Method for preventing pet from turning over garbage can, related equipment and program product
By combining multimodal sensors and environmental sensing units on pet collars, the system identifies pets rummaging through trash cans and outputs measures to stop them, thus solving the problem of pets rummaging through trash cans and improving environmental hygiene and pet health.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- HEFEI IFLYTEK TOYCLOUD TECH
- Filing Date
- 2026-03-31
- Publication Date
- 2026-05-05
AI Technical Summary
Existing pet collars have limited functionality and cannot effectively prevent pets from rummaging through trash cans, leading to environmental hygiene problems and health risks for pets.
By collecting behavioral data from multimodal sensors on pet collars, it can identify whether pets are rummaging through or rummaging through trash cans. Combined with environmental information collected by environmental sensing units, it can confirm whether there is trash in the vicinity and output measures to stop the pet from rummaging through trash cans.
It significantly reduces the false alarm rate, avoids invalid interference, reduces device power consumption, and ensures environmental cleanliness and pet health.
Smart Images
Figure CN121970694A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of smart pet devices, and more specifically, to a method, related equipment, and program product for preventing pets from rummaging through trash cans. Background Technology
[0002] Pets are animals kept by people for emotional or psychological reasons, rather than for economic purposes. A pet collar is a specialized device used to protect, control, and leash a pet.
[0003] With the development of technology, some pet collars are also used to identify pets. In modern cities, some pet owners buy pet collars (pet tags) to help them locate their pets and to allow others to contact the owner if the pet gets lost. Some cities also require pet owners to purchase pet collars for easier pet management.
[0004] Existing pet collars are relatively simple in design, often only displaying partial information about the pet and its owner, such as name, address, and contact information. These collars are too limited in function and offer very little help to their owners. Some pets, like dogs, like to rummage through trash cans. This not only scatters trash around the house, making cleaning more difficult, but pets may also find bones from kitchen waste, such as chicken bones. Eating these bones can easily lead to health problems. Summary of the Invention
[0005] In view of the above problems, this application is made to provide a method, related equipment, and computer program product for preventing pets from rummaging through trash cans, so as to promptly detect the phenomenon of pets rummaging through trash cans and take appropriate measures to stop it, thereby ensuring environmental cleanliness and pet health. The specific solution is as follows:
[0006] Firstly, a method for preventing pets from rummaging through trash cans is provided, applied to a pet collar, and the method includes:
[0007] Obtain behavioral representation data of pets;
[0008] Based on the pet's behavioral representation data, identify whether the pet exhibits a first behavior and obtain a first identification result. The first behavior refers to the pet's pawing behavior or rummaging through the trash can.
[0009] If the first recognition result indicates that the pet exhibits the first behavior, environmental information around the pet is collected through the environmental sensing unit on the pet collar.
[0010] Based on the environmental information, it is determined whether there is trash around the pet. If so, measures are output to stop the pet from rummaging through the trash.
[0011] In one possible design, in another implementation of the first aspect of the embodiments of this application, the process of identifying whether the pet exhibits a first behavior based on the pet's behavioral characterization data and obtaining a first identification result includes:
[0012] The pet's behavioral representation data is fed into a pre-trained behavior classification model to obtain the first recognition result output by the model. The first recognition result indicates whether the pet exhibits the first behavior.
[0013] The behavior classification model is trained using training data of pet behavior representations labeled with behavior tags.
[0014] In one possible design, in another implementation of the first aspect of the embodiments of this application, the process of identifying whether the pet exhibits a first behavior based on the pet's behavioral characterization data and obtaining a first identification result includes:
[0015] The pet's behavioral representation data is matched with reference behavioral data to obtain a matching result. The reference behavioral data are feature parameters that characterize the pet's first behavior.
[0016] Based on the matching results, a first identification result is obtained to determine whether the pet exhibits the first behavior.
[0017] In one possible design, in another implementation of the first aspect of the embodiments of this application, when the first behavior represents the pet's pawing behavior, the reference behavior data includes a first action, which represents the pet's limb pawing action.
