Pet online interaction system operation method

Abstract

The invention discloses a pet online interaction system operation method. The method comprises the following steps: acquiring real-time audio data and video data; a smell sensor is used for collecting smell of a pet, collected smell information is converted into an electric signal, and the electric signal is transmitted to a remote server through a network; according to the collected real-time audio data, video data and smell, monitoring whether the pet meets an online interaction condition or not; when it is monitored that the pets meet the online interaction conditions, a preset pet recommendation method is used for determining a first interaction pet from pets which are currently online and are in an idle state, and the target pet interacts with the interaction pet through an interaction box; and establishing a communication link between the target pet and the interactive pet. By remotely transmitting the smell of the pet, a pet owner can more deeply understand the character and characteristics of the interactive pet and select a more suitable playing partner or partner for a target pet, and the interactive box is a relatively closed environment, so that the target pet is helped to concentrate on interaction with the interactive pet, and external interference is avoided.

Description

Technical Field

[0001] The present invention relates to the field of remote interaction technology, and in particular to a pet online interaction system operation method. Background Art

[0002] As living standards improve, more and more people keep pets. Since many owners need to go out to work, pets have to stay alone at home for a long time. For pets living in urban apartments, it is difficult for them to even have the opportunity to leave the house without their owners. In the long run, the physical and mental health of pets may be affected. Based on the rapid development of technology and the continuous improvement of people's care for pets, pet online interaction systems have gradually become a new way for pet owners to communicate with their pets. Traditional pet-raising methods are often limited to face-to-face interactions, while the fast pace of modern life and high work pressure make it difficult for pet owners to always be with their pets.

[0003] For example, Chinese patent announcement number CN114821815B discloses a method, device, equipment and medium for operating a pet online interaction system, including: obtaining real-time audio and video data sent by the host of the target pet's device end, and using a preset pet monitoring algorithm to monitor whether the target pet in the video screen corresponding to the real-time audio and video data meets the preset online interaction conditions; if the target pet is monitored to meet the preset online interaction conditions, a preset pet recommendation method is used to determine a first interactive pet from pets that are currently online and idle; and a first communication link is established between the device end of the target pet and the device end of the first interactive pet, so that the target pet and the first interactive pet can interact online. This method can recommend interactive friends to pets by analyzing pet sounds and movements, so that pets can interact online with each other, solve the loneliness problem of pets when they are not accompanied by their owners, and enrich the digital life of pets.

[0004] In the comparison documents, there is no collection of pet odors. Odor is one of the important ways for pets to communicate. By restoring the pet's smell, pet owners can understand the pet's characteristics and personality more intuitively, making the selection of the first interactive pet more accurate. Summary of the invention

[0005] The present application provides a method for operating a pet online interactive system. By collecting the odor information of the interactive pet, the pet owner can obtain multi-dimensional information of the interactive pet, making it easier for the pet owner to select an interactive pet for the target pet.

[0006] The present application provides a pet online interaction system operation method, comprising: S101, collecting real-time audio data and video data, obtaining real-time audio data and video data of a pet; Acquire real-time audio and video from the host of the pet device. The host of the pet device collects real-time audio and video data through the integrated audio acquisition module and video acquisition module; S102, using an odor sensor to collect odors from the pet, converting the collected odor information into electrical signals, and transmitting the electrical signals to a remote server via a network; S103, monitoring whether the pet meets the conditions for online interaction based on the collected real-time audio data, video data and smell; S104, according to step S103, it is monitored that the target pet meets the online interaction condition, a preset pet recommendation method is used to determine a first interactive pet from the interactive pets that are currently online and idle, and the target pet interacts with the interactive pet through the interaction box; the pet recommendation method constructs a weighted evaluation function according to the voice matching degree, the action matching degree, the user preference and the popularity; S105, establishing a communication link between the target pet and the first interactive pet; S106, recording in real time the interactive action information and interactive voice information between the target pet and the first interactive pet during the online interaction process.

[0007] Preferably, the monitoring target pet satisfies the conditions for online interaction. If the complete body of the target pet is within the field of view in the video screen, or more than half of the body of the target pet is within the field of view and the head is facing the local video collector on the device side, the conditions are determined to be met; otherwise, the conditions are determined to be not met; if the concentration of the target pet's odor exceeds a preset threshold, the conditions are determined to be met; if the concentration of the target pet's odor does not exceed the threshold, the conditions are determined to be not met.

