Method, apparatus and system for identifying characteristic behaviors of a target object

Abstract

The present disclosure relates to a method, an apparatus, and a system for identifying characteristic behaviors of a target object, as well as a computer-readable storage medium. The method for identifying characteristic behaviors of a target object includes: obtaining data information of periodic characteristic behaviors of the target object; triggering a video surveillance device in the area where the target object is located to turn on when the data information of the periodic characteristic behaviors meets a video surveillance trigger condition; obtaining video surveillance information of the video surveillance device; determining the probability that the target object exhibits a target characteristic behavior based on the video surveillance information; and outputting reference information indicating that the target object exhibits the target characteristic behavior when the probability that the target object exhibits the target characteristic behavior is not less than a probability threshold.

Description

Technical Field

[0001] The present disclosure relates to the field of artificial intelligence technology, and particularly relates to a method, apparatus and system for identifying characteristic behaviors of a target object, and a computer-readable storage medium. Background Art

[0002] In related technologies, the identification of some typical characteristic behaviors of non-human biological target objects (such as the identification of the mounting behavior of cows in estrus) mostly adopts the method of manual observation, which has technical defects of low accuracy and poor real-time performance. Summary of the Invention

[0003] The present disclosure provides a method, apparatus and system for identifying characteristic behaviors of a target object, and a computer-readable storage medium.

[0004] According to one aspect of the present disclosure, there is provided a method for identifying characteristic behaviors of a target object, including:

[0005] Obtaining data information of periodic characteristic behaviors of a target object;

[0006] When the data information of the periodic characteristic behaviors meets the video monitoring trigger condition, triggering the video monitoring device in the area where the target object is located to be turned on;

[0007] Obtaining video monitoring information of the video monitoring device;

[0008] Determining the probability that the target object has a target characteristic behavior according to the video monitoring information;

[0009] When the probability that the target object has a target characteristic behavior is not less than a probability threshold, outputting reference information that the target object has a target characteristic behavior.

[0010] According to another aspect of the present disclosure, there is provided an apparatus for identifying characteristic behaviors of a target object, including:

[0011] A first obtaining unit, configured to obtain data information of periodic characteristic behaviors of a target object;

[0012] A triggering unit, configured to trigger the video monitoring device in the area where the target object is located to be turned on when the data information of the periodic characteristic behaviors meets the video monitoring trigger condition;

[0013] A second obtaining unit, configured to obtain video monitoring information of the video monitoring device;

[0014] A determining unit, configured to determine the probability that the target object has a target characteristic behavior according to the video monitoring information;

[0015] An output unit, configured to output reference information indicating that the target object has performed a target characteristic behavior when the probability that the target object has performed the target characteristic behavior is not less than a probability threshold.

[0016] According to another aspect of the present disclosure, there is provided a recognition system for target object characteristic behaviors, including a monitoring device, a video monitoring device, an output device, and a control device, where:

[0017] The monitoring device is disposed on the body of the target object and is configured to monitor and obtain data information of the periodic characteristic behaviors of the target object;

[0018] The video monitoring device is disposed in the area where the target object is located;

[0019] The control device is respectively connected to the monitoring device, the video monitoring device, and the output device, and is configured to trigger the video monitoring device in the area where the target object is located to turn on when the data information of the periodic characteristic behaviors satisfies the video monitoring trigger condition; determine the probability that the target object has performed the target characteristic behavior according to the video monitoring information; and output reference information indicating that the target object has performed the target characteristic behavior to the output device when the probability that the target object has performed the target characteristic behavior is not less than the probability threshold.

[0020] According to still another aspect of the present disclosure, there is provided a recognition device for target object characteristic behaviors, including a memory; and a processor coupled to the memory, the processor being configured to execute the recognition method for target object characteristic behaviors according to any one of the foregoing technical solutions based on instructions stored in the memory.

[0021] According to still another aspect of the present disclosure, there is provided a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the recognition method for target object characteristic behaviors according to any one of the foregoing technical solutions is implemented.

[0022] According to the foregoing technical solutions of the present disclosure, the accuracy and real-time performance of recognizing the target characteristic behaviors of the target object can be improved.