[0018] When the first behavior represents the pet rummaging through the trash can, the reference behavior data includes a sequence of behaviors consisting of alternating first and second actions, where the second action represents the pet's sniffing action.
[0019] In one possible design, in another implementation of the first aspect of the embodiments of this application, the process of identifying whether the pet exhibits a first behavior based on the pet's behavioral characterization data and obtaining a first identification result includes:
[0020] The configured large model is invoked to instruct the large model to identify whether the pet exhibits a first behavior based on the pet's behavioral representation data, and the first identification result output by the large model is obtained.
[0021] In one possible design, in another implementation of the first aspect of this application, where the environmental information includes environmental images and environmental odors, the process of identifying whether there is litter around a pet based on the environmental information includes:
[0022] The image recognition algorithm is used to detect whether there is garbage in the environmental image and to determine whether the environmental odor matches the set garbage odor characteristics.
[0023] If trash is detected in the environmental image and the environmental odor matches the set characteristics of rotting trash odor, then it is determined that trash exists around the pet.
[0024] In one possible design, in another implementation of the first aspect of the embodiments of this application, the process of collecting environmental information around the pet through the environmental sensing unit on the pet collar when the first identification result indicates that the pet is exhibiting a first behavior includes:
[0025] If the first recognition result indicates that the pet has engaged in the first behavior, the environmental sensing unit on the pet collar is activated to collect environmental information about the pet's surroundings.
[0026] In one possible design, another implementation of the first aspect of the embodiments of this application further includes:
[0027] In addition to providing measures to stop the activity, a warning message is sent to the user terminal associated with the pet collar to remind the user that the pet is currently rummaging through the trash can.
[0028] Secondly, a pet collar is provided, including the collar body;
[0029] The collar body is equipped with multimodal sensors for collecting behavioral representation data of the pet;
[0030] The collar body is also equipped with a processor and an output module. The processor is used to identify whether the pet has a first behavior based on the pet's behavioral representation data and obtain a first identification result. The first behavior refers to the pet's pawing behavior or rummaging through the trash can. If the first identification result indicates that the pet has the first behavior, the processor acquires environmental information around the pet collected by the environmental sensing unit on the pet collar, identifies whether there is trash around the pet based on the environmental information, and if so, controls the output module to output a deterrent measure to stop the pet from rummaging through the trash.
[0031] Thirdly, a readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the method for preventing pets from rummaging through trash cans as described in any of the first aspects of this application.
[0032] Fourthly, a computer program product is provided, comprising a computer program that, when executed by a processor, implements the steps of the method for preventing pets from rummaging through trash cans as described in any of the first aspects of this application.
[0033] By employing the above technical solution, this application combines pet behavior with environmental information to detect pets rummaging through trash cans. By acquiring behavioral representation data of the pet through a pet collar, it can automatically identify whether the pet exhibits a primary behavior, obtaining a first identification result based on the pet's behavior. This primary behavior can be either pawing or rummaging through trash cans. Furthermore, to reduce the false alarm rate of solely behavior-based detection, if the first identification result indicates the presence of the primary behavior, environmental information such as environmental images and odors is collected from the pet collar's environmental sensing unit. Based on this environmental information, it identifies whether there is trash around the pet. If so, it can be confirmed that the probability of the pet currently rummaging through trash cans is high, and thus, intervention measures can be output. This embodiment first detects the pet's primary behavior (pawing or rummaging through trash cans) based on its behavior. If the detection result indicates the presence of the primary behavior, then environmental information is used to confirm whether there is trash around the pet, thereby determining whether intervention measures need to be output. This strategy can significantly reduce the false alarm rate and avoid invalid interference, such as a pet's actions while playing with a ball or looking for toys being mistakenly identified as rummaging through trash cans. In addition, environmental information analysis usually consumes more computing resources than behavior detection. This application first performs lightweight behavior detection and only calls high-cost environmental information analysis after the behavior is triggered, which can significantly reduce device power consumption.
