Intelligent monitoring and identifying method for birds, monitoring system, medium and product

Abstract

The invention relates to the technical field of intelligent monitoring, in particular to a bird intelligent monitoring identification method, a monitoring system, a medium and a product, and the method comprises the steps: obtaining a target image which is an image corresponding to a to-be-monitored area; importing the target image into a trained target recognition model to obtain corresponding bird recognition information; obtaining all bird identification information corresponding to a preset time period, and integrating all bird identification information in the preset time period to obtain a bird identification data set; and obtaining a standard data set corresponding to a preset time period, matching the standard data set with the bird identification data set to obtain a data set matching value, and when the data set matching value is lower than a preset standard value, generating prompt feedback information based on the bird identification data set. According to the invention, the accuracy of monitoring and identifying birds can be improved.

Description

Technical Field

[0001] The present application relates to the field of intelligent monitoring technology, and in particular to a method, monitoring system, medium and product for intelligent bird monitoring and identification. Background Art

[0002] Birds are an important part of biodiversity. They have a key identity and role in the global ecosystem. For example, birds can spread seeds and control pests. By monitoring birds, we can analyze their species, numbers, and distribution changes, which is convenient for assessing the status of biodiversity, and can also timely discover and protect endangered species and prevent the loss of biodiversity. In addition, since birds are more sensitive to environmental changes, the population dynamics and distribution patterns of birds often reflect the health of the current environment. Therefore, monitoring birds is particularly critical.

[0003] In the related art, manual feature extraction is generally used to identify features such as the beak, feet, wings or neck of birds in the target image, and then the details such as the contour, color and texture corresponding to the features are extracted to complete the identification of the birds in the target image. However, when using this method to identify birds, high requirements are placed on the clarity and quality of the target image. That is, any slight noise in the target image, changes in lighting conditions or differences in shooting angles may affect the accuracy of feature extraction, thereby affecting the reliability of subsequent identification. Summary of the invention

[0004] In order to improve the accuracy of bird monitoring and identification, the present application provides a bird intelligent monitoring and identification method, monitoring system, medium and product.

[0005] In the first aspect, the present application provides a method for intelligent monitoring and identification of birds, which adopts the following technical solutions: A bird intelligent monitoring and identification method, comprising: Acquire a target image, where the target image is an image corresponding to the area to be monitored; Importing the target image into a trained target recognition model to obtain corresponding bird recognition information; Obtain all bird identification information corresponding to a preset time period, and integrate all bird identification information within the preset time period to obtain a bird identification data set; A standard data set corresponding to the preset time period is obtained, and the standard data set is matched with the bird identification data set to obtain a data set matching value. When the data set matching value is lower than a preset standard value, prompt feedback information is generated based on the bird identification data set.

[0006] By adopting the above technical solution, by importing the target image into the trained target recognition model, it is convenient to quickly and accurately identify the bird identification information contained in the target image. When the bird identification information is determined by the trained target recognition model, the noise in the target image, the change of lighting conditions or the shooting angle has little effect on the recognition result, and it is also convenient to reduce the impact of human misjudgment or prejudice on the accuracy of the recognition result. Finally, by matching the standard data set with the bird identification data set, it is convenient to timely monitor the possible abnormal situations in the monitored area, and by generating prompt feedback information, it is convenient to remind relevant staff to pay attention to and deal with the abnormalities in time.

[0007] In a possible implementation, the training process of the target recognition model includes: Obtaining an initial recognition model and training sample data, wherein the initial recognition model includes an initial teacher recognition model and an initial student recognition model, and the training sample data includes unlabeled sample images and labeled sample images; Execute the following loop steps until the loss output value corresponding to the initial recognition model is lower than the preset threshold, then stop the loop, and determine the initial recognition model with the loss output value lower than the preset threshold as the target recognition model; The cycle steps include: Importing the unlabeled sample image into the initial teacher recognition model to obtain a pseudo label corresponding to the unlabeled sample image; The initial student recognition model is trained according to the unlabeled sample images and the corresponding pseudo labels as well as the labeled sample images to obtain an updated initial student recognition model; Optimizing the updated initial student identification model based on a preset exponential moving average algorithm to obtain an updated teacher identification model; The loss output value of the initial recognition model is determined based on a preset loss function.

[0008] By adopting the above technical solution, the initial recognition model is trained jointly by unlabeled sample images and labeled sample images, which makes it easy to reduce the demand for training sample data in the model training process, thereby improving the generalization ability of the initial recognition model. In addition, through the loop steps, the student model is continuously trained with pseudo labels, and the initial teacher recognition model is updated with the optimized initial student recognition model. This iterative training method makes it easy for the initial teacher recognition model and the initial student recognition model to promote each other and make progress together, thereby continuously improving the recognition performance of the initial recognition model, that is, to ensure that the final target recognition model has sufficient recognition performance.