[0008] Preferably, the weighted evaluation function = w1 Among them, w1, , and is the weight coefficient.

[0009] Preferably, the method for enabling the target pet to interact with the interactive pet through the interactive box is as follows: S201, setting an interaction box, wherein a main screen, an auxiliary screen, an odor collection module and an odor restoration module are provided on the interaction box, wherein the odor restoration module includes an odor generator and a release port; The main screen is located at the center of the front of the interactive box, the auxiliary screens are distributed around the main screen or on the side of the box, the odor release port is located below the auxiliary screen, the odor release port is electrically connected to the auxiliary screen, and each odor release port corresponds to an auxiliary screen; S202, placing the interactive box around the pet, the odor collection module collects the odor information of the pet, and uploads it to the remote server. After receiving the uploaded odor information, the remote server generates a recommendation list; The collected odor information is converted into electrical signals and transmitted to the remote server in real time through the network; S203, the odor restoration module restores the odor of the interactive pet through a built-in odor generator according to the received odor information of the interactive pet.

[0010] Preferably, the target pet interacts with the interactive pet through the interaction box, records the interaction data, generates a virtual pet based on the interaction data, and performs virtual interaction with the virtual pet when the interactive pet is not online.

[0011] Preferably, the steps of the target pet virtual interaction are: S301, collecting interaction data between the target pet and the interactive pet in real time, and extracting the interaction preference and interaction habit of the target pet based on the interaction data; S302, generating a virtual interactive pet according to the interactive process between the target pet and the interactive pet recorded in step S301; S303, performing virtual interaction according to the virtual interactive pet generated in step S302; S304, collecting interaction data between the target pet and the virtual interactive pet, and analyzing the preferences of the target pet through the collected interaction data; S305, based on the preferences of the target pet analyzed in step S304, the system recommends a matching interactive pet.

[0012] Preferably, the collected basic information of the target pet's preferences is converted into a numerical feature vector, the personality characteristics are converted into a numerical vector using one-hot encoding, a weight is assigned to each feature, a weighted score is calculated based on the weight assignment, and the interactive pets with weighted scores greater than a preset recommendation threshold are recommended as interactive pets.

[0013] Preferably, S401, according to the target pet interaction data collected in step S301, feature extraction is performed on the collected data; S402, identifying the state of the target pet according to the features extracted in step S401; S403, editing the video clips according to the state of the target pet identified in step S402; S404, according to the real-time status of the target pet, match the information from the virtual interaction information library to guide the target pet.

[0014] Preferably, the feature extraction includes sound features, image features and odor features. For the extraction of sound features, the system converts the sound signal from the time domain to the frequency domain through fast Fourier transform, extracts the pitch features, and extracts the volume features by calculating the peak value; for the extraction of image features, a convolutional neural network is used to train the facial image of the target pet; the posture features of the target pet's body are extracted through a key point detection algorithm, and the movement speed of the target pet in the video can be calculated using a feature point tracking algorithm; for the extraction of odor features, the type of odor is identified by using a sensor array.

[0015] Preferably, the video is composed of a series of continuous frames. When the system detects that the state of the target pet has changed, the starting time point of the change is recorded. The starting time point is the time corresponding to the frame where the state begins to change. When the state of the target pet ends or changes to another state, the ending time point of the change is recorded. The ending time point marks the end of the current state. The recorded starting and ending time points are corresponded to the frame numbers of the video. Each frame has a unique number, so that the system can locate the time point on a specific frame.

[0016] One or more technical solutions provided in this application have at least the following technical effects or advantages: By remotely transmitting the pet's scent, pet owners can have a deeper understanding of the interactive pet's personality and characteristics, and select a more suitable playmate or partner for the target pet. The interactive box is a relatively closed environment, which helps the target pet focus on interacting with the interactive pet and avoid external interference. The odor collection and restoration module equipped on the interactive box can accurately capture and restore the pet's scent, allowing pets to communicate more naturally and enhance the realism of the interaction.

[0017] In addition, the system can generate virtual interactive pet information based on the target pet's preferences and interaction quality, providing a realistic virtual interactive experience. Through a comprehensive analysis of the target pet's interaction records, the system can intelligently recommend more suitable interactive pets to improve the success rate of making friends. Even if the interactive pet is not online, the system can still provide a virtual interactive experience.