[0023] Other features and advantages of the present disclosure will become clear through the following detailed description of the embodiments of the present disclosure with reference to the accompanying drawings. Description of the Drawings

[0024] The drawings forming a part of the specification depict embodiments of the present disclosure and, together with the specification, are used to explain the principles of the present disclosure.

[0025] Referring to the accompanying drawings, the present disclosure can be more clearly understood according to the following detailed description, where:

[0026] Figure 1aFlowchart of the method for identifying the characteristic behaviors of the target object in some embodiments of the present disclosure;

[0027] Figure 1b Flowchart of determining the offline model based on the deep learning algorithm in some embodiments of the present disclosure;

[0028] Figure 2 Flowchart of the method for identifying the characteristic behaviors of the target object in some other embodiments of the present disclosure;

[0029] Figure 3 Block diagram of the identification system for the characteristic behaviors of the target object in some embodiments of the present disclosure;

[0030] Figure 4 Block diagram of the identification device for the characteristic behaviors of the target object in some embodiments of the present disclosure;

[0031] Figure 5 Block diagram of the identification device for the characteristic behaviors of the target object in some other embodiments of the present disclosure;

[0032] Figure 6 Block diagram of the computer system in some embodiments of the present disclosure.

[0033] It should be understood that the sizes of the various parts shown in the drawings are not drawn according to actual proportional relationships. In addition, the same or similar reference numerals represent the same or similar components. Detailed Description of the Specific Embodiments

[0034] Now, various exemplary embodiments of the present disclosure will be described in detail with reference to the accompanying drawings. The description of the exemplary embodiments is merely illustrative and in no way limits the present disclosure and its application or use. The present disclosure can be implemented in many different forms and is not limited to the embodiments described herein. These embodiments are provided to make the present disclosure thorough and complete, and to fully convey the scope of the present disclosure to those skilled in the art. It should be noted that: Unless otherwise specifically stated, the relative arrangements of the components and steps set forth in these embodiments should be construed as merely exemplary and not as limitations.

[0035] All terms used in the present disclosure (including technical terms or scientific terms) have the same meaning as understood by those of ordinary skill in the art to which the present disclosure pertains, unless otherwise specifically defined. It should also be understood that terms defined in a general dictionary, such as those, should be interpreted as having a meaning consistent with their meaning in the context of the relevant art and should not be interpreted in an idealized or overly formal sense, unless specifically defined as such herein.

[0036] Technologies, methods, and devices known to those of ordinary skill in the relevant art may not be discussed in detail, but where appropriate, the said technologies, methods, and devices should be regarded as part of the specification.

[0037] In the process of implementing the embodiments of the present disclosure, the inventors of the present application found that in the related art, the recognition of some typical characteristic behaviors of non-human biological target objects mostly adopts the method of manual observation, which has the technical defects of low accuracy, poor real-time performance, and consuming a large amount of manpower.

[0038] To solve this technical problem, the embodiments of the present disclosure provide a method, device, and system for recognizing characteristic behaviors of a target object, as well as a computer-readable storage medium and a computer system.

[0039] As Figure 1a shown, the method for recognizing characteristic behaviors of a target object provided by some embodiments of the present disclosure includes the following steps S101-S105.

[0040] In step S101, data information of periodic characteristic behaviors of the target object is acquired.

[0041] The specific type of the target object is not limited. For example, it can be pasture livestock such as cows, beef cattle, goats, or sheep, or it can be scientific research and breeding animals such as giraffes, sika deer, or antelopes.

[0042] The periodic characteristic behaviors of the target object can be understood as the periodic and regular activity behaviors of the target object, such as walking behavior or rumination behavior. Among them, rumination refers to the process in which some animals return semi-digested food from the stomach to the mouth for chewing again after a period of eating. Rumination can be carried out multiple times and shows a certain periodicity.