[0034] Furthermore, in some real-world scenarios, the trash can may be empty. Therefore, even if behavior detection determines that the pet is currently rummaging through the trash can, but no trash is found after environmental information analysis, there is no need to take any measures to stop it, thus avoiding unnecessary triggering of such measures. Attached Figure Description
[0035] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of this application. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:
[0036] Figure 1 A schematic diagram of an implementation system architecture for a method to prevent pets from rummaging through trash cans, provided in an embodiment of this application;
[0037] Figure 2 This is a schematic diagram of a pet collar structure provided in an embodiment of this application;
[0038] Figure 3 This is a flowchart illustrating a method for preventing pets from rummaging through trash cans, as provided in an embodiment of this application. Detailed Implementation
[0039] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.
[0040] This application provides a method for preventing pets from rummaging through trash cans, which can be applied to, for example... Figure 1 The system architecture shown may include a pet collar 100 and a server 200. The server 200 may include one or more servers (…). Figure 1 (This example uses a server as an illustration).
[0041] The pet collar 100 can be used alone to execute the method for preventing pets from rummaging through trash cans provided in this application embodiment. Alternatively, the pet collar 100 and the server 200 can also be used collaboratively to execute the method for preventing pets from rummaging through trash cans provided in this application embodiment.
[0042] Reference Figure 2 As shown, the pet collar 100 of this application includes a collar body 101, and a multimodal sensor 110, a processor 111 and an output module 112 disposed on the collar body 101.
[0043] Multimodal sensors 110 include, but are not limited to: motion sensors (e.g., six-axis sensors), environmental sensing units (e.g., cameras, LiDAR, odor sensors), etc.
[0044] The motion sensor in the multimodal sensor 110 is used to collect behavioral representation data of the pet.
[0045] The processor 111 communicates with the multimodal sensor 110 to identify whether the pet exhibits a first behavior based on the collected pet behavior representation data, and obtains a first identification result. The first behavior indicates that the pet is pawing or rummaging through the trash can. If the first identification result indicates that the pet exhibits the first behavior, the processor obtains environmental information about the pet's surroundings collected by the environmental sensing unit on the pet's collar, and identifies whether there is trash around the pet based on the environmental information. If there is trash, the processor output module 112 outputs a deterrent measure to stop the pet from rummaging through the trash can.
[0046] The output module 112 includes, but is not limited to: a horn, a physical constraint structure, a gas injection device, etc.
[0047] The loudspeaker can be used to play a deterrent audio message, which can be a pre-recorded message from the pet owner to discourage the pet from rummaging through the trash can.
[0048] Physical restraints can be achieved through structural means such as vibration or electrical stimulation. It should be noted that if such measures would cause harm to the pet, they must be authorized by the pet owner and permitted by relevant laws and regulations before implementation.
[0049] Gas spray devices can be used to spray harmless gases such as bittering agents to discourage pets from rummaging through trash cans.
[0050] In some other possible implementations, the processor 111 can also send the collected pet behavior data to the server 200 via a wireless communication module. The server 200, based on the pet's behavior data, identifies whether the pet exhibits a first behavior and returns the first identification result to the pet collar 100. If the pet collar determines that the pet has exhibited the first behavior based on the first identification result, it can collect surrounding environmental information through an environmental sensing unit and send the environmental information to the server 200. The server 200 identifies whether there is litter around the pet based on the environmental information and returns the identification result to the pet collar 100. If the pet collar 100 determines that there is litter in the environment based on the identification result, its control output module outputs a deterrent measure.
[0051] In addition, once the pet collar 100 confirms that the pet has engaged in the first behavior and identifies trash in the environmental information, it can also send a warning message to the user terminal associated with the pet collar to remind the user that the pet is currently rummaging through the trash can.
