss assessment result is "severe disaster", if the deviation degree data is less than -50%, the output loss assessment result is "no production".

[0065] This embodiment achieves standardization and precision in crop yield damage assessment through accurate deviation degree calculation combined with clear corresponding rules. The unified preset rules eliminate the influence of human factors, ensure the fairness and objectivity of damage assessment, and make the damage assessment results more in line with the actual disaster situation. At the same time, clear steps and standards reduce uncertainty and improve the efficiency of damage assessment, so that damage assessment conclusions can be drawn quickly.

[0066] The present disclosure also provides a method for assessing crop yield loss, comprising the following steps:

[0067] Step S1: using the acquired global wheat ear dataset and actual dataset to train the initial ear number prediction model to obtain a trained ear number prediction model.

[0068] Step S2: receiving the ear image of the area to be verified and the ear image of the area to be processed sent by the user; the difference between the ear image of the area to be processed and the ear image of the area to be verified is that the ear image of the area to be processed also includes a calibration plate with multiple marks on it.

[0069] Step S3: using a preset identifier extraction algorithm to process the ear image of the processing area to obtain position data of multiple identifiers.

[0070] Step S4: Based on the position data of multiple identifiers and preset judgment conditions, determine whether the shooting angle of the fruit ear image of the area to be processed is horizontal; if the shooting angle of the fruit ear image of the area to be processed is horizontal, the fruit ear image of the area to be verified is used as the regional fruit ear image; if the shooting angle of the fruit ear image of the area to be processed is not horizontal, generate and display a prompt that the image does not meet the requirements, and return to step S2.

[0071] Step S5: acquiring a target region ear image having the same shooting area as the regional ear image, wherein the target region ear image includes a calibration plate.

[0072] Step S6: Obtain the actual area of ​​the calibration plate.

[0073] Step S7: Calculate the pixel area of ​​the calibration plate.

[0074] Step S8: Calculate the scale factor according to the actual area and the pixel area.

[0075] Step S9: Calculate the shooting area according to the scale factor and the pixel area of ​​the ear image in the target area.

[0076] Step S10: Acquire a regional ear image and a single ear image of the crop to be assessed for damage. The single ear image includes a white background for highlighting the ears in the single ear image.

[0077] Step S11: using the trained ear number prediction model, processing the regional ear image to obtain the total number of ears; and calculating the number of ears per mu according to the total number of ears and the shooting area of ​​the regional ear image.

[0078] Step S12: using the trained grain number per ear prediction model, the image of a single ear is processed to obtain the grain number per ear.

[0079] Step S13: Calculate the predicted per-acre yield of the crop to be determined for damage based on the number of ears per acre and the number of grains per ear.

[0080] Step S14: Calculate the degree of deviation data based on the per-acre yield forecast value and the preset historical yield value.

[0081] Step S15: Based on the preset deviation degree and loss level correspondence rule, the production loss assessment result is obtained according to the deviation degree data.

[0082] Those skilled in the art will appreciate that, in the above method of the specific implementation manner, the writing order of the steps does not mean a strict execution order, but does not constitute any limitation on the implementation process. The specific execution order of the steps should be determined by their functions and possible internal logic.

[0083] It should be noted that, in practical applications, all possible implementations described above can be combined in any manner to form possible embodiments of the present disclosure, which will not be described one by one here.

[0084] Based on the same concept, the embodiment of the present disclosure also provides a crop yield loss assessment device. Figure 5 The structure diagram of the crop yield loss assessment device provided by the embodiment of the present disclosure is shown in FIG. Figure 5 As shown, the crop yield loss assessment device 500 provided in the embodiment of the present disclosure includes:

[0085] An acquisition module 501 is used to acquire regional ear images and single ear images of crops to be determined for damage;

[0086] The ear number prediction module 502 is used to process the regional ear image using the trained ear number prediction model to obtain the total number of ears; and calculate the number of ears per unit area according to the total number of ears and the shooting area of ​​the regional ear image;

[0087] The grain number per ear prediction module 503 is used to process the single ear image using the trained grain number per ear prediction model to obtain the grain number per ear;

[0088] A prediction module 504 is used to calculate the predicted yield per unit area of ​​the crop to be determined based on the number of ears per unit area and the number of grains per ear;

[0089] The damage assessment module 505 is used to assess the yield damage of the crop to be assessed based on the deviation degree data between the predicted yield per unit area and the preset historical yield value, and obtain the yield damage assessment result.

[0090] In a possible implementation, in the above device, the type of the crop to be determined as damaged is wheat.

[0091] In a possible implementation, the device also includes a training module, which is used to: obtain a data set, the data set including an open source data set and an actual data set; the data set includes multiple fruit ear images of a sample area, the label of each sample is the number of fruit ears, and the average resolution and / or average shooting angle height of the sample images in the actual data set are smaller than the average resolution and / or average shooting height of the sample images in the open source data set; use the data set to train an initial fruit ear number prediction model to obtain the trained fruit ear number prediction model.