[0043] This periodic characteristic behavior of walking can be monitored by Internet of Things wearable devices such as intelligent hoof rings, intelligent ear rings, or intelligent collars, so as to obtain data information such as the number of steps and frequency, which is similar to the principle of people wearing intelligent bracelets to monitor exercise. For this periodic characteristic behavior of rumination, the mouth movements or chewing sounds of the animal can be monitored through an intelligent collar, so as to obtain data information such as the rumination frequency and duration of the animal.

[0044] In some embodiments of the present disclosure, the data information of the periodic characteristic behaviors of the target object may simultaneously include the above-mentioned step counting information and rumination information.

[0045] In step S102, when the data information of the periodic characteristic behavior meets the video monitoring trigger condition, the video monitoring device in the area where the target object is located is triggered to turn on.

[0046] Research has confirmed that when livestock is about to enter or has entered certain typical physiological characteristic periods, the data information of its periodic characteristic behaviors generally shows significant changes.

[0047] Taking cows as an example, cows are animals with a cyclic estrus. Under the influence of balanced estrogen, the estrus cycle of healthy cows is generally about 21 days. The estrus behavior usually lasts for 18 - 24 hours and usually starts at night. Two typical characteristics accompanied by a cow entering the estrus period are increased movement and a decrease in the duration of rumination. By monitoring the step count information and rumination information of cows, it is possible to roughly determine whether a cow is about to enter or has entered the estrus period.

[0048] Based on a similar principle, in some embodiments, it is also possible to determine whether a target object has sleep behavior or foraging behavior by observing the periodic characteristic behaviors of the target object. For example, when the number of steps of a cow does not change for a long time, it can be roughly determined that the cow may be entering sleep; when the number of steps of a cow continues to increase and the number of rumination times is zero for a long time, it can be roughly determined that the cow may be hungry and in a foraging state.

[0049] Therefore, the video monitoring trigger condition can be set according to the above experience. When the data information of the periodic characteristic behavior meets this video monitoring trigger condition, the video monitoring device in the area where the target object is located is triggered to turn on, thereby starting to conduct video monitoring on the target object.

[0050] In step S103, obtain the video monitoring information of the video monitoring device.

[0051] In an embodiment of the present disclosure, the video monitoring information of the video monitoring device can be obtained in real time.

[0052] In another embodiment of the present disclosure, step S103 includes: every time a set time period Ts elapses, obtain the video monitoring information collected by the video monitoring device within this set time period Ts.

[0053] The set time period Ts can be determined in combination with the system processing performance and the requirements of business real-time monitoring. The set time period Ts can take values within the range of 1 - 60 minutes. For example, starting from the time when the video monitoring device is turned on, every 10 minutes, obtain the video monitoring information for the just-passed 10 minutes.

[0054] In step S104, according to the video monitoring information, determine the probability that the target object has the target characteristic behavior.

[0055] Taking cows as an example, when a cow enters the estrus period, it usually exhibits typical behaviors different from normal behaviors (i.e., the target characteristic behaviors that the breeding personnel are concerned about), such as mounting behavior. During an estrus cycle, the number of occurrences of mounting behavior can be as many as 8 times. Other types of animals will also exhibit some typical behaviors different from normal behaviors during the estrus period. Through monitoring by computer vision technology, the occurrence of mounting behavior can be timely known, thereby effectively predicting animal estrus.

[0056] In one embodiment, step S104 includes: determining the probability that the target object exhibits the target characteristic behavior based on the video surveillance information and the standard model for the target object to exhibit the target characteristic behavior, such as determining the probability that a dairy cow exhibits the mounting behavior. Among them, the standard model for the target object to exhibit the target characteristic behavior is obtained based on the deep learning algorithm.

[0057] Deep learning is a type of machine learning, and machine learning is the necessary path to achieve artificial intelligence. The concept of deep learning stems from the research of artificial neural networks. The multi-layer perceptron with multiple hidden layers is a deep learning structure. Deep learning forms more abstract high-level representations of attribute categories or features by combining low-level features to discover the distributed feature representations of data. The motivation for studying deep learning is to build a neural network that simulates the human brain for analysis and learning. It mimics the mechanism of the human brain to interpret data, such as images, sounds, and texts. Typical deep learning models include convolutional neural network models, deep belief network models, and stacked autoencoder network models, etc.

