CN-121970694-A - Method for stopping pet garbage can turning behavior, related equipment and program product

CN121970694ACN 121970694 ACN121970694 ACN 121970694ACN-121970694-A

Abstract

The application discloses a method, related equipment and program product for stopping the behavior of a pet garbage can, which are characterized in that the first behavior (raking behavior or garbage can turning behavior) of the pet is detected based on the behavior of the pet, and whether garbage exists around the pet is confirmed by environment information under the condition that the detection result shows that the first behavior exists, so that whether a stopping measure needs to be output or not is determined, and the strategy can obviously reduce the false alarm rate. In addition, the environment information analysis generally consumes more computing resources than the behavior detection, the lightweight behavior detection is performed first, the high-cost environment information analysis is invoked only after triggering, and the power consumption of the equipment can be obviously reduced. Further, in some practical situations, the dustbin may be in an empty state, so that even if it is determined by behavior detection that the pet is turning the dustbin at present, no garbage is found after environmental information analysis, no stopping measures are needed at this time, and useless triggering stopping measures can be avoided.

Inventors

JIANG ZHIWEN
WANG XIAOFEI
WU YUSHENG

Assignees

安徽淘云科技股份有限公司

Dates

Publication Date: 20260505
Application Date: 20260331

Claims (11)

1. A method for stopping pet garbage can turning behavior, which is characterized by being applied to a pet collar, comprising the following steps: Obtaining behavior characterization data of the pet; Based on the behavior representation data of the pet, identifying whether the pet has a first behavior, and obtaining a first identification result, wherein the first behavior represents a raking behavior or a garbage can turning behavior of the pet; collecting environmental information around the pet through an environmental sensing unit on the pet collar under the condition that the first identification result indicates that the first behavior exists in the pet; Identifying whether garbage exists around the pet based on the environmental information, and if so, outputting a stopping measure to stop the garbage turning behavior of the pet.
2. The method of claim 1, wherein identifying whether the pet has a first behavior based on the behavior characterization data of the pet, the process of obtaining a first identification result, comprises: sending the behavior characterization data of the pet into a pre-trained behavior classification model to obtain a first recognition result output by the model, wherein the first recognition result represents whether the pet has a first behavior; The behavior classification model is obtained by training the behavior characterization training data of the pets marked with the behavior labels.
3. The method of claim 1, wherein identifying whether the pet has a first behavior based on the behavior characterization data of the pet, the process of obtaining a first identification result, comprises: matching the behavior representation data of the pet with reference behavior data, so as to obtain a matching result, wherein the reference behavior data is a characteristic parameter for representing the first behavior of the pet; and based on the matching result, obtaining a first identification result of whether the first behavior exists in the pet.
4. A method according to claim 3, wherein when the first behaviour is indicative of a raking behaviour of a pet, the reference behaviour data comprises a first action indicative of a grabbing and raking action of a limb of the pet; When the first action represents the garbage can turning action of the pet, the reference action data comprises an action sequence of alternating first actions and second actions, and the second actions represent the sniffing actions of the pet.
5. The method of claim 1, wherein identifying whether the pet has a first behavior based on the behavior characterization data of the pet, the process of obtaining a first identification result, comprises: And calling the configured large model to indicate whether the large model identifies the first behavior of the pet based on the behavior characterization data of the pet, and obtaining a first identification result output by the large model.
6. The method of claim 1, wherein the environmental information includes an environmental image and an environmental smell, and the identifying whether the pet has garbage around based on the environmental information comprises: detecting whether garbage exists in the environment image or not through an image recognition algorithm, and judging whether the environment smell is matched with the set garbage spoilage smell characteristics or not; and if the environmental image is detected to contain garbage and the environmental smell is matched with the set garbage putrefactive smell characteristics, determining that garbage exists around the pet.
7. The method of claim 1, wherein, in the case where the first recognition result indicates that the pet has the first behavior, the process of collecting environmental information around the pet through the environmental sensing unit on the pet collar includes: and under the condition that the first recognition result indicates that the first behavior exists in the pet, controlling to start an environment sensing unit on the pet collar so as to acquire environment information around the pet through the environment sensing unit.
8. The method of any one of claims 1-7, further comprising: And on the basis of outputting the stopping measures, sending warning information to a user terminal associated with the pet collar so as to remind the user that the pet is turning over the dustbin currently.
9. A pet collar, comprising a collar body; the necklace body is provided with a multi-mode sensor for collecting behavior characterization data of the pet; The necklace body is further provided with a processor and an output module, the processor is used for identifying whether a first behavior of the pet exists based on behavior representation data of the pet to obtain a first identification result, the first behavior represents a pulling behavior or a garbage can turning behavior of the pet, the environmental information around the pet, collected by an environmental sensing unit on the pet necklace, is obtained when the first identification result represents that the pet exists in the first behavior, whether garbage exists around the pet is identified based on the environmental information, and if so, the output module is controlled to output a deterrent measure to deter the garbage turning behavior of the pet.
10. A readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of the method of deterring pet litter bin activity of any one of claims 1-8.
11. A computer program product comprising a computer program which, when executed by a processor, performs the steps of the method of deterring pet litter box activity as claimed in any one of claims 1 to 8.

