A method, device, equipment and medium for evaluating agricultural equipment status

Abstract

The present application relates to the field of artificial intelligence technology, and in particular to a method, device, equipment and medium for evaluating the state of agricultural equipment. The method includes: obtaining multiple images with a time series of a target device, where the target device is any one of multiple agricultural equipment that requires risk control warning; inputting multiple images into a preset equipment state evaluation model to obtain equipment information of the target device, where the equipment information includes equipment state, equipment type and equipment location; determining the expected equipment state of the target device according to the equipment type, equipment location and time series corresponding to the multiple images of the target device, matching the equipment state of the target device with the expected equipment state, and determining whether the equipment state of the target device meets the expected equipment state; if not, sending the warning signal corresponding to the target device to the terminal device of the manager. The present application can timely discover the potential risks of agricultural equipment and reduce the maintenance cost of agricultural equipment.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a method, device, equipment and medium for evaluating the status of agricultural equipment. Background Art

[0002] With the development of agricultural modernization, more and more agricultural equipment is applied to agricultural production. However, some agricultural equipment works outdoors and is easily affected by the natural environment, such as weather factors such as wind, rain, thunder and lightning, as well as factors such as crop diseases and insect pests. These adverse factors will affect the normal operation of agricultural equipment and even cause damage to agricultural equipment and reduced operation quality.

[0003] In related technologies, risk control of agricultural equipment usually relies on manual observation and empirical judgment, and is often not discovered until some time after the agricultural equipment fails or is damaged, which increases equipment maintenance costs and affects the efficiency and quality of agricultural production. Summary of the invention

[0004] In order to solve the problem that the existing technology cannot detect agricultural equipment risks in a timely manner, the present application provides an agricultural equipment status assessment method, device, equipment and medium.

[0005] In the first aspect, the present application provides a method for evaluating the state of agricultural equipment, which adopts the following technical solution:

[0006] A method for evaluating the state of agricultural equipment, comprising:

[0007] Acquire multiple images of a target device in a time series, wherein the target device is any one of multiple agricultural devices that require wind control and early warning;

[0008] Inputting the multiple images into a preset device status assessment model to obtain device information of the target device, wherein the device information includes device status, device type and device location;

[0009] Determining an expected device state of the target device according to a device type and a device location of the target device and a time sequence corresponding to the multiple images;

[0010] Matching a target item of the device state of the target device with an expected state corresponding to the target item, and judging whether the target item satisfies the expected state corresponding to the target item, wherein the device state of the target device includes one or more of a device switch state, a running direction, an opening angle, and an opening size, and the target item is any one of the device states of the target device;

[0011] If the target item does not meet the expected state corresponding to the target item, it is determined that the device state of the target device does not meet the expected device state, and a warning signal corresponding to the target device is sent to the terminal device of the administrator.

[0012] By adopting the above technical solution, multiple images with time series of the target device are input into the preset device status evaluation model, and the device information of the target device can be obtained. The device information includes device status, device type and device location. The device status is used to indicate the switch status and operation status of the target device. According to the device type, device location and the time series corresponding to the multiple images, it can be determined whether the device status under the time series meets the expected device status corresponding to the time series. When any one of the device statuses corresponding to the target device does not meet the expected status, it is determined that the device status of the target device does not meet the expected device status, indicating that the target device may fail. The warning information corresponding to the target device is sent to the terminal device of the manager, so that the manager can be reminded to come and check in time, and the potential risks of agricultural equipment can be discovered in time, thereby reducing the maintenance cost of agricultural equipment.

[0013] In a preferred example, the present application can be further configured as follows: the method further includes:

[0014] Acquire a plurality of historical images corresponding to each of the plurality of agricultural equipment;

[0015] Preprocessing and feature extraction are performed on the multiple historical images to obtain the preprocessed multiple historical images and the corresponding features of the multiple historical images, and corresponding labels are set for the multiple historical images, respectively, and the labels are used to indicate the equipment information of the agricultural equipment in the corresponding historical images;

[0016] Based on the multiple historical images, the features corresponding to each of the multiple historical images, and the labels, model training is performed to obtain the preset equipment status assessment model.

[0017] By adopting the above technical scheme, multiple historical images are preprocessed and feature extracted to obtain features that characterize key information of the images. Corresponding labels indicating equipment information are set for the multiple historical images respectively. Model training is performed based on the multiple historical images and the corresponding features and labels of the multiple historical images. The obtained preset equipment status assessment model can accurately determine the equipment information of agricultural equipment, thereby improving the efficiency of obtaining equipment information of agricultural equipment.

[0018] In a preferred example, the present application may be further configured as follows: based on the multiple historical images, the features and labels corresponding to the multiple historical images, model training is performed to obtain the preset equipment status assessment model, including:

[0019] Based on the multiple historical images, the features and labels corresponding to the multiple historical images, model training is performed to obtain an initial equipment state assessment model and prediction results corresponding to the multiple historical images;

[0020] Generate a confusion matrix according to the prediction results corresponding to each of the multiple historical images;

[0021] Determining the accuracy, precision, recall and F1 value of the initial equipment state assessment model according to the confusion matrix;

[0022] Determining whether the performance of the initial equipment status assessment model meets the preset performance conditions according to the accuracy, precision, recall and F1 value;

[0023] If not, the initial device state assessment model is optimized to obtain the preset device state assessment model.

[0024] By adopting the above technical solution, the confusion matrix is ​​determined, and the performance of the initial equipment state evaluation model is evaluated based on the accuracy, precision, recall rate and F1 value. When the model performance does not meet the preset performance conditions, the performance of the initial equipment state evaluation model is optimized, which can improve the model performance and accuracy.