[0058] In one embodiment of the present disclosure, the standard model for the target object to exhibit the target characteristic behavior is an offline model determined based on the deep learning algorithm, that is, the standard model has been determined before step S101. As Figure 1b shown, the determination of the standard model includes the following steps S01 - S07.

[0059] Step S01: Obtain the video image data of the target object exhibiting the target characteristic behavior;

[0060] Step S02: Preprocess the video image data, and the preprocessing includes extracting frame image data, making annotations, and removing dirty data, etc.;

[0061] Step S03: Determine the training data set and the test data set according to the preprocessed video image data;

[0062] Step S04: Train and optimize the model according to the hyperparameters and the training data set;

[0063] Step S05: Test the model according to the test data set to obtain the accuracy rate of the model;

[0064] Step S06: Determine whether the accuracy rate of the model reaches the target accuracy rate. If so, the process proceeds to step S07; otherwise, the process returns to step S01;

[0065] Step S07: Determine the model as the standard model and output it.

[0066] In the method for identifying the characteristic behavior of a target object according to an embodiment of the present disclosure, in step S105, when the probability that the target object exhibits a target characteristic behavior is not less than a probability threshold, reference information on the target object's occurrence of the target characteristic behavior is output.

[0067] The probability threshold can be determined based on experience. In the early stage of applying this identification method, the probability threshold can be set relatively low, such as 60%; as the standard model and related algorithms are continuously improved, the probability threshold can be gradually increased, for example, increased to 90% or higher, to improve the accuracy of identifying the target characteristic behavior of the target object. The reference information on the target object's occurrence of the target characteristic behavior can include the probability that the target object exhibits the target characteristic behavior, and can also include the result of further judgment based on this probability, such as the judgment result of whether it is in the estrus period.

[0068] When the probability that the target object exhibits a target characteristic behavior is not less than the probability threshold, it can be basically determined that the target object exhibits the target characteristic behavior. For example, it can be basically determined that a dairy cow exhibits a mounting behavior and is in the estrus period. Outputting the reference information on the target object's occurrence of the target characteristic behavior can timely remind the breeding personnel that the dairy cow is in the estrus period, and the dairy cow can be artificially inseminated in time to improve the calving rate and milk yield.

[0069] The method for identifying the characteristic behavior of a target object according to the above embodiment of the present disclosure has the following beneficial effects:

[0070] When the data information of the periodic characteristic behavior meets the video monitoring trigger condition, the video monitoring device is triggered to turn on. Therefore, the video monitoring device does not need to be turned on all the time, thereby reducing the amount of data processing, saving system resources, and improving the processing efficiency.

[0071] Obtaining video monitoring information once every set time period can balance the system processing performance and the need for real-time business monitoring, save system resources while achieving the monitoring purpose, and improve the system processing speed.

[0072] The standard model for determining the target object's occurrence of the target characteristic behavior based on the deep learning algorithm has relatively high accuracy and intelligence level. Determining the probability that the target object exhibits the target characteristic behavior according to this standard model and the video monitoring information can greatly reduce the workload of breeding personnel or scientific research personnel, thereby helping to reduce the labor cost.

[0073] As Figure 2 shown, the method for identifying the characteristic behavior of a target object provided by some other embodiments of the present disclosure includes the following steps S201 - S209, where:

[0074] Step S201: Obtain the data information of the periodic characteristic behavior of the target object.

[0075] Step S202: Determine whether the data information of the periodic characteristic behavior meets the video monitoring trigger condition. If so, the process proceeds to step S203; otherwise, the process returns to step S201.

[0076] Step S203: Trigger the activation of the video monitoring device in the area where the target object is located.

[0077] Step S204: Every time a set time period elapses, obtain the video monitoring information collected by the video monitoring device during this set time period.

[0078] Step S205: Based on the video monitoring information and the standard model of the target object's target characteristic behavior, determine the probability of the target object's occurrence of the target characteristic behavior.

[0079] Step S206: Determine whether the probability of the target object's occurrence of the target characteristic behavior is less than the probability threshold. If so, the process proceeds to step S207; otherwise, the process proceeds to step S209.