[0052] The pet collar provided in this embodiment first detects the pet's primary behavior (pushing or rummaging through trash cans). If the detection result indicates the presence of this primary behavior, it then uses environmental information to confirm whether there is trash around the pet, thereby determining whether intervention measures should be taken. This strategy can significantly reduce the false alarm rate and avoid invalid interference, such as misinterpreting a pet's actions while playing with a ball or looking for a toy as rummaging through a trash can. Furthermore, environmental information analysis typically consumes more computational resources than behavior detection. This application performs lightweight behavior detection first, only invoking high-cost environmental information analysis after a trigger, which can significantly reduce device power consumption.
[0053] Furthermore, in some real-world scenarios, trash cans may be empty. Therefore, even if behavior detection determines that a pet is currently rummaging through the trash can, but no trash is found after environmental information analysis, there is no need to take any measures to stop it, thus avoiding useless interventions.
[0054] This application provides a method for preventing pets from rummaging through trash cans. Taking the application of this method to a computer device as an example, the computer device can specifically be... Figure 1 The system consists of pet collar 100 or pet collar 100 and server 200. (Refer to...) Figure 3 The specific steps to stop pets from rummaging through trash cans are as follows:
[0055] Step S100: Obtain the pet's behavioral representation data.
[0056] Among them, the pet's behavioral representation data can reflect the pet's current actions and behaviors, which can help identify whether the pet has exhibited a first behavior, such as pawing or rummaging through the trash can.
[0057] For example, when a pet is pawing, the acceleration and frequency of its forelimbs along the Z-axis will exhibit specific parameters. Similarly, when a pet is rummaging through a trash can, its head pitch angle and forelimb movements will display specific postures. By monitoring these behavioral data, we can analyze whether the pet exhibits the primary behavior.
[0058] Among these, behavioral representation data can be behavioral sequences, such as a sequence of various pet behaviors over a period of time. For example, if a pet exhibits a behavioral sequence like: front paw scratching - remaining still for a moment - lowering its head to sniff - scratching again - ..., it indicates a high probability that the pet is rummaging through the trash can.
[0059] Step S110: Based on the pet's behavioral representation data, identify whether the pet exhibits a first behavior and obtain the first identification result.
[0060] The first behavior can be the act of shoveling or pulling.
[0061] Pet scratching behavior refers to the act of using one's limbs to grab or pry at objects, such as a pet scratching at a door frame, a wooden stake, or a trash can.
[0062] Since pets inevitably engage in rummaging through trash cans, this step uses behavioral data to identify instances of this rummaging behavior and recalls scenarios where such actions occurred. Subsequent steps further confirm whether the pet was indeed rummaging through trash cans using environmental information.
[0063] Furthermore, the first behavior could also be rummaging through the trash. It's understandable that pets might paw at the trash while doing so, but pawing doesn't necessarily mean they're rummaging through it. Therefore, this step can set stricter filtering criteria than just pawing, to identify whether the pet is currently rummaging through the trash based on its behavioral data. After determining that the pet is rummaging through the trash based on its behavioral data, subsequent steps can be used to confirm the behavior using environmental information, avoiding misjudgments based solely on the initial identification result obtained from the pet's behavioral data.
[0064] This step, which uses pet behavioral representation data to identify whether a pet exhibits a primary behavior, can be achieved using a variety of different methods.
[0065] One alternative implementation is a rule-based approach, as shown in the following example:
[0066] The pet's behavioral representation data is matched with reference behavioral data to obtain the matching results. The reference behavioral data consists of feature parameters that characterize the pet's primary behavior.
[0067] Based on the matching results, the first identification result is obtained to determine whether the pet exhibits the first behavior.
[0068] In one possible implementation, corresponding reference behavior data can be designed by combining the characteristics of the pet's first behavior (pawing or rummaging through the trash). For example, when the first behavior represents the pet's pawing behavior, the reference behavior data includes the first action, which represents the pet's limb scratching motion.
[0069] For example, when the first behavior represents a pet rummaging through a trash can, the reference behavior data includes a sequence of behaviors that alternate between the first action and the second action, where the second action represents the pet's sniffing action.