[0018] The system can not only fully record the target pet's interactive process, but also identify the different states of the target pet through data analysis, automatically edit video clips, and provide pet owners with personalized video sharing content. Through the matching and playback of virtual interactive information, the system can achieve intelligent state guidance for the target pet, allowing pet owners to view the state of the target pet, share videos, and provide state guidance anytime, anywhere through mobile devices such as mobile phones. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A schematic diagram of a process flow of a pet online interactive system operating method of the present invention; Figure 2 It is a structural schematic diagram of an interactive box according to an embodiment of the present invention; Figure 3 A schematic diagram of a flow chart of interaction between a target pet and an interactive pet according to an embodiment of the present invention; Figure 4 A schematic diagram of a process of virtual interaction with a target pet according to an embodiment of the present invention; Figure 5 This is a schematic diagram of a process for guiding the state of a target pet according to an embodiment of the present invention. DETAILED DESCRIPTION

[0020] To facilitate the understanding of the present invention, the present application will be described more comprehensively below with reference to the relevant drawings; the drawings show preferred embodiments of the present invention, but the present invention can be implemented in many different forms and is not limited to the embodiments described herein; on the contrary, the purpose of providing these embodiments is to enable a more thorough and comprehensive understanding of the disclosed content of the present invention.

[0021] It should be noted that the terms “vertical”, “horizontal”, “up”, “down”, “left”, “right” and similar expressions used in this document are only for illustrative purposes and do not represent the only implementation method.

[0022] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by technicians in the technical field to which the present invention belongs; the terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention; the term "and / or" used herein includes any and all combinations of one or more related listed items.

[0023] Embodiment 1: Figure 1 1 is a flow chart of a method for operating a pet online interactive system according to an embodiment of the present invention, comprising the following steps: S101, collecting real-time audio and video data, obtaining real-time audio and video data of the pet.

[0024] Specifically, real-time audio and video are obtained from the host of the pet device. The host of the pet device collects real-time audio and video data through the integrated audio acquisition module and video acquisition module. The audio acquisition module is composed of a high-sensitivity microphone, which can capture various sounds made by the pet, such as calls, snoring, etc. The microphone converts the captured sound signal into an electrical signal, and digitizes it through an audio codec to finally generate high-quality audio data. The generated audio data is transmitted to the host in real time for processing and analysis; the video acquisition module is composed of a high-definition camera, which can capture the pet's activity screen in real time. The camera captures the pet's image information through the lens and converts it into an electrical signal through an image sensor. The converted electrical signal is processed by the image codec to generate high-quality video data. The video data is also transmitted to the host in real time for processing and analysis, so as to identify the different states of the pet and edit personalized video clips.

[0025] S102, using an odor sensor to collect odors from the pet, converting the collected odor information into electrical signals, and transmitting them to a remote server via a network.

[0026] Furthermore, pets include target pets and interactive pets, and the odor sensor used is an electronic nose, which is a device that can simulate the olfactory function of organisms and can identify and quantify a variety of odor molecules. The electronic nose is composed of a sensor array, a signal processing unit and a data transmission module. The electronic nose is placed in the area where the pet is active to collect the pet's odor. The sensor array of the electronic nose collects odor molecules in the air, and the collected molecules react with the sensor surface to generate a measurable electrical signal. The analog electrical signal generated by the electronic nose is converted into a digital signal through an analog-to-digital converter, and the converted digital signal is encoded for network transmission and storage. The encoding process may include steps such as data compression and encryption to reduce transmission bandwidth requirements and improve data security. The electronic nose device is connected to a remote server via a network, and the encoded odor data is transmitted to the remote server via the network. The remote server receives the odor data transmitted from the electronic nose device, performs preliminary processing, and stores the processed odor data in a database on the server side.

[0027] S103, based on the collected real-time audio data, video data and odor data, monitoring whether the pet meets the conditions for online interaction.

[0028] Monitor whether the video screen and pet odor concentration corresponding to the real-time audio and video data meet the preset online interaction conditions. If the complete body of the pet in the video screen is in the field of view, or more than half of the pet's body is in the field of view and the head is facing the local video collector on the device, the conditions are determined to be met. Otherwise, the conditions are determined to be not met. If the concentration of the pet's odor exceeds the preset threshold, the conditions are determined to be met. If the concentration of the pet's odor does not exceed the threshold, the conditions are determined to be not met. The threshold is set based on historical data and the experience of professionals.

[0029] S104, according to step S103, it is monitored that the pet meets the online interaction condition, and a preset pet recommendation method is used to determine a first interactive pet from the interactive pets that are currently online and idle, and the target pet interacts with the interactive pet through the interaction box.