[0009] In a possible implementation, the updating initial student identification model is optimized based on a preset exponential moving average algorithm to obtain an updated teacher identification model, including: Obtaining student model parameters corresponding to the updated initial student model and teacher model parameters corresponding to the updated teacher identification model; Obtaining a preset update suppression parameter, and importing the update suppression parameter, the student model parameter, and the teacher model parameter into a preset parameter optimization formula to obtain an optimized model parameter; The teacher model parameters are updated based on the optimized model parameters to update the teacher identification model.

[0010] By adopting the above technical solution, by combining the student model parameters and the teacher model parameters, and using the preset parameter optimization formula for comprehensive calculation after introducing the update suppression parameter, more comprehensive and optimized teacher model parameters can be obtained. By making full use of the complementarity between different models, the accuracy of the teacher model parameters can be improved.

[0011] In a possible implementation, when the data set matching value is lower than a preset standard value, the method further includes: Identify the environmental parameter information and vegetation parameter information corresponding to each bird identification information in the bird identification data set, wherein the environmental parameter information includes water area parameters, water quality parameters and soil parameters, and the vegetation parameter information includes vegetation type and vegetation area; Identify weather parameter information corresponding to each bird identification information in the bird identification data set from the collected log data, wherein the weather parameter information includes temperature, precipitation, and light intensity; Acquire a simulation area model corresponding to the area to be monitored, and drive the simulation area model to perform simulation based on the environmental parameter information, the vegetation parameter information, and the weather parameter information, respectively, to obtain an environmental impact simulation data set, a vegetation impact simulation data set, and a weather impact simulation data set; The environmental impact simulation data set, the vegetation impact simulation data set and the weather impact simulation data set are matched with the standard data set respectively to determine the focus simulation data set, and the matching difference between the focus simulation data set and the standard data set is the largest.

[0012] By adopting the above technical solution, when the data set matching value is lower than the preset standard value, the simulation area model can be driven separately according to the environmental parameter information, vegetation parameter information and weather parameter information of the area to be monitored, so as to analyze the impact of factors such as environment, vegetation and weather on bird activities or migration. Finally, the simulation data set with the greatest impact on bird activities or migration will be fed back, so that relevant staff can adjust the possible anomalies in the area to be monitored according to the simulation data set, thereby reducing the impact of the parameter information on birds and maintaining the ecological balance in the area to be monitored.

[0013] In one possible implementation, the method further includes: Based on the environmental parameter information, the vegetation parameter information and the weather parameter information, the simulation area model is driven to perform simulation, and an environmental impact layer, a vegetation impact layer and a weather impact layer are obtained respectively; When a layer overlay instruction is detected, the layer overlay instruction is identified to determine the affected layer to be overlaid, and the layer to be overlaid is overlaid to obtain overlay layer feedback information.

[0014] By adopting the above technical solution, by displaying the simulation results corresponding to the environmental parameter information, vegetation parameter information and weather parameter information in the form of influence layers, it is convenient to intuitively display the impact of different parameter information on birds. In addition, different influence layers can be combined and superimposed according to viewing needs, so as to analyze the impact of the combination of different parameter information on the activities of birds, thereby facilitating in-depth understanding and analysis of the rules and connections between the parameter information.

[0015] In one possible implementation, the method further includes: When the bird identification information corresponding to the target image contains a preset abnormal feature, locating the abnormal position based on the target image; Obtaining an inspection position of a preset inspection device, generating an inspection instruction based on the inspection position and the abnormal position, controlling the preset inspection device to perform inspection according to the inspection instruction, and taking a close-up photo of the abnormal position to obtain a close-up image; Abnormal feedback information is generated based on the preset abnormal feature and the close-range image.

[0016] By adopting the above technical solution, by performing feature recognition and monitoring on the target image, it is convenient to timely discover possible abnormal situations in the target image. When there are abnormal situations in the target image, inspection instructions can be timely planned according to the inspection position and the abnormal position, so as to realize remote control of the inspection equipment to take pictures of the abnormal position, thereby facilitating a more intuitive and clearer understanding of the abnormal situation, and then facilitating timely response and elimination of the abnormal situation.

[0017] In a second aspect, the present application provides a monitoring system, which adopts the following technical solution: A monitoring system, comprising: 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 above-mentioned bird intelligent monitoring and identification method.

[0018] In a third aspect, the present application provides a computer-readable storage medium, which adopts the following technical solution: A computer-readable storage medium includes: a computer program that can be loaded by a processor and execute the above-mentioned bird intelligent monitoring and identification method.

[0019] In a fourth aspect, the present application provides a computer program product, which adopts the following technical solution: A computer program product includes a computer program, and when the computer program is executed by a processor, the computer program implements the above-mentioned bird intelligent monitoring and identification method.