[0070] Table 1 below illustrates several examples of rules for determining whether a pet is rummaging through the trash can. Of course, rules corresponding to the pet's rummaging behavior can also be designed based on the characteristics of that behavior; these will not be shown here.
[0071] Table 1
[0072]
[0073] It should be noted that Table 1 above only uses a pet dog as an example to illustrate one possible judgment rule, and the specific values in the trigger conditions are only an optional example. This application can pre-design rules to determine whether a pet is rummaging through trash cans, tailored to different types of pets, and can collect feedback data during actual application to personalize and adjust the rules to suit the current pet.
[0074] Another alternative implementation involves using a pre-trained neural network model to predict whether a pet will exhibit its first behavior. For example:
[0075] The pet's behavioral representation data is fed into a pre-trained behavior classification model to obtain the first recognition result output by the model. The first recognition result indicates whether the pet exhibits the first behavior.
[0076] The behavior classification model is trained using training data of pet behavior representations labeled with behavior tags.
[0077] Among them, the behavior classification model can adopt a network structure of encoder and classification head.
[0078] The encoder is used to input behavioral representation data, and after encoding, it yields a behavioral feature vector.
[0079] The classification head is used to output a binary classification result based on the behavior feature vector: presence of the first behavior and absence of the first behavior.
[0080] The behavior classification model can use the cross-entropy loss function during training.
[0081] Another alternative implementation involves using a large model to predict whether a pet will exhibit its first behavior. For example:
[0082] The configured large model is invoked to instruct the large model to identify whether the pet exhibits a first behavior based on the pet's behavioral representation data, and the first identification result output by the large model is obtained.
[0083] This embodiment can accurately identify whether a pet exhibits first behavior by leveraging the powerful knowledge background and semantic understanding capabilities of a large model.
[0084] Step S120: If the first recognition result indicates that the pet has a first behavior, collect environmental information around the pet through the environmental sensing unit on the pet collar.
[0085] Specifically, in the aforementioned steps, when identifying the pet's first behavior based on behavioral representation data, in order to further confirm whether the pet is indeed rummaging through the trash can, environmental information around the pet is collected through the environmental sensing unit on the pet collar.
[0086] In one possible implementation, this step could involve activating the environmental sensing unit on the pet's collar when the first recognition result indicates that the pet has engaged in the first behavior, thereby collecting environmental information about the pet's surroundings. That is, the environmental sensing unit can normally be off, and is only activated to capture environmental information when the first recognition result indicates that the pet has engaged in the first behavior, thus reducing system power consumption.
[0087] Step S130: Identify whether there is garbage around the pet based on environmental information. If so, output measures to stop the pet from rummaging through the garbage.
[0088] Environmental information includes, but is not limited to, environmental images and odors. This information can be used to identify whether litter is present around the pet. If litter is found, it indicates a high probability that the pet is currently rummaging through the trash, at which point appropriate measures will be taken to stop it.
[0089] If no trash is identified around the pet in this step:
[0090] 1) If the first behavior is the pet pawing, it means that the first identification result may be a false alarm, that is, the pet did not actually paw, and it is even less likely to be rummaging through the trash can; another possibility is that the first identification result is accurate, but the pet's pawing behavior is not rummaging through the trash can, but rather pawing at the door frame, playing with a ball, or other types of pawing behavior.
[0091] 2) If the first behavior is rummaging through the trash can, it indicates that the first identification result may be a false alarm, meaning the pet is not actually rummaging through the trash can, and therefore trash cannot be detected in the environmental information. Alternatively, the first identification result may be accurate, meaning that although the pet is indeed rummaging through the trash can, the trash can may be empty at the moment. In this case, it is also possible not to stop it.