[0030] The system determines a set of voice words and a set of action words for the pet. When the pet interacts, the system collects the pet's voice samples in real time. The voice samples contain various sounds made by the pet in different situations and emotions. The system uses signal processing technology to extract the Mel-frequency cepstral coefficients (MFCC) in the voice samples. The MFCC is a common voice feature that can effectively represent the spectral envelope information of the voice. Based on the extracted MFCC features, each voice sample generates a corresponding voice feature vector. The system performs cluster analysis on all voice feature vectors through a preset clustering algorithm (DBSCAN), classifies similar voice feature vectors into one category, and generates voice clustering vectors. Each voice clustering vector is a language. A number of phonetic words constitute the pet's phonetic dictionary; the system collects video information in real time, and the video information contains the dynamic performance of the pet in various activities. The system extracts optical flow information and color information of each frame of the video information. The optical flow information reflects the movement trajectory and speed of the pet, while the color information helps to identify the appearance characteristics of the pet. The extracted optical flow and color information are input into a preset deep neural network. After deep neural network training, it can output a feature vector representing the pet's action characteristics. The system uses a preset clustering algorithm to perform cluster analysis on the action feature vector to generate a preset number of action clustering vectors. Each action clustering vector is an action word, and multiple action words constitute the pet's action dictionary.

[0031] Calculate the matching degree between the target pet's voice word set and the voice ability of the potential interactive pet, using the cosine similarity method to calculate. Calculate the matching degree between the target pet's action word set and the behavioral ability of the potential interactive pet. Based on the pet owner's previous preferences for pets, such as the favorite pet type, color, size, etc., the preference information is obtained by analyzing the user's historical interaction records. Based on the pet's popularity, such as the number of pet interactions, user evaluations, etc., the pet's popularity reflects the general appeal and interaction quality of the pet. Construct a weighted evaluation function based on the voice matching degree, action matching degree, user preference and popularity. The weighted evaluation function = w1 Among them, w1, , and is the weight coefficient. According to the calculated weighted evaluation function, the pet with the highest score is selected as the first interactive pet.

[0032] S105, establishing a communication link between the target pet and the first interactive pet.

[0033] The device end of the target pet initiates a connection request to the device end of the first interactive pet. After receiving the connection request, the device end of the first interactive pet verifies the validity and security of the request. After the verification is passed, the device end of the first interactive pet accepts the connection request. After the target pet device end and the first interactive pet device end confirm the connection, the system will use the network communication protocol (TCP / IP,) to establish a stable communication link.

[0034] S106, recording in real time the interactive action information and interactive voice information between the target pet and the first interactive pet during the online interaction process.

[0035] The technical solutions in the above embodiments of the present application have at least the following technical effects or advantages: By remotely transmitting the pet's scent, pet owners can obtain more information about the interactive pet and thus select the first interactive pet. By transmitting the scent of the interactive pet, pet owners can more intuitively understand the characteristics and personality of the interactive pet and can quickly select the first interactive pet. Suitable interactive pets have a positive impact on the pet's mental health and physical health. With the assistance of scent information, pet owners can more easily find suitable interactive pets for the target pets.

[0036] Embodiment 2: Based on the transmission of odor in embodiment 1, an interaction box is set in this embodiment, and an odor collection module and an odor restoration module are provided on the interaction box. The interaction box provides a physically isolated space to ensure that the pet will not be disturbed or harmed by external factors during the interaction process.

[0037] like Figure 2 and Figure 3 As shown, the method of making the target pet interact with the interactive pet through the interaction box is as follows: S201, setting an interaction box, on which a main screen, an auxiliary screen, an odor collection module and an odor restoration module are set, and the odor restoration module includes an odor generator and a release port.

[0038] Furthermore, the main screen is located at the center of the front of the interactive box, and adopts high-definition touch screen technology, which not only provides a clear visual effect, but also allows pet owners to browse and select information conveniently through touch operation. The main screen will display the basic information of the pet, such as name, age, breed, etc., and can also play videos and display pictures of the pet, helping pet owners to understand the status and living habits of the pet more comprehensively; the auxiliary screens are distributed around the main screen or on the side of the box, mainly displaying high-definition pictures of specific body parts of the pet, so that the pet owner can observe the different parts of the pet more carefully, thereby enhancing the understanding and care of the pet. Each auxiliary screen corresponds to a specific body part, such as the head, paws, body, etc., making the information display more intuitive and orderly; the odor release port is located below or beside the auxiliary screen, and the odor release port is electrically connected to the auxiliary screen. Each release port corresponds to an auxiliary screen. When the screen displays a picture of a specific part, the corresponding odor release port will release the odor of the part. The design of the odor release port not only enhances the sensory experience of the interactive box, but also enables the pet owner to feel the presence and breath of the pet more realistically.