[0025] In a preferred example, the present application may be further configured to: pre-process and extract features from the plurality of historical images, including:

[0026] Performing histogram equalization, adjusting brightness contrast, and adding Gaussian noise on the multiple historical images in sequence to obtain multiple pre-processed historical images;

[0027] Feature extraction is performed on multiple preprocessed historical images.

[0028] By adopting the above technical solution, multiple historical images are subjected to histogram equalization, brightness contrast adjustment and Gaussian noise addition in sequence, so that the image quality can be improved, so that the features extracted from the pre-processed multiple historical images are more accurate and comprehensive.

[0029] In a preferred example, the present application may be further configured as follows: the step of acquiring a plurality of images of the target device in a time sequence includes:

[0030] Determine whether the target device is in an on state or an off state;

[0031] If the target device is in a closed state, multiple images of the target device in a time series are acquired according to a preset acquisition frequency, and environmental information corresponding to the target device is acquired in real time;

[0032] Determine whether the environment information corresponding to the target device meets the preset activation condition;

[0033] If the conditions are met, the target device is turned on, the preset acquisition frequency is updated, and a plurality of images of the target device in a time series are acquired according to the updated preset acquisition frequency.

[0034] By adopting the above technical solution, different preset acquisition frequencies are used to capture images of the target device when the target device is in the on state and the off state, so that risks of the target device can be discovered in a timely manner, and the amount of calculation of the electronic device is reduced by updating the preset acquisition frequency.

[0035] In a preferred example, the present application may be further configured as follows: the environmental information includes light, wind speed, wind direction and rainfall;

[0036] Determining whether the environment information corresponding to the target device meets the preset activation condition includes:

[0037] Determining the relative position of the target device and the agricultural greenhouse;

[0038] If the relative position is that the target device is set on or outside the agricultural greenhouse, it is determined whether the illumination, wind speed, wind direction and rainfall corresponding to the target device meet the respective corresponding preset ranges;

[0039] If the illumination, wind speed, wind direction and rainfall corresponding to the target device all meet their respective corresponding preset ranges, it is determined that the environmental information corresponding to the target device meets the preset activation condition.

[0040] By adopting the above technical scheme, corresponding environmental information is set for the target device according to the relative positions of the target device and the agricultural greenhouse. When the environmental information corresponding to the target device meets the preset start-up conditions, the target device is turned on, thereby realizing automatic start-up of the target device, reducing labor costs, and improving the management level of the agricultural greenhouse.

[0041] In a second aspect, the present application provides an agricultural equipment status assessment device, which adopts the following technical solution:

[0042] An agricultural equipment status assessment device, comprising:

[0043] An acquisition module, used to acquire multiple images of a target device in a time series, wherein the target device is any one of multiple agricultural devices that require risk control and early warning;

[0044] An input module, used to input the multiple images into a preset device status assessment model to obtain device information of the target device, wherein the device information includes device status, device type and device location;

[0045] A determination module, configured to determine an expected device state of the target device according to a device type and a device location of the target device and a time sequence corresponding to the plurality of images;

[0046] a matching module, used to match a target item of the device state of the target device with an expected state corresponding to the target item, and determine whether the target item satisfies the expected state corresponding to the target item, wherein the device state of the target device includes one or more of a device switch state, a running direction, an opening angle, and an opening size, and the target item is any one of the device states of the target device;

[0047] The sending module is used to determine that the device state of the target device does not meet the expected device state if the target item does not meet the expected state corresponding to the target item, and send the warning signal corresponding to the target device to the terminal device of the administrator.

[0048] In a third aspect, the present application provides an electronic device, which adopts the following technical solution:

[0049] one or more processors;

[0050] Memory;

[0051] At least one application, wherein the at least one application is stored in a memory and configured to be executed by at least one processor, and the at least one application is configured to: execute the agricultural equipment status assessment method as described in any one of the first aspects.

[0052] In a fourth aspect, the present application provides a computer-readable storage medium, which adopts the following technical solution:

[0053] A computer-readable storage medium stores a computer program, which, when executed in a computer, causes the computer to execute the agricultural equipment status assessment method as described in any one of the first aspects.

[0054] In summary, this application includes the following beneficial technical effects:

[0055] The present application can obtain the device information of the target device by inputting multiple images with time series of the target device into a preset device status assessment model, the device information including the device status, device type and device location, the device status being used to indicate the switch status and operation status of the target device, and can determine whether the device status under the time series satisfies the expected device status corresponding to the time series according to the device type, device location and the time series corresponding to the multiple images; when any one of the device statuses corresponding to the target device does not satisfy the expected status, it is determined that the device status of the target device does not meet the expected device status, indicating that the target device may fail, and the warning information corresponding to the target device is sent to the terminal device of the manager, so that the manager can be reminded in time to come and check, so that the potential risks of agricultural equipment can be discovered in time, and the maintenance cost of agricultural equipment can be reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 It is a flowchart of an agricultural equipment status assessment method provided in an embodiment of the present application;

[0057] Figure 2 It is a structural schematic diagram of an agricultural equipment status assessment system provided in an embodiment of the present application;

[0058] Figure 3 It is a structural schematic diagram of an agricultural equipment status assessment device provided in an embodiment of the present application;

[0059] Figure 4 It is a structural schematic diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0060] The following is combined with Figure 1 -Attached Figure 4 This application is described in further detail.

