Method, device, equipment and medium for sending urine retention test results

Abstract

This disclosure provides embodiments of a method, apparatus, device, and medium for transmitting urinary retention test results. One specific implementation of the method includes: acquiring a set of captured image sequences; generating at least one behavioral information and at least one behavioral frequency information; filtering out subsequences of captured images; generating excretion time information, excretion shape information, excretion foreign object information, and excretion location information; generating daily activity behavior information; inputting at least one behavioral information, at least one behavioral frequency information, excretion time information, excretion shape information, excretion foreign object information, excretion location information, and daily activity behavior information into a logistic regression model and a tree model to generate a first urinary retention test result and a second urinary retention test result; generating an actual urinary retention test result; and, in response to determining that urinary retention symptoms are present, sending cat information and the actual urinary retention test result to a user terminal. This implementation can accurately and efficiently generate urinary retention test results corresponding to a target cat.

Description

Technical Field

[0001] Embodiments of the present disclosure relate to the field of computer technology, and in particular to a method, device, equipment, and medium for sending urine retention detection results. Background Art

[0002] At present, urinary retention in cats is a disease that may be caused by a variety of reasons, including but not limited to: insufficient drinking water, urinary system problems, and stress reactions. How to detect whether a cat has urinary retention symptoms is also known as one of the main development directions in the current disease field. For the detection of urinary retention symptoms in cats, the usual method is to detect whether a cat has urinary retention symptoms through human observation or diagnosis by relevant personnel.

[0003] However, when using the above method to detect whether a cat has urinary retention symptoms, the following technical problems often occur:

[0004] First, it requires relevant personnel to observe the cat's behavior information for a long time, which leads to low efficiency of manual detection and inaccurate results.

[0005] Second, after a cat shows symptoms of urinary retention, it is also crucial to accurately determine the corresponding cause of the symptoms.

[0006] The above information disclosed in this Background section is only for enhancement of understanding of the background of the inventive concept and therefore it may contain information that does not form the prior art that is already known in this country to a person of ordinary skill in the art. Summary of the invention

[0007] The content of this disclosure is used to introduce concepts in a brief form, which will be described in detail in the detailed implementation section below. The content of this disclosure is not intended to identify the key features or essential features of the technical solution claimed for protection, nor is it intended to limit the scope of the technical solution claimed for protection.

[0008] Some embodiments of the present disclosure propose methods, devices, equipment and media for sending urine retention detection results to solve one or more of the technical problems mentioned in the above background technology section.

[0009] In a first aspect, some embodiments of the present disclosure provide a method for sending urine retention detection results, comprising: obtaining a set of captured picture sequences corresponding to a target cat in a target time period, wherein the captured picture sequences in the captured picture sequence set correspond to the cameras in the camera device set, and the camera device set comprises: at least one camera device placed in the activity area corresponding to the target cat, and a target camera device placed in a cat litter box; based on the target captured picture sequence, generating at least one behavior information and corresponding at least one behavior frequency information for the target cat, wherein the target captured picture sequence is a captured picture sequence corresponding to the target camera device; in response to determining that the at least one behavior information includes excretion behavior information, filtering out a captured picture subsequence associated with the excretion behavior information from the target captured picture sequence; based on the captured picture subsequence, generating Excretion time information, excretion shape information, excretion foreign body information, excretion location information; based on at least one shooting picture sequence, generate daily activity behavior information for the above-mentioned target cat, wherein there is a corresponding relationship between the above-mentioned at least one shooting picture sequence and at least one camera device; input the above-mentioned at least one behavior information, the above-mentioned at least one behavior frequency information, the above-mentioned excretion time information, the above-mentioned excretion shape information, the above-mentioned excretion foreign body information, the above-mentioned excretion location information, and the above-mentioned daily activity behavior information into a logistic regression model and a tree model to generate a first urine retention detection result and a second urine retention detection result; based on the above-mentioned first urine retention detection result and the above-mentioned second urine retention detection result, generate an actual urine retention detection result; in response to determining that the above-mentioned actual urine retention detection result indicates the presence of urine retention symptoms, send the cat information corresponding to the above-mentioned target cat and the above-mentioned actual urine retention detection result to the user terminal of the target care object.

[0010] In a second aspect, some embodiments of the present disclosure provide a device for sending urine retention detection results, comprising: an acquisition unit, configured to acquire a set of captured picture sequences corresponding to a target cat in a target time period, wherein the captured picture sequences in the captured picture sequence set correspond to the cameras in the camera device set, and the camera device set comprises: at least one camera device placed in the activity area corresponding to the target cat, and a target camera device placed in the cat litter box; a first generation unit, configured to generate at least one behavior information and corresponding at least one behavior frequency information for the target cat according to the target captured picture sequence, wherein the target captured picture sequence is a captured picture sequence corresponding to the target camera device; a screening unit, configured to screen out a captured picture subsequence associated with the excretion behavior information from the target captured picture sequence in response to determining that the at least one behavior information includes excretion behavior information; a second generation unit, configured to generate a screening subsequence according to the captured picture subsequence. excretion time information, excretion shape information, excretion foreign body information, excretion position information; a third generating unit is configured to generate daily activity behavior information for the above-mentioned target cat according to at least one shooting picture sequence, wherein there is a corresponding relationship between the above-mentioned at least one shooting picture sequence and at least one camera device; a fourth generating unit is configured to input the above-mentioned at least one behavior information, the above-mentioned at least one behavior frequency information, the above-mentioned excretion time information, the above-mentioned excretion shape information, the above-mentioned excretion foreign body information, the above-mentioned excretion position information, and the above-mentioned daily activity behavior information into a logistic regression model and a tree model to generate a first urine retention detection result and a second urine retention detection result; a fifth generating unit is configured to generate an actual urine retention detection result according to the above-mentioned first urine retention detection result and the above-mentioned second urine retention detection result; a sending unit is configured to send the cat information corresponding to the above-mentioned target cat and the above-mentioned actual urine retention detection result to the user terminal of the target care object in response to determining that the above-mentioned actual urine retention detection result indicates the presence of urine retention symptoms.