[0039] S202, placing the interactive box near the pet, the odor collection module collects the pet's odor information and uploads it to the remote server. After receiving the uploaded odor information, the remote server generates a recommendation list.

[0040] The odor collection module has a built-in odor sensor, which is an electronic nose. The electronic nose can accurately collect the odor information of the interactive pet. The collected odor information will be converted into electrical signals and transmitted to the remote server in real time through the network.

[0041] After receiving the uploaded odor information, the remote server generates a recommendation table arranged according to the preferences of the target pets based on the pet's basic information (such as breed, age, gender, etc.) and the pet owner's friendship needs (such as looking for a playmate of the same breed, looking for a pet with a similar personality, etc.).

[0042] S203, the odor restoration module restores the odor of the interactive pet through a built-in odor generator according to the received odor information of the interactive pet.

[0043] In the interactive box, the pet owner can choose to view pictures of the tail by touching the screen. The odor restoration module will release the odor of the tail through the odor release port based on the collected odors, allowing the pet owner to further understand the characteristics of the pet, such as the softness of the hair, the health of the skin, etc.

[0044] The technical solutions in the above embodiments of the present application have at least the following technical effects or advantages: The environment in the interaction box is relatively closed, which helps the target pet to focus more on the interaction with the interactive pet without being distracted by external sounds, smells or other interference factors, and helps to more accurately capture and analyze the pet's behavior and voice information; The odor collection module and odor restoration module installed on the interactive box can accurately collect and restore the pet's odor. In a closed space, the transmission of odor is more direct and effective, which helps pets communicate more naturally through odor, thereby enhancing the realism of the interaction.

[0045] Embodiment 3: Based on the interactive box set in Embodiment 2, the interaction between the target pet and the interactive pet is made more convenient, the communication process of the pet is recorded, its favorite pet friend is evaluated, and the interaction process with the interactive pet is recorded, including sound, image, smell, and action, to form virtual friend information, so that virtual interaction can be carried out when the interactive pet is not online.

[0046] like Figure 4 As shown, the steps of the target pet virtual interaction are: S301, collecting the interaction data between the target pet and the interactive pet in real time, and extracting the interaction preference and interaction habit of the target pet based on the interaction data.

[0047] The interactive data includes data such as sound, image, smell and movement. Sound collection uses a high-sensitivity microphone array to capture the sound of the target pet in all directions, including calls, snoring, barking, etc.; image collection utilizes a high-definition camera to take a full-body photo of the target pet in real time. The camera has a night vision function to ensure that the target pet's image can be clearly captured even in a dark environment; smell collection uses an electronic nose (smell sensor) device; movement collection is through motion capture technology, which is an infrared sensor to record the movement data of the target pet in real time. The movement data should include the pet's walking, jumping, tail wagging, licking and other behaviors.

[0048] S302, generating a virtual interactive pet according to the interaction process between the target pet and the interactive pet recorded in step S301.

[0049] Based on the collected interactive information, the preferences of the target pet are analyzed, and virtual interactive objects are created using 3D modeling according to the preferences of the target pet. The virtual interactive objects are modeled using 3ds Max, and the basic framework of the virtual character is constructed using geometry. The basic model is refined using the software's sculpting tools, including the virtual pet's hair, skin texture, and eyes. The surface of the 3D model is mapped onto a two-dimensional plane for mapping, and the model is rendered using a rendering engine to generate a virtual interactive object.

[0050] S303, performing virtual interaction according to the virtual interactive pet generated in step S302.

[0051] When the interactive pet is not online, the system automatically plays the voice of the virtual friend through the interactive box according to the preset schedule or the activity status of the target pet, uses the main screen and auxiliary screen on the interactive box to display the image of the virtual interactive object, and releases the scent of the virtual interactive object through the scent release port on the interactive box when the target pet shows loneliness or anxiety.

[0052] S304, collecting interaction data between the target pet and the virtual interactive pet, and analyzing the preferences of the target pet through the collected interaction data.