[0061] This specific embodiment is merely an explanation of the present application and is not a limitation of the present application. After reading this specification, those skilled in the art may make modifications to the present embodiment without any creative contribution as needed, but such modifications are protected by the patent law as long as they are within the scope of the claims of the present application.

[0062] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0063] In addition, the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article, unless otherwise specified, generally means that the associated objects before and after are in an "or" relationship.

[0064] The present application embodiment provides a method for evaluating the state of agricultural equipment, such as Figure 1 As shown, the method provided in the embodiment of the present application is performed by an electronic device, which can be a server or a terminal device, wherein the server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. The terminal device can be a smart phone, a tablet computer, a laptop computer, a desktop computer, etc., but is not limited thereto. The terminal device and the server can be directly or indirectly connected via wired or wireless communication, which is not limited in the embodiment of the present application. The method includes steps S101 to S105, wherein:

[0065] S101. Acquire multiple images of a target device in a time series, where the target device is any one of multiple agricultural devices that require risk control and early warning.

[0066] In this embodiment, a camera device can be set for each agricultural equipment according to the respective positions of the multiple agricultural equipment to collect multiple images of the agricultural equipment in a time series. The agricultural equipment status assessment method provided in this embodiment can be applicable to greenhouse planting scenes. The agricultural equipment may include equipment set inside the greenhouse and equipment set outside the greenhouse. Among them, the equipment set inside the greenhouse may include indoor sprinkler irrigation devices and indoor fans; the equipment set outside the greenhouse may include equipment set on the greenhouse and equipment set outside the greenhouse, such as sunshades and ventilation windows.

[0067] S102: Input multiple images into a preset device status assessment model to obtain device information of the target device, where the device information includes device status, device type, and device location.

[0068] In this embodiment, the preset device state evaluation model is pre-trained and used to evaluate the device information of the device. Multiple images are input into the preset device state evaluation model, and the device type and device position of the device contained in the multiple images can be automatically identified. The preset correspondence between the device type and the device state can also be preset. After multiple images of the target device are input into the preset device state evaluation model, the device state corresponding to the target device can be determined according to the preset correspondence, wherein the device state of the target device can be divided into an instantaneous state and a sequential state. The instantaneous state is a state that can be determined based on one image, including the switch state, the opening size, and the opening angle. The sequential state is a state that needs to be determined based on multiple images, including the running direction. For the instantaneous state of the target device, the instantaneous state corresponding to the image with the latest acquisition time among the multiple images can be used as the device state of the target device.

[0069] Specifically, the preset correspondence between the device type and the device information can be set according to actual needs, and this embodiment does not make any specific limitation. Optionally, for indoor fans, the corresponding device status can be set to a switch status; for indoor sprinkler irrigation devices, the corresponding device status can be set to a switch status, an operating direction (the direction of change in the angle of the water spray of the indoor sprinkler irrigation device); for sunshades, the corresponding device status can be set to a switch status (device on / device off), an operating direction (the opening and closing direction of the sunshade), and an opening size (expressed as a percentage); for ventilation windows, the corresponding device status can be set to an opening angle.

[0070] S103: Determine an expected device state of the target device according to the device type and device location of the target device and a time sequence corresponding to the multiple images.

[0071] In this embodiment, a variety of crops are planted in the greenhouse, and the crops have different requirements for agricultural equipment. For example, some crops need more irrigation, and some crops need strong light. Therefore, the requirements of the crops corresponding to the equipment can be determined according to the equipment type and equipment location, and then the expected equipment state can be set. The expected equipment state can be set by time period. For example, for a certain sunshade curtain, the crops corresponding to the sunshade curtain do not need strong light. In the time period of strong light at noon, such as 10 am to 2 pm, the opening size of the sunshade curtain can be set to 100%, and in the time period of weak light in the morning, such as 7 am to 10 am, the opening size of the sunshade curtain can be set to 30%.

[0072] Furthermore, if the device type of the target device is an indoor fan, the latest moment in the time sequence corresponding to the multiple images is used as the acquisition moment corresponding to the device state of the target device, and the expected device state of the target device can be determined based on the acquisition moment and the device type and device location of the target device.

[0073] S104. Match the target item of the device state of the target device with the expected state corresponding to the target item to determine whether the target item satisfies the expected state corresponding to the target item. The device state of the target device includes one or more of the device switch state, running direction, opening angle, and opening size. The target item is any one of the device states of the target device.

[0074] In this embodiment, if the expected device state of the target device is the on state, and the device state of the target device is also the on state, it indicates that the device state of the target device meets the expected device state; if the expected device state of the target device is the on state, and the device state of the target device is the off state, it indicates that the device state of the target device does not meet the expected device state. When the device state of the target device does not meet the expected device state, the warning information corresponding to the target device can be sent to the terminal device of the administrator, and the warning signal can be in the form of a WEB site notification, an APP push notification, a WeChat public account notification, or a text message notification.

[0075] S105. If the target item does not meet the expected state corresponding to the target item, it is determined that the device state of the target device does not meet the expected device state, and a warning signal corresponding to the target device is sent to the terminal device of the administrator.

[0076] The warning information of the target device may include the status information and expected device status of the target device, as well as the device type and device location of the target device, so that the management personnel can quickly locate the location and abnormal situation of the target device according to the warning signal.