[0080] Step S207: Determine whether the continuous activation duration of the video monitoring device is greater than the monitoring duration threshold. If so, the process proceeds to step S208; otherwise, the process returns to step S204.

[0081] Among them, the monitoring duration threshold can be determined according to the upper limit of the cycle of the target object's occurrence of the target characteristic behavior. For example, if the estrus behavior of dairy cows usually lasts for 18 - 24 hours, the monitoring duration threshold can be set to 24 hours.

[0082] Step S208: Turn off the video monitoring device.

[0083] Step S209: Output the reference information on the target object's occurrence of the target characteristic behavior. For example, output the reference information to a display for display, or output it to an audio device for voice reminder. In one embodiment, the reference information is pushed to a mobile terminal application in the form of WeChat or text message.

[0084] Similarly to the foregoing embodiments, by using the recognition method of this embodiment of the present disclosure, the accuracy and real-time performance of the recognition of the target object's target characteristic behavior can be improved, thereby timely reminding the breeding personnel or scientific research personnel to take certain measures. In addition, the video monitoring device is conditionally turned on and off, and the video monitoring information is obtained once every set time period, which can balance the system processing performance and the need for real-time business monitoring, saving system resources while achieving the monitoring purpose.

[0085] Such as Figure 3As shown in the figure, some embodiments of the present disclosure also provide a recognition system for the characteristic behaviors of a target object, including a control device 34, which is configured to trigger the activation of a video surveillance device in the area where the target object is located when the data information of the periodic characteristic behaviors meets the video surveillance trigger condition; determine the probability that the target object exhibits a target characteristic behavior based on the video surveillance information; and output reference information indicating that the target object exhibits the target characteristic behavior when the probability that the target object exhibits the target characteristic behavior is not less than a probability threshold.

[0086] The specific type of the control device 34 is not limited. For example, it can be a computer, a local area network server, or a cloud server, etc.

[0087] By using the recognition system of the embodiments of the present disclosure, the accuracy and real-time performance of recognizing the target characteristic behaviors of the target object can be improved, and the degree of intelligence is relatively high. It can timely remind the breeding personnel or scientific research personnel to take certain measures, so as to achieve the purpose of increasing the breeding output or scientific research.

[0088] Please refer to Figure 3 As shown in the figure, in one embodiment, the recognition system for the characteristic behaviors of a target object further includes: a monitoring device 31, which is disposed on the body of the target object and is configured to monitor and obtain the data information of the periodic characteristic behaviors of the target object and send it to the control device 34.

[0089] The specific type of the monitoring device 31 is not limited. According to different working principles of the monitoring device 31, the monitoring device may include a three-axis acceleration sensor or an acoustic wave sensor. For example, it can be an Internet of Things wearable device such as a smart hoof ring, a smart earring, or a smart collar, etc.

[0090] Please refer to Figure 3 As shown in the figure, in one embodiment, the recognition system for the characteristic behaviors of a target object further includes: a video surveillance device 32, which is disposed in the area where the target object is located and is connected to the control device 34. The video surveillance device 32 can be a network camera, and the number can be one or more.

[0091] Please refer to Figure 3 As shown in the figure, in one embodiment, the recognition system for the characteristic behaviors of a target object further includes: an output device 33, which is configured to receive the reference information indicating that the target object exhibits the target characteristic behavior output by the control device 34 and output the reference information indicating that the target object exhibits the target characteristic behavior in a media form. The media form output by the output device 33 is not limited and may include at least one of text, sound, and image. The output device can be, for example, a mobile terminal such as a mobile phone or a tablet computer, or a display screen or an audio device, etc.

[0092] As Figure 4 As shown in the figure, some embodiments of the present disclosure also provide a recognition device for the characteristic behaviors of a target object, including:

[0093] A first acquisition unit 41, configured to acquire data information of the periodic characteristic behaviors of a target object;

[0094] A triggering unit 42, configured to trigger the video monitoring device in the area where the target object is located to turn on when the data information of the periodic characteristic behaviors meets the video monitoring triggering condition;

[0095] A second acquisition unit 44, configured to acquire video monitoring information of the video monitoring device;

[0096] A determination unit 44, configured to determine the probability that the target object has a target characteristic behavior according to the video monitoring information;

[0097] An output unit 45, configured to output reference information that the target object has a target characteristic behavior when the probability that the target object has a target characteristic behavior is not less than a probability threshold.