[0092] In a possible implementation, the device further includes an area calculation module, which is used to: obtain a target area ear image whose shooting area is the same as the regional ear image, wherein the target area ear image includes a first calibration plate; obtain an actual area of ​​the first calibration plate; calculate a pixel area of ​​the first calibration plate; calculate a proportional factor based on the actual area and the pixel area; and calculate the shooting area based on the proportional factor and the pixel area of ​​the target area ear image.

[0093] In a possible implementation, in the above-mentioned device, the acquisition module is used to: receive the fruit ear image of the area to be verified and the fruit ear image of the area to be processed sent by the user; the difference between the fruit ear image of the area to be processed and the fruit ear image of the area to be verified is that the fruit ear image of the area to be processed also includes a second calibration plate, and the second calibration plate has multiple logos; use a preset logo extraction algorithm to process the fruit ear image of the area to be processed to obtain the position data of the multiple logos; based on the position data of the multiple logos and preset judgment conditions, judge whether the shooting angle of the fruit ear image of the area to be processed is horizontal; if the shooting angle of the fruit ear image of the area to be processed is horizontal, then use the fruit ear image of the area to be verified as the regional fruit ear image; if the shooting angle of the fruit ear image of the area to be processed is not horizontal, generate and display a prompt that the image does not meet the requirements, and return to the step of receiving the fruit ear image of the area to be verified and the fruit ear image of the area to be processed sent by the user.

[0094] In a possible implementation manner, in the above device, the single ear of fruit image includes a background for highlighting the ear of fruit in the single ear of fruit image.

[0095] In a possible implementation, in the above-mentioned device, the damage assessment module is used to: calculate the deviation degree data based on the unit area yield forecast value and the preset historical yield value; based on the preset deviation degree and loss level correspondence rule, obtain the yield damage assessment result according to the deviation degree data; wherein the yield damage assessment result includes: no disaster, light disaster, moderate disaster, severe disaster, and complete loss.

[0096] It should be noted that any of the above-mentioned crop yield loss assessment devices can implement the above-mentioned crop yield loss assessment method one-to-one, which will not be repeated here.

[0097] The present disclosure provides a crop yield loss assessment device, which can accurately calculate the number of ears per unit area and the number of grains per ear of a single ear by acquiring regional ear images and single ear images of the crops to be assessed, respectively, using a trained ear number prediction model and an ear-grain number prediction model to process them. On this basis, the yield prediction value per unit area is further calculated. Then, the yield loss is assessed based on the deviation degree data between the yield prediction value and the preset historical yield value, and an accurate and reliable yield loss assessment result is obtained. It can be seen that the embodiments of the present disclosure provide a scientific loss assessment method, which can effectively reduce the error of human estimation, significantly improve the efficiency and accuracy of loss assessment, and provide scientific, objective and valuable data support for relevant decisions such as claim settlement and loss assessment.

[0098] The specific definition of the crop yield loss assessment device can be found in the definition of the crop yield loss assessment method above, which will not be repeated here. Each module in the above-mentioned crop yield loss assessment device can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0099] Figure 6 FIG. 1 shows a schematic diagram of the structure of an electronic device provided by an embodiment of the present disclosure. Figure 6 As shown, at the hardware level, the electronic device includes a processor, and optionally also includes an internal bus, a network interface, and a memory. Among them, the memory may include a memory, such as a high-speed random access memory (RAM), and may also include a non-volatile memory (non-volatile memory), such as at least one disk storage, etc. Of course, the electronic device may also include hardware required for other services.

[0100] The processor, network interface and memory can be interconnected through an internal bus, which can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 6Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0101] The memory is used to store programs. Specifically, the program may include program code, and the program code includes computer operation instructions. The memory may include internal memory and non-volatile memory, and provides instructions and data to the processor.

[0102] The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs it, forming a crop yield loss assessment device at the logical level. The processor executes the program stored in the memory and is specifically used to execute the above method.

[0103] The processor may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by an integrated logic circuit of hardware in the processor or an instruction in the form of software. The above processor may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The disclosed methods, steps and logic block diagrams in the embodiments of the present disclosure may be implemented or executed. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. The steps of the method disclosed in conjunction with the embodiments of the present disclosure may be directly embodied as being executed by a hardware decoding processor, or may be executed by a combination of hardware and software modules in a decoding processor. The software module may be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.

[0104] The electronic device can execute the crop yield loss assessment method provided by the multiple embodiments of the present disclosure, and realize the crop yield loss assessment device in Figure 5 The functions of the illustrated embodiments will not be described in detail in the embodiments of the present disclosure.

[0105] The embodiments of the present disclosure also propose a computer-readable storage medium, which stores one or more programs, and the one or more programs include instructions. When the instructions are executed by an electronic device including multiple application programs, the electronic device can execute the crop yield loss assessment method provided by multiple embodiments of the present disclosure.

[0106] Those skilled in the art will appreciate that the embodiments of the present disclosure may be provided as methods, systems, or computer program products. Therefore, the present disclosure may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present disclosure may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0107] The present disclosure is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present disclosure. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0108] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0109] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0110] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0111] The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0112] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape, disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0113] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the presence of other identical elements in the process, method, commodity or device including the element.