[0020] In summary, the present application includes at least one of the following beneficial technical effects: By importing the target image into the trained target recognition model, it is convenient to quickly and accurately identify the bird identification information contained in the target image. When the bird identification information is determined by the trained target recognition model, the noise in the target image, the changes in lighting conditions or the shooting angle have little impact on the recognition result. It is also convenient to reduce the impact of human misjudgment or prejudice on the accuracy of the recognition result. Finally, by matching the standard data set with the bird identification data set, it is convenient to timely monitor possible abnormal situations in the monitored area, and by generating prompt feedback information, it is convenient to remind relevant staff to pay attention to and deal with abnormalities in a timely manner.

[0021] When the dataset matching value is lower than the preset standard value, the simulation area model can be driven separately according to the environmental parameter information, vegetation parameter information and weather parameter information of the monitored area to analyze the impact of environmental, vegetation and weather factors on bird activities or migration. Finally, the simulation dataset with the greatest impact on bird activities or migration will be fed back to facilitate relevant staff to adjust the possible anomalies in the monitored area according to the simulation dataset, thereby reducing the impact of the parameter information on birds and maintaining the ecological balance in the monitored area. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a flowchart of a method for intelligent bird monitoring and identification in an embodiment of the present application; Figure 2 is a schematic diagram of a process for determining a simulation data set of interest in an embodiment of the present application; Figure 3 It is a structural schematic diagram of a monitoring system in an embodiment of the present application. DETAILED DESCRIPTION

[0023] The following is combined with Figures 1 to 3 This application is described in further detail.

[0024] After reading this specification, those skilled in the art may make non-creative modifications to this embodiment as needed, but such modifications are protected by patent law as long as they are within the scope of the claims of this application.

[0025] 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.

[0026] It should be noted that in the optional embodiments of the present application, the object information and other related data involved, when the embodiments in the present application are applied to specific products or technologies, need to obtain the permission or consent of the object, and the collection, use and processing of the relevant data need to comply with the relevant laws, regulations and standards of the relevant countries and regions. In other words, if the embodiments of the present application involve data related to the object, it needs to be obtained with the authorization and consent of the object, the authorization and consent of the relevant departments, and in compliance with the relevant laws, regulations and standards of the country and region. If personal information is involved in the embodiments, the acquisition of all personal information needs to obtain the consent of the individual. If sensitive information is involved, the separate consent of the information subject needs to be obtained. The embodiments also need to be implemented with the authorization and consent of the object.

[0027] Specifically, the embodiment of the present application provides a method for intelligent monitoring and identification of birds, which is executed by a monitoring system, and the monitoring system 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 that provides 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, and the embodiment of the present application does not limit this.

[0028] refer to Figure 1 , Figure 1 : is a flow chart of a method for intelligent bird monitoring and identification in an embodiment of the present application, the method comprising steps S110-S140, wherein: Step S110: Acquire a target image, where the target image is an image corresponding to the area to be monitored.

[0029] Specifically, the area to be monitored is an area where bird monitoring and identification is required, which may be a wetland, a forest mountain area, an urban park or a green space, etc. The specific area is not specifically limited in the embodiment of the present application. The target image can be collected by an image acquisition device set in the area to be monitored and uploaded to the monitoring system, and the target image needs to contain a bird to be identified.

[0030] Step S120: importing the target image into the trained target recognition model to obtain corresponding bird recognition information.

[0031] Specifically, the trained target recognition model can accurately identify the bird information contained in the target image, such as the features of the bird's beak, feet, wings or neck, as well as the contour, color and texture details corresponding to each extracted feature. The target recognition model is a semi-supervised fine-grained training method, and a double-threshold loss calculation method is used to verify whether the training results are qualified. Therefore, the accuracy of determining bird recognition information using the trained target recognition model is high. The embodiment of the present application provides a training process for a target recognition model, which may specifically include: Obtain an initial recognition model and training sample data, the initial recognition model includes an initial teacher recognition model and an initial student recognition model, and the training sample data includes unlabeled sample images and labeled sample images; perform the following loop steps until the loss output value corresponding to the initial recognition model is lower than a preset threshold, then stop the loop, and determine the initial recognition model with a loss output value lower than the preset threshold as the target recognition model; The cycle steps include: The unlabeled sample images are imported into the initial teacher recognition model to obtain the pseudo-labels corresponding to the unlabeled sample images; the initial student recognition model is trained according to the unlabeled sample images and the corresponding pseudo-labels as well as the labeled sample images to obtain an updated initial student recognition model; the updated initial student recognition model is optimized based on a preset exponential moving average algorithm to obtain an updated teacher recognition model; the loss output value of the initial recognition model is determined based on a preset loss function.