[0098] Similarly, by using the recognition device of this embodiment of the present disclosure, the accuracy and real-time performance of recognizing the target characteristic behaviors of the target object can be improved, so as to timely remind the breeding personnel or scientific research personnel to take certain measures, and achieve the purpose of improving the breeding yield or scientific research.

[0099] As Figure 5 shown, some embodiments of the present disclosure further provide a recognition device for the characteristic behaviors of a target object, including: a memory 51 and a processor 52 coupled to the memory 51, and the processor 52 is configured to execute the recognition method for the characteristic behaviors of the target object according to any one of the foregoing embodiments based on the instructions stored in the memory 51.

[0100] It should be understood that each step in the foregoing recognition method for the characteristic behaviors of the target object can be implemented by the processor, and can be implemented in any one of the ways of software, hardware, firmware or a combination thereof.

[0101] In addition to the foregoing recognition method and device for the characteristic behaviors of the target object, some embodiments of the present disclosure can also be in the form of a computer program product implemented on one or more non-volatile storage media containing computer program instructions. Therefore, some embodiments of the present disclosure further provide a computer-readable storage medium, on which a computer program is stored, and when the program is executed by the processor, it implements the recognition method for the characteristic behaviors of the target object according to any one of the foregoing technical solutions.

[0102] Figure 6 The schematic diagram of a computer system according to some embodiments of the present disclosure is shown.

[0103] As Figure 6As shown, the computer system can be embodied in the form of a general-purpose computing device, which can be used to implement the method for identifying the characteristic behaviors of the target object in the above embodiments. The computer system includes a memory 61, a processor 62, and a bus 60 that connects different system components.

[0104] The memory 61 may include, for example, a system memory, a non-volatile storage medium, etc. The system memory stores, for example, an operating system, application programs, a boot loader, and other programs. The system memory may include a volatile storage medium, such as a random access memory (RAM) and / or a cache memory. The non-volatile storage medium stores, for example, the instructions for implementing the corresponding embodiments of the display method. The non-volatile storage medium includes, but is not limited to, a disk memory, an optical memory, a flash memory, etc.

[0105] The processor 62 can be implemented in the form of a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor discrete hardware components. Correspondingly, each module, such as a judgment module and a determination module, can be implemented by a central processing unit (CPU) running the instructions for executing the corresponding steps in the memory, or by a dedicated circuit for executing the corresponding steps.

[0106] The bus 60 can use any bus structure among a variety of bus structures. For example, the bus structure includes, but is not limited to, an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus.

[0107] The computer system may further include an input / output interface 63, a network interface 64, a storage interface 65, etc. The input / output interface 63, the network interface 64, the storage interface 65, the memory 61, and the processor 62 can be connected through the bus 60. The input / output interface 63 can provide a connection interface for input / output devices such as a display, a mouse, and a keyboard. The network interface 64 provides a connection interface for various networking devices. The storage interface 65 provides a connection interface for external storage devices such as a floppy disk, a USB flash drive, and an SD card.

[0108] So far, various embodiments of the present disclosure have been described in detail. To avoid obscuring the concept of the present disclosure, some details known in the art have not been described. Those skilled in the art can fully understand how to implement the technical solutions disclosed herein based on the above description.

[0109] Although some specific embodiments of the present disclosure have been described in detail by way of examples, those skilled in the art should understand that the above examples are for illustrative purposes only and not for limiting the scope of the present disclosure. Those skilled in the art should understand that the above embodiments can be modified or equivalent substitutions can be made for some technical features without departing from the scope and spirit of the present disclosure. The scope of the present disclosure is defined by the appended claims.