[0077] The embodiment of the present application can obtain device information of the target device by inputting multiple images with time series of the target device into a preset device status assessment model, where the device information includes device status, device type and device location, and the device status is used to indicate the switch status and operation status of the target device. According to the device type, device location and the time series corresponding to the multiple images, it can be determined whether the device status under the time series satisfies the expected device status corresponding to the time series. When any one of the device statuses corresponding to the target device does not satisfy the expected status, it is determined that the device status of the target device does not meet the expected device status, indicating that the target device may fail. The warning information corresponding to the target device is sent to the terminal device of the manager, so that the manager can be reminded in time to come and check, so that the potential risks of agricultural equipment can be discovered in time, and the maintenance cost of agricultural equipment can be reduced.

[0078] In a possible implementation manner of the embodiment of the present application, the method further includes:

[0079] Acquire a plurality of historical images corresponding to a plurality of agricultural equipment;

[0080] Preprocessing and feature extraction are performed on the multiple historical images to obtain the preprocessed multiple historical images and the corresponding features of the multiple historical images, and corresponding labels are set for the multiple historical images, respectively, where the labels are used to indicate equipment information of the agricultural equipment in the corresponding historical images;

[0081] Based on multiple historical images, the features corresponding to each of the multiple historical images, and labels, model training is performed to obtain a preset equipment status assessment model.

[0082] In this embodiment, the process of obtaining the preset equipment status assessment model may include: obtaining multiple historical images corresponding to multiple agricultural equipment, and the multiple historical images cover different time points, different equipment types, and different equipment states. The step of preprocessing the multiple historical images may include histogram equalization, adjusting brightness contrast, and adding Gaussian noise to improve image quality and perform subsequent feature extraction. Specifically, a deep learning model (such as a convolutional neural network) can be used to extract features from the preprocessed multiple historical images. The convolutional neural network can extract features from low to high levels (such as edges, textures, shapes, and object parts, etc.) from the image. These features can be used to identify the equipment type and equipment status of the agricultural equipment in the image. A corresponding label can be set for each historical image, and the label includes an equipment type label and an equipment status label, wherein the equipment status label can be determined according to the equipment type of the agricultural equipment in the historical image. For historical images containing indoor fans, the set equipment status label may include a switch status, for historical images containing indoor sprinkler irrigation devices, the set equipment status label may include a switch status and a running direction, for historical images containing sunshades, the set equipment status label may include a switch status and a running direction, and for historical images containing ventilation windows, the set equipment status label may include an opening angle.

[0083] Based on the extracted features and the labels corresponding to the historical images, a supervised learning algorithm (such as YOLOv8 model, support vector machine, logistic regression, decision tree) is used to train a classifier model, i.e., a preset equipment status assessment model, which can determine the equipment type, equipment status, and equipment location of the corresponding agricultural equipment based on the input image.

[0084] The embodiment of the present application can obtain features characterizing key information of the images by preprocessing and extracting features from multiple historical images, and respectively set corresponding labels indicating equipment information for the multiple historical images. Model training is performed based on the multiple historical images and the features and labels corresponding to each of the multiple historical images. The obtained preset equipment status assessment model can accurately determine the equipment information of the agricultural equipment, thereby improving the efficiency of obtaining the equipment information of the agricultural equipment.

[0085] A possible implementation of the embodiment of the present application is to perform model training based on multiple historical images, features corresponding to each of the multiple historical images, and labels to obtain a preset equipment status assessment model, including:

[0086] Based on the multiple historical images, the features and labels corresponding to the multiple historical images, model training is performed to obtain an initial equipment status assessment model and prediction results corresponding to the multiple historical images;

[0087] Generate a confusion matrix based on the prediction results corresponding to multiple historical images;

[0088] According to the confusion matrix, determine the accuracy, precision, and recall of the initial equipment status assessment model;

[0089] Determine whether the performance of the initial equipment status assessment model meets the preset performance conditions based on the accuracy, precision, and recall rate;

[0090] If not, the initial equipment status assessment model is optimized to obtain a preset equipment status assessment model.

[0091] In this embodiment, a model trained based on multiple historical images, features and labels corresponding to each of the multiple historical images can be used as an initial device state assessment model. Specifically, a positive sample image and a negative sample image can be set. The positive sample image contains multiple historical images, features and labels corresponding to each of the multiple historical images. The labels include device type labels, device status labels and positive example labels. The positive example labels indicate that both the device category labels and the device status labels are correct. The negative sample image and the multiple historical images contained in the positive sample image can be the same or different. The negative sample image contains multiple historical images, features and labels corresponding to each of the multiple historical images. The labels include device type labels, device labels and negative example labels. The negative example labels indicate that both the device category labels and the device status labels are wrong. A model trained based on features and labels corresponding to each of the positive sample image and the negative sample image is used as an initial device state assessment model. The initial device state assessment model can be used to determine device information in an image according to an input image. At the same time, during the training process of the initial device state assessment model, the initial device state assessment model can output a prediction result for each historical image. The prediction result includes whether the historical image is a positive example or a negative example.

[0092] Furthermore, according to the prediction results of each historical image, a confusion matrix can be obtained. The confusion matrix consists of four elements: true positive examples TP, false negative examples FN, false positive examples FP and true negative examples TN, where true positive examples are the number of times the model predicts a positive sample image as a positive example, false negative examples are the number of times the model predicts a positive sample image as a negative example, false positive examples are the number of times the model predicts a negative sample image as a positive example, and true negative examples are the number of times the model predicts a negative sample image as a positive example. By calculating the four elements of the confusion matrix, various evaluation indicators can be obtained, including precision, recall rate, accuracy and F1 value. Accuracy: indicates the ratio of the number of sample images correctly predicted by the model to the total number of sample images, that is, Accuracy=(TP+TN) / (TP+FP+FN+TN); Recall: indicates the proportion of positive sample images correctly predicted by the model, that is, Recall=TP / (TP+FN); Precision: indicates the proportion of true positive samples in the sample images predicted by the model as positive samples, that is, Precision=TP / (TP+FP); F1 value: indicates the harmonic mean of accuracy and recall. The initial equipment state assessment model can also be cross-validated to obtain the cross-validation error to evaluate the performance estimate of the initial equipment state assessment model on the entire training data set.