[0032] Among them, the training sample data includes unlabeled sample images and labeled sample images. In the embodiment of the present application, a mixture of unlabeled sample images and labeled sample images is used to train the initial recognition model, thereby reducing the demand for training sample data during the training process. The initial recognition model is composed of an initial teacher recognition model and an initial student recognition model. The process of training the initial recognition model using mixed training sample data can be regarded as a semi-supervised learning process, wherein it is necessary to predict the unlabeled sample images according to the initial teacher recognition model, generate pseudo labels corresponding to the unlabeled pseudo images, and bind the obtained pseudo labels to the corresponding unlabeled sample images to the pseudo label sample images, and then train the initial student recognition model based on the pseudo label sample images and the labeled sample images. In order to improve the training rate and effect of the initial student recognition model, before training the initial student recognition model based on the pseudo label sample images, the pseudo label sample images and the labeled sample images are preprocessed. The operation of preprocessing the pseudo label sample images is different from that of preprocessing the labeled sample images. When preprocessing the pseudo label sample images, it is necessary to perform threshold filtering on the pseudo labels of the pseudo label sample images, that is, to remove pseudo labels exceeding the preset standard threshold to improve the quality of the pseudo labels, thereby improving the quality of the pseudo label sample images. When preprocessing the labeled sample image, the label edge features in the labeled sample image can be identified to determine whether the label corresponding to the labeled sample image is a weak label, and the weak label enhancement processing can be performed on the existing weak label, wherein the weak label enhancement processing includes but is not limited to HSV adjustment, rotation, translation, flipping and Mosaic, and Mosaic processing includes random cropping, scaling and image merging operations. The method of preprocessing the labeled sample image is not specifically limited in the embodiments of the present application.

[0033] After preprocessing the pseudo-label sample images and the labeled sample images, the initial student recognition model is trained based on the results obtained by the preprocessing until the student loss function value of the initial student recognition model is lower than the preset student threshold. The specific student threshold is not specifically limited in the embodiment of the present application and can be set by relevant technical personnel. Among them, the student loss function value is the weighted sum of the supervised loss function value and the unsupervised loss function value, wherein the supervised loss function value is obtained after training based on the labeled sample images, and the unsupervised loss function value is obtained after training based on the pseudo-label sample images. The calculation formula of the student loss function value is: L=Ls+WuLu, wherein L is used to characterize the student loss function value, Ls is the supervised loss function value, Lu is the unsupervised loss function value, Wu is the weight of the unsupervised loss, and the specific Wu value is not specifically limited in the embodiment of the present application and can be set by relevant technical personnel according to actual needs.

[0034] When the student loss function value of the initial student identification model is lower than the preset student threshold, the initial student identification model is optimized according to the preset exponential moving average algorithm to obtain an updated teacher identification model to optimize the teacher identification model. The embodiment of the present application provides a method for optimizing and updating the initial student identification model based on the preset exponential moving average algorithm, which may specifically include: Obtain the student model parameters corresponding to the updated initial student model, and the teacher model parameters corresponding to the updated teacher identification model; obtain the preset update inhibition parameters, import the update inhibition parameters, the student model parameters and the teacher model parameters into the preset parameter optimization formula to obtain the optimized model parameters; update the teacher model parameters based on the optimized model parameters to update the teacher identification model.

[0035] Specifically, update suppression parameters, student model parameters, and teacher model parameters can be imported into ,in, Used to characterize the update inhibition parameter. In the embodiment of the present application, the update inhibition parameter is 0.5. The specific numerical value is not specifically limited in the embodiment of the present application and can be set by relevant staff according to actual needs; θt is used to characterize the teacher model parameters, θs is used to characterize the student model parameters, and θxt is used to characterize the optimization model parameters, that is, the optimized teacher model parameters. The teacher model parameters are the weights, biases, activation functions, etc. of the initial teacher recognition model. The student model parameters correspond to the teacher model parameters, but the number and complexity are lower than the teacher model parameters. The teacher model parameters can be obtained or the student model parameters can be obtained during the semi-supervised training process. By updating or adjusting the teacher model parameters through the determined optimization model parameters, it is convenient to update and optimize the initial teacher recognition model.

[0036] The updating strategy of the initial teacher identification model is to integrate the parameters of the initial teacher identification model in the previous (t-1) iteration and the initial student identification model in the tth iteration. The initial teacher identification model can be regarded as an integrated model of the initial student identification model at different training stages, that is, part of the weight of the initial teacher identification model comes from the weight of the historical initial teacher identification model, and the other part comes from the weight of the initial student identification model in the current stage. Therefore, with the interaction between the initial student identification model and the initial teacher identification model, the two types of models can evolve together and continuously improve the prediction accuracy.

[0037] After optimizing the teacher identification model, the updated teacher identification model is used again to predict the pseudo-labels corresponding to other unlabeled sample images to obtain pseudo-label sample images, and the initial student model is trained again using the pseudo-label sample images and other labeled sample images. The cycle is executed, and at the end of each cycle, the loss output value corresponding to the initial identification model is calculated according to the preset loss function until the loss output value corresponding to the initial identification model is lower than the preset threshold, wherein the preset loss function can be a dual-threshold loss calculation function. The specific loss calculation function and the preset threshold are not specifically limited in the embodiments of the present application, as long as it is possible to verify whether the initial identification model has completed training.