[0053] The interaction data between the target pet and the virtual pet are collected, and the interaction data include but are not limited to the behavior records of the target pet (such as playing, eating, resting, etc.), the types of interaction objects (such as humans, virtual pets, toys, etc.), the time and place of interaction, and the results of interaction (such as the reaction of the target pet, emotional changes, etc.). The collected data are analyzed by statistical analysis methods, and the frequency data of the target pet's interaction with the virtual pet in different time periods are obtained. Statistics show that the average time the target pet interacts with the virtual pet every day is 2 hours, and the standard deviation of the interaction time is 0.5 hours, indicating that there is a certain volatility in the interaction time between the target pet and the virtual pet.

[0054] S305, recommending a matching interactive pet according to the preferences of the target pet analyzed in step S304.

[0055] The basic information collected about the preferences of the target pet is converted into a numerical feature vector, the personality traits are converted into a numerical vector using one-hot encoding, and the age, body shape and health status are directly represented numerically. A weight is assigned to each feature based on expert experience and actual data results. The personality traits have a high weight. The value of each feature is multiplied by the corresponding weight to obtain a weighted score. The cosine similarity is calculated, and the cosine value of the two feature vectors is calculated. The interactive pets with cosine values ​​greater than the preset recommendation threshold are recommended as interactive pets.

[0056] In some embodiments, the information of the target pet and the three candidate pets is converted into numerical feature vectors, the feature vectors are weighted and scored, and the cosine similarity is used to calculate the feature vector similarity between the target pet and the three candidate pets to obtain the similarity scores: candidate pet A (0.65), candidate pet B (0.70), candidate pet C (0.85). The target pet likes to be active in the afternoon and evening, and candidate pets with lively personalities and who like to interact in the afternoon and evening are preferred. The scores are dynamically adjusted, and the adjusted scores are candidate pet A (0.65, unchanged), candidate pet B (0.75, slightly improved), and candidate pet C (0.90, significantly improved). According to the weighted scores, similarity calculation results, and dynamic adjustment factors, a final recommendation list is generated, and the final recommendation list is: candidate pet C (highest match), candidate pet B (second best), and candidate pet A (lowest).

[0057] The technical solutions in the above embodiments of the present application have at least the following technical effects or advantages: Generate personalized virtual friend information based on the target pet's preferences and interaction quality, and provide a realistic virtual interaction experience. By analyzing the target pet's interaction records, more suitable friends can be intelligently recommended to improve the target pet's success rate in making friends. By comprehensively recording the pet's communication process and interaction quality, the system can generate personalized virtual friend information and provide a realistic virtual interaction experience when the interactive pet is not online.

[0058] Embodiment 4: By recording the interaction process between the target pet and the interactive pet, the interaction process is edited into a video and sent to the pet owner. The pet owner can judge the state of the target pet based on the received video, and according to the state of the target pet, the operating system releases the image, sound, smell and other information of the virtual pet to guide the state of the target pet, such as Figure 5 shown.

[0059] S401, based on the pet interaction data collected in step S301, feature extraction is performed on the collected data.

[0060] Specifically, for the extraction of sound features, the system converts the sound signal from the time domain to the frequency domain through fast Fourier transform (FFT), extracts the pitch features, and extracts the volume features by calculating the peak values; for the extraction of image features, the convolutional neural network (CNN) is used to train the pet's facial images, which can recognize the pet's happy, sad, angry and surprised expressions, and the key point detection algorithm is used to extract the pet's body posture features. The feature point tracking algorithm can be used to calculate the pet's movement speed in the video; for the extraction of odor features, the type of odor is identified by using a sensor array.

[0061] S402, identifying the state of the target pet according to the features extracted in step S401.

[0062] Among them, the state of the target pet is identified using a machine learning algorithm, and the extracted features are trained using a neural network machine learning algorithm. In the training stage, the neural network is optimized by introducing known state data as learning samples, and the extracted features are input into the optimized neural network. The state of the target pet is identified based on the neural network. When the neural network detects that the target pet emits high-pitched, fast-paced calls, and shows active body posture and faster movement speed, it may judge that the target pet is in an excited state. On the contrary, if the target pet emits low-pitched, slow-paced calls, has a relaxed body posture, and moves slowly, the system may judge that the target pet is in a calm state.

[0063] S403, editing the video clips according to the state of the target pet identified in step S402.