[0093] In addition, preset ranges and preset quantity thresholds corresponding to performance indicators can also be pre-set, and performance indicators include precision, recall rate, accuracy, F1 value, and cross-validation error. When the number of performance indicators that do not meet the corresponding preset range exceeds the preset quantity threshold, the performance of the initial equipment state evaluation model does not meet the preset performance conditions. Further, the initial equipment state evaluation model can be optimized, and the optimization method includes model parameter adjustment, model data enhancement, model integration, etc., so as to obtain an optimized preset equipment state evaluation model.

[0094] The embodiment of the present application determines the confusion matrix and performs performance evaluation on the initial equipment state evaluation model according to the accuracy, precision, recall rate and F1 value. When the model performance does not meet the preset performance conditions, the performance of the initial equipment state evaluation model is optimized, thereby improving the model performance and accuracy.

[0095] A possible implementation of the embodiment of the present application is to preprocess and extract features from multiple historical images, including:

[0096] Performing histogram equalization, adjusting brightness contrast, and adding Gaussian noise on multiple historical images in turn to obtain multiple preprocessed historical images;

[0097] Feature extraction is performed on multiple preprocessed historical images.

[0098] In this embodiment, the discrete form of histogram equalization can be expressed by the following formula:

[0099]

[0100] in, The output image is The pixel value of gray level, is a transformation function used to convert the grayscale of the input image Mapped to the grayscale of the output image , is the number of gray levels, usually 256, that is, an integer from 0 to 255, The gray level in the input image is The probability of a pixel appearing is Indicates that the gray level in the input image is The number of pixels, Represents the total number of pixels of the input image. This formula converts the grayscale histogram of the output historical image into a histogram with an approximately uniform distribution, thereby achieving image contrast enhancement.

[0101] The formula for adjusting brightness and contrast can be expressed as:

[0102]

[0103] in, represents the contrast factor, A value greater than 1 increases the contrast. Less than 1 means reducing the contrast. Represents the original pixel value, represents the pixel value after contrast adjustment, Represents the brightness adjustment value. The brightness adjustment value can be positive or negative. A positive brightness adjustment value indicates an increase in brightness, while a negative brightness adjustment value indicates a decrease in brightness.

[0104] The probability density function of adding Gaussian noise can be expressed as:

[0105]

[0106] in, is the standard deviation, is a random variable, is the mean.

[0107] The embodiment of the present application can improve the image quality by performing histogram equalization, adjusting brightness contrast and adding Gaussian noise on multiple historical images in sequence, so that the features extracted from the pre-processed multiple historical images are more accurate and comprehensive.

[0108] A possible implementation of the embodiment of the present application is to obtain multiple images of a target device in a time sequence, including:

[0109] Determine whether the target device is turned on or off;

[0110] If the target device is in a closed state, multiple images of the target device in a time series are acquired according to a preset acquisition frequency, and the environmental information corresponding to the target device is acquired in real time;

[0111] Determine whether the environment information corresponding to the target device meets the preset activation conditions;

[0112] If the conditions are met, the target device is turned on, the preset acquisition frequency is updated, and a plurality of images of the target device in a time series are acquired according to the updated preset acquisition frequency.

[0113] In this embodiment, at the initial moment, an image of the target device can be obtained, and through the preset device evaluation model, it can be determined whether the target device is in the on state or the off state. The preset acquisition frequencies can be set in advance for the target devices in the on state and the off state respectively. The preset acquisition frequency of the target device in the off state can be used as the first preset acquisition frequency, and the preset acquisition frequency of the target device in the on state can be used as the second preset acquisition frequency, and the second preset acquisition frequency is greater than the first preset acquisition frequency.

[0114] In one possible situation, if the target device is in a closed state, the image acquisition device is controlled to acquire multiple images of the target device with a time series according to a first preset acquisition frequency. The corresponding relationship between the device type and the environmental information can be preset. According to the relative positions of the agricultural equipment and the agricultural greenhouse, the agricultural equipment can be divided into two types. One is that the relative position of the agricultural equipment and the agricultural greenhouse is that the agricultural equipment is set inside the agricultural greenhouse, and the other is that the relative position of the agricultural equipment and the agricultural greenhouse is that the agricultural equipment is set on or outside the agricultural greenhouse. For the type of equipment whose relative position is that the agricultural equipment is set inside the agricultural greenhouse, such as an indoor fan and an indoor sprinkler, the corresponding environmental information may include temperature, humidity, wind speed, and light. For the type of equipment whose relative position is that the agricultural equipment is set on or outside the agricultural greenhouse, such as a sunshade and a ventilation window, the corresponding environmental information may include light, wind speed, wind direction, and rainfall. According to actual needs, corresponding environmental information can also be set for each device type, which is not specifically limited in this embodiment. For the environmental information corresponding to any device type, a first preset range can be set in advance for each item in the environmental information, and the first preset range is the range corresponding to the opening condition. Furthermore, after obtaining the environmental information corresponding to the target device, it is determined whether each item of the environmental information corresponding to the target device meets the corresponding first preset range. If each item of the environmental information corresponding to the target device meets the corresponding first preset range, it is determined that the environmental information corresponding to the target device meets the preset activation condition. If any item of the environmental information corresponding to the target device does not meet the corresponding first preset range, it is determined that the environmental information corresponding to the target device does not meet the preset activation condition.