[0038] Step S130: Acquire all bird identification information corresponding to a preset time period, and integrate all bird identification information within the preset time period to obtain a bird identification data set.

[0039] Specifically, the preset time period can be a period of time before the current moment, and the duration corresponding to the preset time period can be 10 days or 20 days. The specific duration is not specifically limited in the embodiment of the present application. After sorting all the bird identification information according to the time when each bird identification information is obtained, a bird identification data set can be obtained, and the bird identification data set is convenient for overall analysis of the activities or migration of birds in the monitoring area.

[0040] Step S140: obtaining a standard data set corresponding to a preset time period, matching the standard data set with the bird identification data set to obtain a data set matching value, and when the data set matching value is lower than a preset standard value, generating prompt feedback information based on the bird identification data set.

[0041] Specifically, the standard data set can be an average data set determined by relevant staff based on historical monitoring and identification data. For example, the preset time period is October 1, 2023 to October 31, 2023, and the standard data set is the average of the monitoring and identification data from October 1 to October 30, 2013 to 2022. The specific content is not specifically limited in the embodiments of the present application. By matching the bird identification dataset corresponding to the preset time period with the corresponding standard dataset, it is convenient to analyze and predict whether there are any abnormalities in the activities or migration of birds within the preset time period. This can be determined specifically by the dataset matching value. When the dataset matching value is not lower than the preset standard value, it indicates that the migration of bird activities in the monitored area within the preset time period is normal. For example, xx migratory birds generally stay in the monitored area for a short period of time from October 1 to October 30 to prepare for the subsequent long-distance flight. The average number of birds staying is 200. If it is determined based on the bird identification dataset that only 100 xx migratory birds stay in the monitored area within the preset time period, the corresponding dataset matching value may be lower than the preset standard value. That is, a prompt feedback information may be generated at this time to remind relevant staff to promptly deal with the ecological and environmental problems that may be faced by the monitored area.

[0042] For the embodiments of the present application, by importing the target image into a trained target recognition model, it is convenient to quickly and accurately identify the bird identification information contained in the target image. When the bird identification information is determined by the trained target recognition model, the noise in the target image, the changes in lighting conditions or the shooting angle have little impact on the recognition result. It is also convenient to reduce the impact of human misjudgment or prejudice on the accuracy of the recognition result. Finally, by matching the standard data set with the bird identification data set, it is convenient to timely monitor possible abnormal situations in the monitored area, and by generating prompt feedback information, it is convenient to remind relevant staff to pay attention to and handle the abnormalities in a timely manner.

[0043] Furthermore, when the data set matching value is lower than the preset standard value, the method provided in the embodiment of the present application further includes steps S1 to S4, such as Figure 2 As shown, where: Step S1: Identify the environmental parameter information and vegetation parameter information corresponding to each bird identification information in the bird identification data set, the environmental parameter information includes water area parameters, water quality parameters and soil parameters, and the vegetation parameter information includes vegetation type and vegetation area.

[0044] Specifically, the bird identification data set includes bird identification information corresponding to each acquisition moment within a preset time period, wherein the target image can be acquired within the preset time period according to the preset acquisition frequency, and the bird identification information is acquired based on the acquisition frequency. For example, the target image is acquired every 3 days, and the preset time period is 30 days, that is, the bird identification data set corresponding to the preset time period includes 10 bird identification information. Since the bird identification information is obtained from the target image, it is convenient to determine the environmental parameter information and vegetation parameter information contained in the corresponding target image by analyzing the bird identification information.

[0045] The water area parameters, i.e., the water area, can be identified from the corresponding target image by feature recognition. The specific feature recognition algorithm is not specifically limited in the embodiment of the present application. For example, an edge detection algorithm can be used. The water quality parameters and soil parameters can be collected by sensors arranged in the area to be monitored and uploaded to the monitoring system. Since the food source of birds may come from organisms in the water or soil, such as fish or insects, once the water quality or soil in the area to be monitored deteriorates, it is likely to affect the number of fish or insects, thereby affecting the food supply of birds. Therefore, when it is determined by analyzing the bird identification data set in the area to be monitored that the data set matching value is lower than the preset standard value, it may indicate that there is an abnormality in the water quality or soil in the area to be monitored, so it is necessary to analyze the water quality or soil in the area to be monitored.

[0046] Similarly, since vegetation can provide birds with places to nest, forage, rest, and hide, and different types of vegetation can support the survival of different types of birds, when the vegetation in the monitored area is polluted, damaged, or degraded, it may affect the bird's habitat and food source. Therefore, when it is determined that the data set matching value is lower than the preset standard value, it may indicate that the vegetation in the monitored area is abnormal, so it is necessary to analyze the vegetation in the monitored area. Among them, the vegetation parameter information can be identified from the target image corresponding to the bird identification information by feature recognition, and the specific feature recognition algorithm is not specifically limited in the embodiment of this application.