[0092] In one possible implementation, taking environmental information including environmental images as an example, the process of identifying whether there is litter around a pet based on environmental information includes:
[0093] Image recognition algorithms are used to detect whether litter exists in environmental images. For example, an environmental image is fed into a pre-trained litter recognition model, and the model outputs a recognition result indicating whether litter exists in the input image. The litter recognition model can be pre-trained using training images.
[0094] In another possible implementation, taking environmental information including ambient odor as an example, the process of identifying whether there is litter around a pet based on environmental information includes:
[0095] Determine if the ambient odor matches a set characteristic of decaying garbage odor. For example, check if the concentration of decaying characteristic gases in the ambient odor reaches a set concentration value. If it does, the ambient odor is determined to match the characteristic of decaying garbage odor, indicating that there is garbage around the pet.
[0096] In another possible implementation, taking environmental information including environmental images and odors as an example, the process of identifying whether there is litter around a pet based on environmental information includes:
[0097] Image recognition algorithms are used to detect whether there is garbage in environmental images and to determine whether the environmental odor matches the set characteristics of garbage rotten odor.
[0098] If trash is detected in the environmental image and the environmental odor matches the set characteristics of rotting trash odor, then it is determined that there is trash around the pet.
[0099] By simultaneously considering whether there is litter in the environmental image and whether the environmental odor matches the characteristics of rotting litter, misidentification of single-modal data can be avoided, thus improving the accuracy of litter identification results.
[0100] In this embodiment, there are multiple ways in which the pet collar can output a restraining measure, as detailed in the previous descriptions, which will not be repeated here.
[0101] In one alternative implementation, in addition to outputting deterrent measures, a warning message can also be sent to the user terminal associated with the pet collar to remind the user that the pet is currently rummaging through the trash can.
[0102] The method for preventing pets from rummaging through trash cans provided in this embodiment first detects the pet's primary behavior (pushing or rummaging through trash cans). If the detection result indicates the presence of this primary behavior, environmental information is then used to confirm whether there is trash around the pet, thereby determining whether intervention measures should be taken. This strategy can significantly reduce the false alarm rate and avoid invalid interference, such as misinterpreting a pet's actions while playing with a ball or looking for a toy as rummaging through a trash can. Furthermore, environmental information analysis typically consumes more computational resources than behavior detection. This application performs lightweight behavior detection first, only invoking high-cost environmental information analysis after a trigger, which can significantly reduce device power consumption.
[0103] Furthermore, in some real-world scenarios, trash cans may be empty. Therefore, even if behavior detection determines that a pet is currently rummaging through the trash can, but no trash is found after environmental information analysis, there is no need to take any measures to stop it, thus avoiding useless interventions.
[0104] This application also provides a computer program product including computer-readable instructions, which, when executed on an electronic device, cause the electronic device to perform the steps of any of the methods provided in this application for preventing pets from rummaging through trash cans.
[0105] This application also provides a computer-readable storage medium carrying one or more computer programs. When the one or more computer programs are executed by an electronic device, the electronic device can implement the steps of any of the methods for preventing pets from rummaging through trash cans provided in this application.
[0106] It should also be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. In addition, in the device embodiment drawings provided in this application, the connection relationship between modules indicates that they have a communication connection, which can be implemented as one or more communication buses or signal lines.
[0107] Through the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary general-purpose hardware, or it can be implemented by special-purpose hardware including application-specific integrated circuits, special-purpose CPUs, special-purpose memory, special-purpose components, etc. Generally, any function performed by a computer program can be easily implemented by corresponding hardware, and the specific hardware structure used to implement the same function can also be diverse, such as analog circuits, digital circuits, or special-purpose circuits. However, for this application, software program implementation is more often the preferred implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a readable storage medium, such as a computer floppy disk, USB flash drive, mobile hard disk, ROM, RAM, magnetic disk, or optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, training equipment, or network device, etc.) to execute the methods described in the various embodiments of this application.
[0108] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product.
[0109] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, training device, or data center to another website, computer, training device, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a training device or data center that integrates one or more available media. The available media may be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., DVDs), or semiconductor media (e.g., solid-state drives (SSDs)).