[0115] Furthermore, if the environmental information corresponding to the target device meets the preset start-up conditions, the target device is turned on and the preset acquisition frequency is updated. The first preset acquisition frequency before the update is updated to the second preset acquisition frequency, and multiple images with a time series of the target device are acquired according to the updated second preset acquisition frequency; if the environmental information corresponding to the target device does not meet the preset start-up conditions, multiple images with a time series of the target device continue to be acquired according to the first preset acquisition frequency.

[0116] In another possible situation, when the target device is in the on state, the image acquisition device is controlled to acquire multiple images of the target device with a time series according to a second preset acquisition frequency, and the environmental information corresponding to the target device is acquired in real time. For the environmental information corresponding to any device type, a second preset range can be set in advance for each item in the environmental information, and the second preset range is the range corresponding to the shutdown condition. Further, after acquiring the environmental information corresponding to the target device, it is determined whether each item in the acquired environmental information corresponding to the target device satisfies the corresponding second preset range. If each item in the acquired environmental information corresponding to the target device satisfies the corresponding second preset range, it is determined that the environmental information corresponding to the target device satisfies the preset shutdown condition. If any item in the acquired environmental information corresponding to the target device does not satisfy the corresponding second preset range, it is determined that the environmental information corresponding to the target device does not satisfy the preset shutdown condition.

[0117] Furthermore, if the environmental information corresponding to the target device meets the preset shutdown condition, the target device is shut down, and the preset acquisition frequency is updated, the second preset acquisition frequency before the update is updated to the first preset acquisition frequency, and multiple images with a time series of the target device are acquired according to the updated first preset acquisition frequency; if the environmental information corresponding to the target device does not meet the preset shutdown condition, multiple images with a time series of the target device continue to be acquired according to the second preset acquisition frequency.

[0118] The embodiment of the present application uses different preset acquisition frequencies to capture images of the target device when the target device is in an on state and a off state, so that risks of the target device can be discovered in a timely manner, while reducing the amount of computation of the electronic device by updating the preset acquisition frequency.

[0119] In a possible implementation of the embodiment of the present application, the environmental information includes light, wind speed, wind direction and rainfall;

[0120] Determine whether the environment information corresponding to the target device meets the preset activation conditions, including:

[0121] Determine the relative position of the target equipment and the agricultural greenhouse;

[0122] If the relative position is that the target device is set on or outside the agricultural greenhouse, it is determined whether the light, wind speed, wind direction and rainfall corresponding to the target device meet the respective corresponding preset ranges;

[0123] If the illumination, wind speed, wind direction and rainfall corresponding to the target device all meet their respective preset ranges, it is determined that the environmental information corresponding to the target device meets the preset activation condition.

[0124] In this embodiment, there are three ways to control agricultural equipment. The first way is manual control, including turning on the equipment, turning off the equipment and controlling the running direction of the equipment; the second way is network single-point control, where humans input control instructions through the WEB or APP, and the electronic device controls the agricultural equipment after receiving the manually input control instructions; the third way is timing condition control, that is, after the environmental information corresponding to the target device meets the preset start-up conditions, the target device is controlled to turn on.

[0125] The embodiment of the present application sets corresponding environmental information for the target device according to the relative positions of the target device and the agricultural greenhouse. When the environmental information corresponding to the target device meets the preset start-up conditions, the target device is turned on, thereby realizing automatic start-up of the target device, reducing labor costs, and improving the management level of the agricultural greenhouse.

[0126] Figure 2 is a structural diagram of an agricultural equipment status assessment system provided in an embodiment of the present application, such as Figure 2 As shown, equipment control can be achieved through manual control, network single-point control and timing condition control. The monitoring module can collect images of agricultural equipment through image acquisition equipment, including immediate capture, timed capture and timed video streaming. The acquired images are input into the preset equipment status evaluation model to obtain the evaluation results of the equipment status. The evaluation results are compared with the expected results. If the comparison fails, an early warning module can be used to issue an early warning signal to the terminal device of the manager. At the same time, the data in the above process are archived and recorded and saved in the database.

[0127] The above embodiment introduces an agricultural equipment status assessment method from the perspective of method flow, and the following embodiment introduces an agricultural equipment status assessment device from the perspective of a virtual module or a virtual unit. Please refer to the following embodiment for details.

[0128] The present application embodiment provides an agricultural equipment status assessment device, such as Figure 3 As shown, the device may include:

[0129] An acquisition module 301 is used to acquire multiple images of a target device in a time series, where the target device is any one of multiple agricultural devices that require wind control and early warning;

[0130] An input module 302 is used to input multiple images into a preset device status assessment model to obtain device information of a target device, where the device information includes device status, device type, and device location;

[0131] A determination module 303, configured to determine an expected device state of the target device according to a device type and a device location of the target device and a time sequence corresponding to a plurality of images;

[0132] A matching module 304 is used to match a target item of a device state of a target device with an expected state corresponding to the target item, and determine whether the target item satisfies the expected state corresponding to the target item. The device state of the target device includes one or more of a device switch state, a running direction, an opening angle, and an opening size. The target item is any one of the device states of the target device.