[0047] Step S2: Identify weather parameter information corresponding to each bird identification information in the bird identification data set from the collected log data, where the weather parameter information includes temperature, precipitation, and light intensity.

[0048] Specifically, the weather parameter information corresponding to each bird identification information, that is, the weather parameter information corresponding to each acquisition time, can be collected by the weather sensor set in the monitored area, uploaded to the monitoring system and converted into collection log data, which can be retrieved from the collection log data based on the acquisition time. Since excessively high or low temperatures can affect the physiological functions, foraging, and migration of birds, and precipitation and light intensity can affect the food supply and habitat of birds in the monitored area, when it is determined that the data set matching value is lower than the preset standard value, it is also necessary to analyze the weather parameters in the monitored area.

[0049] Based on the above method, the environmental parameter information, vegetation parameter information and weather parameter information corresponding to each acquisition moment within the preset time period can be determined, and the environmental parameter information, vegetation parameter information and weather parameter information corresponding to each acquisition moment can be integrated to obtain the integrated environmental parameter information, integrated vegetation parameter information and integrated weather parameter information corresponding to the preset time period.

[0050] Step S3: Obtain the simulation area model corresponding to the monitored area, and drive the simulation area model to perform simulation based on the environmental parameter information, vegetation parameter information, and weather parameter information, to obtain the environmental impact simulation data set, vegetation impact simulation data set, and weather impact simulation data set, respectively.

[0051] Specifically, the simulation area model corresponding to the area to be monitored has a consistent ecological environment with the area to be monitored, and can simulate the activities of birds in the area to be monitored. The simulation area model can be determined by relevant staff based on historical experimental data and uploaded to the monitoring system.

[0052] For environmental parameter information, the simulation regional model is driven by the environmental parameter information for simulation, that is, the simulation regional model is driven by the integrated environmental parameter information for simulation, that is, the simulation regional model is simulated only according to the actual environmental parameter information within the preset time period, and the activity of birds in the monitored area within the preset time period is simulated and reviewed. Among them, while driving the simulation regional model according to the environmental parameter information, it is necessary to adopt the default vegetation parameter information and the default weather parameter information to jointly drive the simulation regional model. The default vegetation parameter information and the default weather parameter information can be determined by relevant staff from historical collected data, which are the average values ​​corresponding to the same historical stage. For example, the preset time period is October 1, 2023-October 30, 2023, and a certain acquisition time is October 1. The default vegetation type and default vegetation area corresponding to October 1 are the average values ​​of all vegetation types and vegetation areas corresponding to October 1 in the historical collected data from 2010 to 2020. After simulating the simulation area model according to the integrated environmental parameter information, the obtained environmental impact simulation data set is a simulated bird identification data set corresponding to the monitored area under the influence of the integrated environmental parameter information.

[0053] Based on the above content, a vegetation impact simulation data set and a weather impact simulation data set can be obtained. When determining the vegetation impact data set, it is necessary to drive the simulation regional model based on the integrated vegetation parameter information, the default environmental parameter information and the default weather parameter information; when determining the weather impact simulation data set, it is necessary to drive the simulation regional model based on the integrated weather parameter information, the default environmental parameter information and the simulated vegetation parameter information.

[0054] Step S4: respectively matching the environmental impact simulation data set, the vegetation impact simulation data set and the weather impact simulation data set with the standard data set to determine the focus simulation data set, and the matching difference between the focus simulation data set and the standard data set is the largest.

[0055] Specifically, after determining the environmental impact simulation data set, vegetation impact simulation data set and weather impact simulation data set, the environmental impact simulation data set, vegetation impact simulation data set and weather impact simulation data set can be matched with the standard data set respectively to obtain the environmental matching difference, vegetation matching difference and weather matching difference respectively. Finally, the environmental matching difference, vegetation matching difference and weather matching difference are compared to determine the simulation data set of interest with the largest matching difference. For example, when the simulation data set of interest is the vegetation impact simulation data set, it can be determined that the main reason for the data set matching value in the monitored area to be lower than the preset standard value is that the vegetation in the monitored area has an abnormality. By providing feedback on the simulation data set of interest that has the greatest impact on bird activities or migration, it is convenient for relevant staff to adjust the possible anomalies in the monitored area according to the simulation data set of interest, thereby reducing the impact of the parameter information of interest on birds and maintaining the ecological balance in the monitored area.

[0056] Furthermore, in order to intuitively display the effects of different parameter information on birds, the method provided in the embodiment of the present application also includes: The simulation area model is driven to perform simulation based on the environmental parameter information, vegetation parameter information and weather parameter information to obtain the environmental impact layer, vegetation impact layer and weather impact layer respectively; when a layer overlay instruction is detected, the layer overlay instruction is identified to determine the impact layer to be overlaid, and the layer to be overlaid is overlaid to obtain the overlay layer feedback information.