[0064] Specifically, when the system detects that the state of the target pet has changed, the system records the starting time point of the change. The starting time point is the time corresponding to the frame where the state begins to change. The video is composed of a series of continuous frames. When the state of the target pet ends or changes to another state, the ending time point of the change is recorded. The ending time point marks the end of the current state. The recorded starting and ending time points are corresponded to the frame numbers of the video. Each frame has a unique number, so that the system can locate the time point on a specific frame. At the same time, the timestamps of the start and end of each state are recorded. The timestamps are in seconds or milliseconds. The timestamps indicate the specific time point when the state change occurs in the video. According to the marked time points, the corresponding clips of each state are intercepted from the original video, the intercepted videos are edited to improve the visual effect, and the edited videos are sorted in chronological order, so that the pet owner can view the state of the target pet in chronological order.

[0065] In some embodiments, in a video of a pet cat playing in the house, the system recognizes the following state changes through a neural network: the pet cat starts to run from a stationary state, the starting frame of this state change is the 150th frame, and the timestamp is 00:00:02.500 (i.e., 2 seconds and 500 milliseconds from the start of the video); the pet cat stops to sniff the ground after running for a while, and the end frame of this state change is the 300th frame, and the timestamp is 00:00:05.000 (i.e., 5 seconds from the start of the video); the pet cat starts to run again, and the starting frame of this state change is the 301st frame, and the timestamp is 00:00:05.001 (due to the high frame rate, the timestamps may only differ by 1 millisecond).

[0066] S404, according to the real-time status of the target pet, matching corresponding information from the virtual interaction information library to guide the target pet.

[0067] According to the current state, personality, preferences and habits of the target pet, features of the current state of the target pet are extracted, and a matching model is established based on the extracted feature vector. The features of the current state, personality, preferences and habits of the target pet are quantified into comparable indicators as an input to the matching model. The image, sound, and smell information features in the virtual interaction information library are also quantified into comparable indicators as another input to the matching model. The cosine similarity is used to calculate the similarity of the two input information, and a matching threshold is set. When the calculated similarity is greater than the preset matching threshold, it is regarded as a matching result.

[0068] Furthermore, the pet owner can use the interactive box system interface to show the matching results to the target pet in a preset order and time interval, play sounds and release odors to guide the target pet. During the guidance process, the target pet's reaction and state changes are continuously monitored. The system feeds back the real-time reaction and state changes of the target pet to the pet owner in the form of charts, so that the pet owner can understand the guidance effect. According to the real-time feedback, the pet owner can adjust the guidance strategy in time, such as changing the order of playing content, adjusting the volume or the concentration of odor, to optimize the guidance effect. The system regularly generates a summary report on the target pet's state guidance, which can help the pet owner better understand the needs of the target pet.

[0069] The technical solutions in the above embodiments of the present application have at least the following technical effects or advantages: The system can comprehensively record the pet's interaction process, identify the different states of the target pet through data analysis, automatically edit video clips according to the state of the target pet, provide personalized video sharing content for pet owners, and realize intelligent pet state guidance through matching and playback of virtual interactive information. Pet owners can check the state of the target pet, share video clips and provide state guidance anytime and anywhere through mobile devices such as mobile phones.

[0070] Combining pet interaction records, video editing and virtual interaction technology, the system provides pet owners with a new and intelligent way to manage their pets' status. By comprehensively recording the pet's interaction process, identifying different states and editing video clips, the system can help pet owners better understand their pets' behavioral patterns and emotional states, and achieve intelligent guidance of their pets' status through the matching and playback of virtual interaction information.

[0071] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for operating a pet online interactive system, characterized in that: include: S101, collecting real-time audio data and video data, obtaining real-time audio data and video data of a pet; Acquire real-time audio and video from the host of the pet device. The host of the pet device collects real-time audio and video data through the integrated audio acquisition module and video acquisition module; S102, using an odor sensor to collect odors from the pet, converting the collected odor information into electrical signals, and transmitting the electrical signals to a remote server via a network; S103, monitoring whether the pet meets the conditions for online interaction based on the collected real-time audio data, video data and smell; S104, according to step S103, it is monitored that the target pet meets the online interaction condition, a first interactive pet is determined from the interactive pets that are currently online and idle using a preset pet recommendation method, and the target pet interacts with the interactive pet through the interaction box; The pet recommendation method constructs a weighted evaluation function according to the voice matching degree, the action matching degree, the user preference and the popularity; S105, establishing a communication link between the target pet and the first interactive pet; S106, recording in real time the interactive action information and interactive voice information between the target pet and the first interactive pet during the online interaction process.