Description

Method for stopping pet garbage can turning behavior, related equipment and program product Technical Field The application relates to the technical field of intelligent devices for pets, in particular to a method for stopping the action of turning over a dustbin for pets, related equipment and a program product. Background Pets are organisms that people are housed for mental purposes, not for economic purposes. A pet collar is a special appliance for protecting, controlling and towing pets. With the development of technology, some pet collars are also used to identify the identity of the pet. In order to facilitate the owner to find the pet, part of the owners can buy the pet collar (pet card) to the pet belt, so that the pet can be conveniently positioned, and other people can conveniently contact the owner through the pet collar (pet card) when the pet is lost. Some cities also require owners to handle pet collars in order to facilitate management of the pets. The existing pet collar is simpler in design, and more part of information of the pet and part of information of the owner, such as names, addresses, contact modes and the like of the owner, are recorded. Such pet collars are too single in function and often have limited assistance to the owner. Part of the pets can turn over the garbage can, such as dogs, and the pets can turn over the garbage can, so that the garbage in the garbage can is scattered indoors, the difficulty of cleaning the room is increased, the pets can turn over bones in kitchen garbage, such as chicken bones, and the bones are eaten by the pets, so that health problems are easily caused. Disclosure of Invention In view of the foregoing, the present application has been made to provide a method, related apparatus, and computer program product for deterring a pet from turning over a trash can, discovering the phenomenon of turning over the trash can in time, and taking appropriate deterrent measures to deter, thereby ensuring environmental cleanliness and health of the pet. The specific scheme is as follows: In a first aspect, a method for stopping the action of turning over a dustbin by a pet is provided, and is applied to a pet collar, and the method comprises the following steps: Obtaining behavior characterization data of the pet; Based on the behavior representation data of the pet, identifying whether the pet has a first behavior, and obtaining a first identification result, wherein the first behavior represents a raking behavior or a garbage can turning behavior of the pet; collecting environmental information around the pet through an environmental sensing unit on the pet collar under the condition that the first identification result indicates that the first behavior exists in the pet; Identifying whether garbage exists around the pet based on the environmental information, and if so, outputting a stopping measure to stop the garbage turning behavior of the pet. In another implementation manner of the first aspect of the embodiment of the present application, based on the behavior characterization data of the pet, the process of identifying whether the first behavior exists in the pet and obtaining the first identification result includes: sending the behavior characterization data of the pet into a pre-trained behavior classification model to obtain a first recognition result output by the model, wherein the first recognition result represents whether the pet has a first behavior; The behavior classification model is obtained by training the behavior characterization training data of the pets marked with the behavior labels. In another implementation manner of the first aspect of the embodiment of the present application, based on the behavior characterization data of the pet, the process of identifying whether the first behavior exists in the pet and obtaining the first identification result includes: matching the behavior representation data of the pet with reference behavior data, so as to obtain a matching result, wherein the reference behavior data is a characteristic parameter for representing the first behavior of the pet; and based on the matching result, obtaining a first identification result of whether the first behavior exists in the pet. In another implementation of the first aspect of the embodiments of the present application, when the first behavior represents a raking behavior of a pet, the reference behavior data includes a first action, the first action representing a grabbing action of a limb of the pet; When the first action represents the garbage can turning action of the pet, the reference action data comprises an action sequence of alternating first actions and second actions, and the second actions represent the sniffing actions of the pet. In another implementation manner of the first aspect of the embodiment of the present application, based on the behavior characterization data of the pet, the process of identifying whether the first behavior