[0114] It will be appreciated by those skilled in the art that the embodiments of the present disclosure may be provided as methods, systems or computer program products. Therefore, the present disclosure may take the form of a complete hardware embodiment, a complete software embodiment or an embodiment combining software and hardware. Moreover, the present disclosure may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0115] The above are only embodiments of the present disclosure and are not intended to limit the present disclosure. For those skilled in the art, the present disclosure may have various modifications and variations. Any modification, equivalent substitution, improvement, etc. made within the spirit and principle of the present disclosure shall be included in the scope of the claims of the present disclosure.

Claims

1. A method for assessing crop yield loss, characterized in that: The method comprises: Acquire regional ear images and single ear images of crops to be assessed for damage; The trained ear number prediction model is used to process the regional ear image to obtain the total number of ears; and the number of ears per unit area is calculated according to the total number of ears and the shooting area of ​​the regional ear image; Using the trained grain number per ear prediction model, the single ear image is processed to obtain the grain number per ear; Calculating the predicted yield per unit area of ​​the crop to be determined for damage according to the number of ears per unit area and the number of grains per ear; According to the deviation degree data between the predicted yield per unit area and the preset historical yield value, the yield loss of the crops to be assessed is assessed to obtain a yield loss assessment result.

2. The method according to claim 1, characterized in that The type of crop to be determined as damaged is wheat.

3. The method according to claim 1, characterized in that The trained ear number prediction model is generated according to the following method: Acquire a data set, the data set comprising an open source data set and an actual data set; the data set comprises a plurality of fruit ear images of a sample area, the label of each sample is the number of fruit ears, and the average resolution and / or average shooting angle height of the sample images in the actual data set are smaller than the average resolution and / or average shooting angle height of the sample images in the open source data set; The data set is used to train the initial ear number prediction model to obtain the trained ear number prediction model.

4. The method according to any one of claims 1 to 3, characterized in that: The shooting area is calculated according to the following method: Acquire a target region ear image having the same shooting area as the regional ear image, wherein the target region ear image includes a first calibration plate; Obtaining the actual area of ​​the first calibration plate; Calculating the pixel area of ​​the first calibration plate; Calculating a scaling factor according to the actual area and the pixel area; The shooting area is obtained by calculation according to the scale factor and the pixel area of ​​the ear image of the target area.

5. The method according to claim 1, characterized in that The regional ear image is obtained according to the following method: Receive the fruit ear image of the area to be verified and the fruit ear image of the area to be processed sent by the user; the difference between the fruit ear image of the area to be processed and the fruit ear image of the area to be verified is that the fruit ear image of the area to be processed also includes a second calibration plate, and the second calibration plate has a plurality of marks; Using a preset identification extraction algorithm, the ear image of the to-be-processed area is processed to obtain position data of the multiple identifications; Based on the position data of the multiple markers and the preset judgment condition, judging whether the shooting angle of the ear image of the to-be-processed area is horizontal; If the shooting angle of the ear image of the area to be processed is horizontal, the ear image of the area to be verified is used as the regional ear image; If the shooting angle of the ear image of the area to be processed is not horizontal, a prompt indicating that the image does not meet the requirements is generated and displayed, and the process returns to the step of receiving the ear image of the area to be verified and the ear image of the area to be processed sent by the user.

6. The method according to claim 1, characterized in that The single ear of fruit image includes a background for highlighting the ear of fruit in the single ear of fruit image.

7. The method according to claim 1, characterized in that The method of determining the yield loss of the crop to be determined based on the deviation degree data between the predicted yield per unit area and the preset historical yield value to obtain the yield loss determination result includes: Calculating deviation degree data according to the predicted value of yield per unit area and the preset historical yield value; Based on the preset rules corresponding to the degree of deviation and the level of loss, the production loss assessment result is obtained according to the degree of deviation data; wherein the production loss assessment result includes: no disaster, minor disaster, moderate disaster, severe disaster, and total loss.

8. A device for determining crop yield loss, characterized in that: The device comprises: An acquisition module, used for acquiring regional ear images and single ear images of crops to be assessed for damage; The ear number prediction module is used to process the regional ear image using the trained ear number prediction model to obtain the total number of ears; and calculate the number of ears per unit area according to the total number of ears and the shooting area of ​​the regional ear image; The grain number per ear prediction module is used to process the single ear image using the trained grain number per ear prediction model to obtain the grain number per ear; A prediction module, used for calculating the predicted yield per unit area of ​​the crop to be determined for damage according to the number of ears per unit area and the number of grains per ear; The damage assessment module is used to assess the yield damage of the crops to be assessed based on the deviation degree data between the predicted yield per unit area and the preset historical yield value, and obtain the yield damage assessment result.

9. An electronic device, comprising: processor; as well as A memory arranged to store computer executable instructions, characterized in that when the executable instructions are executed, the processor executes the steps of the crop yield loss assessment method as described in any one of claims 1-7.

10. A computer-readable storage medium storing one or more programs, characterized in that: When the one or more programs are executed by an electronic device including a plurality of application programs, the electronic device executes the steps of the crop yield loss assessment method as described in any one of claims 1-7.

Images