2. A pet online interactive system operation method as claimed in claim 1, characterized in that: The monitoring target pet satisfies the online interaction condition. If the complete body of the target pet is within the field of view in the video screen, or more than half of the body of the target pet is within the field of view and the head is facing the local video collector on the device side, the condition is determined to be met; otherwise, the condition is determined not to be met; If the concentration of the target pet odor exceeds a preset threshold, it is determined that the condition is met, and if the concentration of the target pet odor does not exceed the threshold, it is determined that the condition is not met.

3. A pet online interactive system operation method as claimed in claim 1, characterized in that: The weighted evaluation function = w1 Among them, w1, , and is the weight coefficient.

4. A pet online interactive system operation method as claimed in claim 1, characterized in that: The method to make the target pet interact with the interactive pet through the interaction box is as follows: S201, setting an interaction box, wherein a main screen, an auxiliary screen, an odor collection module and an odor restoration module are provided on the interaction box, wherein the odor restoration module includes an odor generator and a release port; The main screen is located at the center of the front of the interactive box, the auxiliary screens are distributed around the main screen or on the side of the box, the odor release port is located below the auxiliary screen, the odor release port is electrically connected to the auxiliary screen, and each odor release port corresponds to an auxiliary screen; S202, placing the interactive box around the pet, the odor collection module collects the odor information of the pet, and uploads it to the remote server. After receiving the uploaded odor information, the remote server generates a recommendation list; The collected odor information is converted into electrical signals and transmitted to the remote server in real time through the network; S203, the odor restoration module restores the odor of the interactive pet through a built-in odor generator according to the received odor information of the interactive pet.

5. A pet online interactive system operation method as claimed in claim 4, characterized in that: The target pet interacts with the interactive pet through the interactive box, records the interactive data, generates a virtual pet based on the interactive data, and performs virtual interaction with the virtual pet when the interactive pet is not online.

6. A pet online interactive system operation method as claimed in claim 5, characterized in that: The steps of the target pet virtual interaction are: S301, collecting interaction data between the target pet and the interactive pet in real time, and extracting the interaction preference and interaction habit of the target pet based on the interaction data; S302, generating a virtual interactive pet according to the interactive process between the target pet and the interactive pet recorded in step S301; S303, performing virtual interaction according to the virtual interactive pet generated in step S302; S304, collecting interaction data between the target pet and the virtual interactive pet, and analyzing the preferences of the target pet through the collected interaction data; S305, based on the preferences of the target pet analyzed in step S304, the system recommends a matching interactive pet.

7. A pet online interactive system operation method as claimed in claim 6, characterized in that: The collected basic information of the target pet's preferences is converted into a numerical feature vector, the personality characteristics are converted into a numerical vector using one-hot encoding, a weight is assigned to each feature, a weighted score is calculated based on the weight assignment, and the interactive pets whose weighted scores are greater than the preset recommendation threshold are recommended as interactive pets.

8. A pet online interactive system operation method as claimed in claim 6, characterized in that: S401, according to the target pet interaction data collected in step S301, feature extraction is performed on the collected data; S402, identifying the state of the target pet according to the features extracted in step S401; S403, editing the video clips according to the state of the target pet identified in step S402; S404, according to the real-time status of the target pet, match the information from the virtual interaction information library to guide the target pet.

9. A pet online interactive system operation method as claimed in claim 8, characterized in that: The feature extraction includes sound features, image features and odor features. For the extraction of sound features, the system converts the sound signal from the time domain to the frequency domain through fast Fourier transform, extracts the pitch features, and extracts the volume features by calculating the peak value; for the extraction of image features, a convolutional neural network is used to train the facial image of the target pet; the posture features of the target pet's body are extracted through a key point detection algorithm, and the movement speed of the target pet in the video can be calculated using a feature point tracking algorithm; for the extraction of odor features, the type of odor is identified by using a sensor array.

10. A pet online interactive system operation method as claimed in claim 8, characterized in that: The video is composed of a series of continuous frames. When the system detects that the state of the target pet has changed, the starting time point of the change is recorded. The starting time point is the time corresponding to the frame where the state begins to change. When the state of the target pet ends or changes to another state, the ending time point of the change is recorded. The ending time point marks the end of the current state. The recorded starting and ending time points are corresponded to the frame numbers of the video. Each frame has a unique number, so that the system can locate the time point on a specific frame.

Images