Claims

1. A method for identifying characteristic behaviors of a target object, comprising: Obtaining data information of the periodic characteristic behaviors of the target object, where the data information of the periodic characteristic behaviors comes from a monitoring device disposed on the body of the target object; When the data information of the periodic characteristic behaviors satisfies the video monitoring trigger condition, triggering the video monitoring device in the area where the target object is located to turn on; Obtaining the video monitoring information of the video monitoring device; Determining the probability that the target object exhibits a target characteristic behavior according to the video monitoring information; When the probability that the target object exhibits a target characteristic behavior is not less than the probability threshold, outputting reference information indicating that the target object exhibits a target characteristic behavior.

2. The recognition method according to claim 1, wherein The obtaining of the video monitoring information of the video monitoring device includes: Every time a set time period elapses, obtaining the video monitoring information collected by the video monitoring device within the set time period.

3. The recognition method according to claim 2, wherein, The set time period is 1 - 60 minutes.

4. The identification method according to claim 1, further comprising: When the probability that the target object exhibits a target characteristic behavior is less than the probability threshold and the continuous on - time of the video monitoring device is greater than the monitoring time threshold, turning off the video monitoring device; When the probability that the target object exhibits a target characteristic behavior is less than the probability threshold and the continuous on - time of the video monitoring device is not greater than the monitoring time threshold, returning to the step of obtaining the video monitoring information of the video monitoring device.

5. The recognition method according to claim 1, wherein, The determining of the probability that the target object exhibits a target characteristic behavior according to the video monitoring information includes: Determining the probability that the target object exhibits a target characteristic behavior according to the video monitoring information and a standard model of the target object exhibiting a target characteristic behavior; Wherein, the standard model of the target object exhibiting a target characteristic behavior is obtained based on a deep learning algorithm.

6. The identification method according to any one of claims 1 - 5, wherein: The target object includes livestock in a ranch or research animals; The data information of the periodic characteristic behaviors of the target object includes step - counting information and / or rumination information; The target characteristic behavior includes estrus behavior, sleep behavior or foraging behavior.

7. An identification device for characteristic behaviors of a target object, comprising: A first obtaining unit, configured to obtain data information of the periodic characteristic behaviors of the target object, where the data information of the periodic characteristic behaviors comes from a monitoring device disposed on the body of the target object; A triggering unit, configured to trigger the video monitoring device in the area where the target object is located to turn on when the data information of the periodic characteristic behaviors satisfies the video monitoring trigger condition; A second obtaining unit, configured to obtain the video monitoring information of the video monitoring device; A determining unit, configured to determine the probability that the target object exhibits a target characteristic behavior according to the video monitoring information; An output unit, configured to output reference information indicating that the target object exhibits a target characteristic behavior when the probability that the target object exhibits a target characteristic behavior is not less than the probability threshold.

8. An identification system for characteristic behaviors of a target object, comprising: A monitoring device, disposed on the body of the target object, configured to monitor and obtain data information of the periodic characteristic behaviors of the target object and send it to a control device; A control device is configured to trigger the activation of a video surveillance device in the area where the target object is located when the data information of the periodic characteristic behavior meets the video surveillance trigger condition; determine the probability of the target object performing a target characteristic behavior based on the video surveillance information; and output reference information indicating that the target object has performed the target characteristic behavior when the probability of the target object performing the target characteristic behavior is not less than a probability threshold.

9. The recognition system according to claim 8, wherein, The monitoring device includes a triaxial acceleration sensor or an acoustic wave sensor.

10. The recognition system according to any one of claims 8-9, further comprising: A video surveillance device, which is installed in the area where the target object is located and is connected to the control device.

11. The recognition system according to claim 10, further comprising: An output device, configured to receive the reference information indicating that the target object has performed the target characteristic behavior output by the control device and output the reference information in a media form.

12. An identification device for target object characteristic behaviors, comprising: A memory; and A processor coupled to the memory, the processor being configured to execute the method for identifying target object characteristic behaviors according to any one of claims 1-6 based on instructions stored in the memory.

13. A computer-readable storage medium, having stored thereon a computer program, which when executed by a processor implements the method for identifying target object characteristic behaviors according to any one of claims 1-6.

Images