[0110] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can be referred to each other.
Claims
1. A method for preventing pets from rummaging through trash cans, characterized in that, Methods for applying to pet collars include: Obtain behavioral representation data of pets; Based on the pet's behavioral representation data, identify whether the pet exhibits a first behavior and obtain a first identification result. The first behavior refers to the pet's pawing behavior or rummaging through the trash can. If the first recognition result indicates that the pet exhibits the first behavior, environmental information around the pet is collected through the environmental sensing unit on the pet collar. Based on the environmental information, it is determined whether there is trash around the pet. If so, measures are output to stop the pet from rummaging through the trash.
2. The method according to claim 1, characterized in that, The process of identifying whether a pet exhibits a first behavior based on the pet's behavioral representation data and obtaining a first identification result includes: The pet's behavioral representation data is fed into a pre-trained behavior classification model to obtain the first recognition result output by the model. The first recognition result indicates whether the pet exhibits the first behavior. The behavior classification model is trained using training data of pet behavior representations labeled with behavior tags.
3. The method according to claim 1, characterized in that, The process of identifying whether a pet exhibits a first behavior based on the pet's behavioral representation data and obtaining a first identification result includes: The pet's behavioral representation data is matched with reference behavioral data to obtain a matching result. The reference behavioral data are feature parameters that characterize the pet's first behavior. Based on the matching results, a first identification result is obtained to determine whether the pet exhibits the first behavior.
4. The method according to claim 3, characterized in that, When the first behavior represents the pet's pawing behavior, the reference behavior data includes a first action, which represents the pet's limbs scratching or pawing. When the first behavior represents the pet rummaging through the trash can, the reference behavior data includes a sequence of behaviors that alternate between the first action and the second action, where the second action represents the pet's sniffing action.
5. The method according to claim 1, characterized in that, The process of identifying whether a pet exhibits a first behavior based on the pet's behavioral representation data and obtaining a first identification result includes: The configured large model is invoked to instruct the large model to identify whether the pet exhibits a first behavior based on the pet's behavioral representation data, and the first identification result output by the large model is obtained.
6. The method according to claim 1, characterized in that, The environmental information includes environmental images and environmental odors. The process of identifying whether there is litter around a pet based on this environmental information includes: The image recognition algorithm is used to detect whether there is garbage in the environmental image and to determine whether the environmental odor matches the set garbage odor characteristics. If trash is detected in the environmental image and the environmental odor matches the set characteristics of rotting trash odor, then it is determined that trash exists around the pet.
7. The method according to claim 1, characterized in that, When the first recognition result indicates that the pet exhibits a first behavior, the process of collecting environmental information around the pet through the environmental sensing unit on the pet collar includes: If the first recognition result indicates that the pet has engaged in the first behavior, the environmental sensing unit on the pet collar is activated to collect environmental information about the pet's surroundings.
8. The method according to any one of claims 1-7, characterized in that, Also includes: In addition to providing measures to stop the activity, a warning message is sent to the user terminal associated with the pet collar to remind the user that the pet is currently rummaging through the trash can.
9. A pet collar, characterized in that, Including the collar itself; The collar body is equipped with multimodal sensors for collecting behavioral representation data of the pet; The collar body is also equipped with a processor and an output module. The processor is used to identify whether the pet has a first behavior based on the pet's behavioral representation data and obtain a first identification result. The first behavior refers to the pet's pawing behavior or rummaging through the trash can. If the first identification result indicates that the pet has the first behavior, the processor acquires environmental information around the pet collected by the environmental sensing unit on the pet collar, identifies whether there is trash around the pet based on the environmental information, and if so, controls the output module to output a deterrent measure to stop the pet from rummaging through the trash.
10. A readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements each step of the method for preventing pets from rummaging through the trash can as described in any one of claims 1 to 8.
11. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the method for preventing pets from rummaging through the trash can as described in any one of claims 1 to 8.