[0133] The sending module 305 is used to determine that the device state of the target device does not meet the expected device state if the target item does not meet the expected state corresponding to the target item, and send a warning signal corresponding to the target device to the terminal device of the administrator.

[0134] In a preferred example, the present application can be further configured as follows: the device also includes a training module, specifically used for:

[0135] Acquire a plurality of historical images corresponding to a plurality of agricultural equipment;

[0136] Preprocessing and feature extraction are performed on the multiple historical images to obtain the preprocessed multiple historical images and the corresponding features of the multiple historical images, and corresponding labels are set for the multiple historical images, respectively, where the labels are used to indicate equipment information of the agricultural equipment in the corresponding historical images;

[0137] Based on multiple historical images, the features corresponding to each of the multiple historical images, and labels, model training is performed to obtain a preset equipment status assessment model.

[0138] In a preferred example, the present application can be further configured as follows: when the training module performs model training based on multiple historical images, features corresponding to each of the multiple historical images, and labels to obtain a preset device status assessment model, it is specifically used to:

[0139] Based on the multiple historical images, the features and labels corresponding to the multiple historical images, model training is performed to obtain an initial equipment status assessment model and prediction results corresponding to the multiple historical images;

[0140] Generate a confusion matrix based on the prediction results corresponding to multiple historical images;

[0141] According to the confusion matrix, determine the accuracy, precision, recall and F1 value of the initial equipment status assessment model;

[0142] Determine whether the performance of the initial equipment status assessment model meets the preset performance conditions based on the accuracy, precision, recall and F1 value;

[0143] If not, the initial equipment status assessment model is optimized to obtain a preset equipment status assessment model.

[0144] In a preferred example, the present application can be further configured as follows: when the training module performs preprocessing and feature extraction on a plurality of historical images, it is specifically used to:

[0145] Performing histogram equalization, adjusting brightness contrast, and adding Gaussian noise on multiple historical images in turn to obtain multiple preprocessed historical images;

[0146] Feature extraction is performed on multiple preprocessed historical images.

[0147] In a preferred example, the present application may be further configured as follows: when the acquisition module 301 acquires multiple images of the target device in a time sequence, it is specifically used to:

[0148] Determine whether the target device is turned on or off;

[0149] If the target device is in a closed state, multiple images of the target device in a time series are acquired according to a preset acquisition frequency, and the environmental information corresponding to the target device is acquired in real time;

[0150] Determine whether the environment information corresponding to the target device meets the preset activation conditions;

[0151] If the conditions are met, the target device is turned on, the preset acquisition frequency is updated, and a plurality of images of the target device in a time series are acquired according to the updated preset acquisition frequency.

[0152] In a preferred example, the present application can be further configured as follows: when the acquisition module 301 determines whether the environment information corresponding to the target device meets the preset activation condition, it is specifically used to:

[0153] Determine the relative position of the target equipment and the agricultural greenhouse;

[0154] If the relative position is that the target device is set on or outside the agricultural greenhouse, it is determined whether the light, wind speed, wind direction and rainfall corresponding to the target device meet the respective corresponding preset ranges;

[0155] If the illumination, wind speed, wind direction and rainfall corresponding to the target device all meet their respective preset ranges, it is determined that the environmental information corresponding to the target device meets the preset activation condition.

[0156] An agricultural equipment status assessment device provided in an embodiment of the present application is applicable to the above method embodiment and will not be described in detail here.

[0157] An electronic device is provided in an embodiment of the present application, such as Figure 4 As shown, Figure 4The electronic device 400 shown includes: a processor 401 and a memory 403. The processor 401 and the memory 403 are connected, such as through a bus 402. Optionally, the electronic device 400 may also include a transceiver 404. It should be noted that in actual applications, the transceiver 404 is not limited to one, and the structure of the electronic device 400 does not constitute a limitation on the embodiments of the present application.

[0158] Processor 401 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. It may implement or execute various exemplary logic blocks, modules and circuits described in conjunction with the disclosure of this application. Processor 401 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0159] The bus 402 may include a path to transmit information between the above components. The bus 402 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus 402 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 Only one thick line is used in the diagram, but it does not mean that there is only one bus or only one type of bus.

[0160] The memory 403 may be a ROM (Read Only Memory) or other types of static storage devices that can store static information and instructions, a RAM (Random Access Memory) or other types of dynamic storage devices that can store information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, optical disk storage (including compressed optical disk, laser disk, optical disk, digital versatile disk, Blu-ray disk, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto.

[0161] The memory 403 is used to store the application code for executing the solution of the present application, and the execution is controlled by the processor 401. The processor 401 is used to execute the application code stored in the memory 403 to implement the contents shown in the above-mentioned embodiment of the method for evaluating the state of agricultural equipment.

[0162] Figure 4 The electronic device shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0163] An embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer-readable storage medium is run on a computer, the computer can execute the corresponding content in the aforementioned method embodiment.