[0057] Specifically, the simulation is driven by the simulation regional model based on the environmental parameter information, that is, the regional simulation model is driven by the integrated environmental parameter information, the default vegetation parameter information and the default weather parameter information. The corresponding environmental impact layer is an intuitive display method of the environmental impact simulation data set, that is, the environmental impact layer is a simulated image of the monitored area within a preset time period, which is mainly used to show the impact of the integrated environmental parameter information on bird activities or migration. When determining the environmental impact layer, the output type of the simulation regional model can be adjusted, that is, the original output environmental impact simulation data set is adjusted to the output environmental image layer. Based on the above content, the vegetation impact layer and the weather impact layer can be obtained.

[0058] The layer overlay instruction can be sent to the monitoring system by the relevant access personnel through the access terminal. The layer overlay instruction contains the overlay requirement, that is, the layer that needs to be overlaid. For example, the overlay requirement can be to overlay the environmental impact layer with the vegetation impact layer to analyze the impact of the combination of different parameter information on the activities of birds. The overlay requirement contained in the layer overlay instruction can be determined by semantic recognition.

[0059] Furthermore, in order to promptly respond to and eliminate abnormal conditions in the target image, the method provided in the embodiment of the present application further includes: When the bird identification information corresponding to the target image contains preset abnormal features, the abnormal position is located based on the target image; the inspection position of the preset inspection equipment is obtained, and an inspection instruction is generated based on the inspection position and the abnormal position, and the preset inspection equipment is controlled to inspect according to the inspection instruction, and the abnormal position is photographed at close range to obtain a close-up image; and abnormal feedback information is generated based on the preset abnormal features and the close-up image.

[0060] Specifically, in the process of acquiring the target image in real time, it is also necessary to identify and analyze the abnormal conditions that may exist in the target image. Since the image acquisition devices set in the monitored area are mostly fixed devices, when the preset abnormal features appear in the target image, there may be blurring. At this time, a patrol route can be formulated based on the abnormal position of the preset abnormal feature and the patrol position of the patrol device to control the patrol device to move from the patrol position to the abnormal position along the patrol route for close-up shooting, without moving the image acquisition device or adjusting the shooting parameters of the image acquisition device, etc., that is, by controlling the patrol device to take close-up shots of the abnormal position, it is convenient to intuitively view the abnormal situation while not affecting the normal target image acquisition process. Among them, the patrol device can be a drone image acquisition device, and the specific patrol device is not specifically limited in the embodiments of the present application, as long as it can take close-up shots of the designated position according to the patrol instruction.

[0061] Among them, the preset abnormal features can be distorted bird posture, missing legs or wings, dull feathers, bloody feathers, etc. When the target image contains the preset abnormal features, it may indicate that the corresponding bird in the target image is injured or killed. The specific preset abnormal features are not specifically limited in the embodiments of this application and can be set by relevant staff. By performing feature recognition and monitoring on the target image, it is convenient to timely discover possible abnormal situations in the target image. When there are abnormal situations in the target image, inspection instructions can be planned in time according to the inspection position and the abnormal position, so as to realize remote control of the inspection equipment to shoot the abnormal position, so as to facilitate a more intuitive and clearer understanding of the abnormal situation, and then to facilitate timely response and elimination of the abnormal situation.

[0062] The present application provides a monitoring system, such as Figure 3 As shown, Figure 3The monitoring system 300 shown includes: a processor 301 and a memory 303. The processor 301 and the memory 303 are connected, such as through a bus 302. Optionally, the monitoring system 300 may also include a transceiver 304. It should be noted that in actual applications, the transceiver 304 is not limited to one, and the structure of the monitoring system 300 does not constitute a limitation on the embodiments of the present application.

[0063] The processor 301 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. The processor 301 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.

[0064] The bus 302 may include a path to transmit information between the above components. The bus 302 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus 302 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 3 The fact that only one line is used in the diagram does not mean that there is only one bus or only one type of bus.

[0065] The memory 303 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.

[0066] The memory 303 is used to store the application code for executing the solution of the present application, and the execution is controlled by the processor 301. The processor 301 is used to execute the application code stored in the memory 303 to implement the contents shown in the above method embodiment.

[0067] The monitoring system includes, but is not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. It can also be a server, etc. Figure 3 The monitoring system 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.

[0068] 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.

[0069] The embodiment of the present application provides a computer program product, which includes a computer program, which implements the method in any of the above embodiments when executed by a processor. Compared with the related art, in the embodiment of the present application, by importing the target image into a trained target recognition model, it is convenient to quickly and accurately identify the bird recognition information contained in the target image. When the bird recognition information is determined by the trained target recognition model, the noise in the target image, the change of lighting conditions or the shooting angle have a small impact on the recognition result, and it is also convenient to reduce the impact of human misjudgment or prejudice on the accuracy of the recognition result. Finally, by matching the standard data set with the bird recognition data set, it is convenient to timely monitor the abnormal conditions that may exist in the monitored area, and by generating prompt feedback information, it is convenient to remind relevant staff to pay attention to and deal with the abnormalities in a timely manner.