[0164] It should be understood that, although the steps in the flowchart of the accompanying drawings are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

[0165] The above are only some implementation methods of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A method for evaluating the state of agricultural equipment, characterized in that: include: Acquire multiple images of a target device in a time series, wherein the target device is any one of multiple agricultural devices that require wind control and early warning; Inputting the multiple images into a preset device status assessment model to obtain device information of the target device, wherein the device information includes device status, device type and device location; Determining an expected device state of the target device according to a device type and a device location of the target device and a time sequence corresponding to the multiple images; Matching a target item of the device state of the target device with an expected state corresponding to the target item, and judging whether the target item satisfies the expected state corresponding to the target item, wherein the device state of the target device includes one or more of a device switch state, a running direction, an opening angle, and an opening size, and the target item is any one of the device states of the target device; If the target item does not meet the expected state corresponding to the target item, it is determined that the device state of the target device does not meet the expected device state, and a warning signal corresponding to the target device is sent to the terminal device of the administrator; The step of acquiring a plurality of images of a target device in a time sequence comprises: Determine whether the target device is in an on state or an off state; If the target device is in a closed state, multiple images of the target device in a time series are acquired according to a preset acquisition frequency, and environmental information corresponding to the target device is acquired in real time; Determine whether the environment information corresponding to the target device meets the preset activation condition; If the conditions are met, the target device is turned on, the preset acquisition frequency is updated, and a plurality of images of the target device in a time series are acquired according to the updated preset acquisition frequency; Wherein, updating the preset acquisition frequency includes: determining the preset acquisition frequency of the target device in the closed state as the first preset acquisition frequency, updating the first preset acquisition frequency to a second preset acquisition frequency, and the second preset acquisition frequency is greater than the first preset acquisition frequency.

2. The agricultural equipment status assessment method according to claim 1, characterized in that: The method further comprises: Acquire a plurality of historical images corresponding to each of the plurality of agricultural equipment; Preprocessing and feature extraction are performed on the multiple historical images to obtain the preprocessed multiple historical images and the corresponding features of the multiple historical images, and corresponding labels are set for the multiple historical images, respectively, and the labels are used to indicate the equipment information of the agricultural equipment in the corresponding historical images; Based on the multiple historical images, the features corresponding to each of the multiple historical images, and the labels, model training is performed to obtain the preset equipment status assessment model.

3. The agricultural equipment status assessment method according to claim 2, characterized in that: Based on the multiple historical images, the features and labels corresponding to the multiple historical images, model training is performed to obtain the preset equipment status assessment model, including: Based on the multiple historical images, the features and labels corresponding to the multiple historical images, model training is performed to obtain an initial equipment state assessment model and prediction results corresponding to the multiple historical images; Generate a confusion matrix according to the prediction results corresponding to each of the multiple historical images; Determining the accuracy, precision, recall and F1 value of the initial equipment state assessment model according to the confusion matrix; Determining whether the performance of the initial equipment status assessment model meets preset performance conditions according to the accuracy, precision, recall and F1 value; If not, the initial device state assessment model is optimized to obtain the preset device state assessment model.

4. The agricultural equipment status assessment method according to claim 2, characterized in that: Preprocessing and feature extraction are performed on the multiple historical images, including: Performing histogram equalization, adjusting brightness contrast, and adding Gaussian noise on the multiple historical images in sequence to obtain multiple pre-processed historical images; Feature extraction is performed on multiple preprocessed historical images.

5. The agricultural equipment status assessment method according to claim 1, characterized in that: The environmental information includes light, wind speed, wind direction and rainfall; Determining whether the environment information corresponding to the target device meets the preset activation condition includes: Determining the relative position of the target device and the agricultural greenhouse; If the relative position is that the target device is set on or outside the agricultural greenhouse, it is determined whether the illumination, wind speed, wind direction and rainfall corresponding to the target device meet the respective corresponding preset ranges; If the illumination, wind speed, wind direction and rainfall corresponding to the target device all meet their respective corresponding preset ranges, it is determined that the environmental information corresponding to the target device meets the preset activation condition.

6. An agricultural equipment status assessment device, characterized in that: include: An acquisition module, used to acquire multiple images of a target device in a time series, wherein the target device is any one of multiple agricultural devices that require risk control and early warning; An input module, used for inputting the plurality of images into a preset device status assessment model to obtain device information of the target device, wherein the device information includes device status, device type and device location; A determination module, configured to determine an expected device state of the target device according to a device type and a device location of the target device and a time sequence corresponding to the plurality of images; a matching module, used to match a target item of the device state of the target device with an expected state corresponding to the target item, and determine whether the target item satisfies the expected state corresponding to the target item, wherein the device state of the target device includes one or more of a device switch state, a running direction, an opening angle, and an opening size, and the target item is any one of the device states of the target device; A sending module, configured to determine that the device state of the target device does not meet the expected device state if the target item does not meet the expected state corresponding to the target item, and send a warning signal corresponding to the target device to a terminal device of a manager; The acquisition module, when executing the acquisition of a plurality of images having a time sequence of the target device, is specifically used to: Determine whether the target device is in an on state or an off state; If the target device is in a closed state, multiple images of the target device in a time series are acquired according to a preset acquisition frequency, and environmental information corresponding to the target device is acquired in real time; Determine whether the environment information corresponding to the target device meets the preset activation condition; If the conditions are met, the target device is turned on, the preset acquisition frequency is updated, and a plurality of images of the target device in a time series are acquired according to the updated preset acquisition frequency; Among them, when the acquisition module executes the updating of the preset acquisition frequency, it is specifically used to: determine the preset acquisition frequency of the target device in the closed state as the first preset acquisition frequency, update the first preset acquisition frequency to the second preset acquisition frequency, and the second preset acquisition frequency is greater than the first preset acquisition frequency.

7. An electronic device, characterized in that: include: at least one processor; Memory; At least one application, wherein the at least one application is stored in a memory and configured to be executed by at least one processor, and the at least one application is configured to: execute the agricultural equipment status assessment method according to any one of claims 1-5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed in a computer, the computer is caused to execute the agricultural equipment status assessment method according to any one of claims 1 to 5.

Images