[0070] 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.

[0071] The above description is only a partial implementation method 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 intelligent bird monitoring and identification, characterized in that: include: Acquire a target image, where the target image is an image corresponding to the area to be monitored; Importing the target image into a trained target recognition model to obtain corresponding bird recognition information; Obtain all bird identification information corresponding to a preset time period, and integrate all bird identification information within the preset time period to obtain a bird identification data set; A standard data set corresponding to the preset time period is obtained, and the standard data set is matched with the bird identification data set to obtain a data set matching value. When the data set matching value is lower than a preset standard value, prompt feedback information is generated based on the bird identification data set.

2. The method for intelligent bird monitoring and identification according to claim 1, characterized in that: The training process of the target recognition model includes: Obtaining an initial recognition model and training sample data, wherein the initial recognition model includes an initial teacher recognition model and an initial student recognition model, and the training sample data includes unlabeled sample images and labeled sample images; Execute the following loop steps until the loss output value corresponding to the initial recognition model is lower than the preset threshold, then stop the loop, and determine the initial recognition model with the loss output value lower than the preset threshold as the target recognition model; The cycle steps include: Importing the unlabeled sample image into the initial teacher recognition model to obtain a pseudo label corresponding to the unlabeled sample image; The initial student recognition model is trained according to the unlabeled sample images and the corresponding pseudo labels as well as the labeled sample images to obtain an updated initial student recognition model; Optimizing the updated initial student identification model based on a preset exponential moving average algorithm to obtain an updated teacher identification model; The loss output value of the initial recognition model is determined based on a preset loss function.

3. A method for intelligent bird monitoring and identification according to claim 2, characterized in that: The method of optimizing the updated initial student identification model based on a preset exponential moving average algorithm to obtain an updated teacher identification model includes: Obtaining student model parameters corresponding to the updated initial student model and teacher model parameters corresponding to the updated teacher identification model; Obtaining a preset update suppression parameter, and importing the update suppression parameter, the student model parameter, and the teacher model parameter into a preset parameter optimization formula to obtain an optimized model parameter; The teacher model parameters are updated based on the optimized model parameters to update the teacher identification model.

4. The method for intelligent bird monitoring and identification according to claim 1, characterized in that: When the data set matching value is lower than a preset standard value, the method further includes: Identify the environmental parameter information and vegetation parameter information corresponding to each bird identification information in the bird identification data set, wherein the environmental parameter information includes water area parameters, water quality parameters and soil parameters, and the vegetation parameter information includes vegetation type and vegetation area; Identify weather parameter information corresponding to each bird identification information in the bird identification data set from the collected log data, wherein the weather parameter information includes temperature, precipitation, and light intensity; Acquire a simulation area model corresponding to the area to be monitored, and drive the simulation area model to perform simulation based on the environmental parameter information, the vegetation parameter information, and the weather parameter information, respectively, to obtain an environmental impact simulation data set, a vegetation impact simulation data set, and a weather impact simulation data set; The environmental impact simulation data set, the vegetation impact simulation data set and the weather impact simulation data set are matched with the standard data set respectively to determine the focus simulation data set, and the matching difference between the focus simulation data set and the standard data set is the largest.

5. The method for intelligent bird monitoring and identification according to claim 4, characterized in that: Also includes: Based on the environmental parameter information, the vegetation parameter information and the weather parameter information, the simulation area model is driven to perform simulation, and an environmental impact layer, a vegetation impact layer and a weather impact layer are obtained respectively; When a layer overlay instruction is detected, the layer overlay instruction is identified to determine the affected layer to be overlaid, and the layer to be overlaid is overlaid to obtain overlay layer feedback information.

6. The method for intelligent bird monitoring and identification according to claim 1, characterized in that: Also includes: When the bird identification information corresponding to the target image contains a preset abnormal feature, locating the abnormal position based on the target image; Obtaining an inspection position of a preset inspection device, generating an inspection instruction based on the inspection position and the abnormal position, controlling the preset inspection device to perform inspection according to the inspection instruction, and taking a close-up photo of the abnormal position to obtain a close-up image; Abnormal feedback information is generated based on the preset abnormal feature and the close-range image.

7. A monitoring system, characterized in that: The monitoring system includes: 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 a bird intelligent monitoring and identification method according to any one of claims 1-6.

8. A computer-readable storage medium, characterized in that: include: A computer program is stored which can be loaded by a processor and execute a method for intelligent monitoring and identifying birds as described in any one of claims 1 to 6.

9. A computer program product, characterized in that The invention comprises a computer program, which, when executed by a processor, implements the steps of a method for intelligent monitoring and identifying birds according to any one of claims 1 to 6.

Images