[0011] In a third aspect, some embodiments of the present disclosure provide an electronic device comprising: one or more processors; a storage device on which one or more programs are stored, and when the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any implementation manner in the first aspect.

[0012] In a fourth aspect, some embodiments of the present disclosure provide a computer-readable medium having a computer program stored thereon, wherein when the program is executed by a processor, the method described in any implementation manner in the first aspect is implemented.

[0013] The above-mentioned embodiments of the present disclosure have the following beneficial effects: the urine retention detection result sending method of some embodiments of the present disclosure can accurately and efficiently generate the urine retention detection result corresponding to the target cat. Specifically, the reason why the relevant urine retention detection result is not accurate enough is that the relevant personnel need to observe the behavior information of the cat for a long time, resulting in low efficiency of manual detection and inaccurate detection results. Based on this, the urine retention detection result sending method of some embodiments of the present disclosure first obtains a shooting picture sequence set corresponding to the target cat in the target time period, wherein the shooting picture sequence in the above-mentioned shooting picture sequence set has a corresponding relationship with the camera in the camera set, and the above-mentioned camera set includes: at least one camera placed in the activity area corresponding to the above-mentioned target cat, and a target camera placed in the cat litter box. Here, by setting the target camera and at least one camera, the global behavior monitoring of the target cat can be realized, and according to the obtained shooting picture sequence set, the real-time detection of the urine retention symptoms of the target cat can be realized. Then, according to the target shooting picture sequence, at least one behavior information and corresponding at least one behavior frequency information for the above-mentioned target cat can be accurately generated. Among them, the above-mentioned target shooting picture sequence is a shooting picture sequence corresponding to the above-mentioned target camera. Here, the obtained at least one behavior information and the corresponding behavior frequency information are used as detection indicators for determining the subsequent urinary retention symptoms, so as to accurately generate the urinary retention detection results in the subsequent time. Then, in response to determining that the above-mentioned at least one behavior information includes excretion behavior information, a shooting picture subsequence associated with the excretion behavior information is screened out from the above-mentioned target shooting picture sequence, so as to further determine the urinary retention symptoms from the perspective of excretion behavior, and ensure the accuracy of the subsequent urinary retention detection results. Then, according to the above-mentioned shooting picture subsequence, the excretion time information, excretion shape information, excretion foreign body information, and excretion position information can be accurately generated as detection indicators for subsequent urinary retention symptoms, so as to accurately generate urinary retention detection results in the subsequent time. Next, according to at least one shooting picture sequence, the daily activity behavior information for the above-mentioned target cat can be accurately generated. Among them, there is a corresponding relationship between the above-mentioned at least one shooting picture sequence and at least one camera device. Here, the urinary retention symptoms are further determined by the dimension of daily activity behavior to ensure the accuracy of subsequent detection. Furthermore, by inputting the above-mentioned at least one behavior information, the above-mentioned at least one behavior frequency information, the above-mentioned excretion time information, the above-mentioned excretion shape information, the above-mentioned excretion foreign body information, the above-mentioned excretion location information, and the above-mentioned daily activity behavior information into the logistic regression model and the tree model, the first urine retention detection result and the second urine retention detection result can be accurately generated. Here, the corresponding urine retention detection result is accurately generated by a dual-model method, avoiding the occurrence of a situation where a single model may not detect accurately enough. Furthermore, based on the above-mentioned first urine retention detection result and the above-mentioned second urine retention detection result, the actual urine retention detection result can be accurately generated.Finally, in response to determining that the actual urine retention detection result indicates the presence of urine retention symptoms, the cat information corresponding to the target cat and the actual urine retention detection result are sent to the user terminal of the target care object to achieve timely treatment of the urine retention symptoms of the target cat. In summary, through the generation of indicators related to urine retention symptoms from various aspects and the use of dual models, the urine retention detection result corresponding to the target cat can be accurately and efficiently generated. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and that components and elements are not necessarily drawn to scale.

[0015] Figure 1 is a flow chart of some embodiments of the method for sending urine retention detection results according to the present disclosure;

[0016] Figure 2 is a schematic structural diagram of some embodiments of a urine retention detection result sending device according to the present disclosure;

[0017] Figure 3 It is a schematic diagram of the structure of an electronic device suitable for implementing some embodiments of the present disclosure. DETAILED DESCRIPTION

[0018] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments set forth herein. On the contrary, these embodiments are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0019] It should also be noted that, for ease of description, only the parts related to the invention are shown in the drawings. In the absence of conflict, the embodiments and features in the embodiments of the present disclosure can be combined with each other.

[0020] It should be noted that the concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0021] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0022] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0023] The present disclosure will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.

[0024] refer to Figure 1 , shows a process 100 of some embodiments of the method for sending urine retention detection results according to the present disclosure. The method for sending urine retention detection results comprises the following steps:

[0025] Step 101, obtaining a set of shot picture sequences corresponding to a target cat and in a target time period.

[0026] In some embodiments, the execution subject (for example, an electronic device) of the above-mentioned urine retention detection result sending method can obtain a set of shooting picture sequences corresponding to the target cat in the target time period through a wired connection method or a wireless connection method. Among them, the target cat can be a cat to be tested for urine retention symptoms. The target time period can be a preset time period. For example, the target time period can be from 8 am to 4 pm every day. There is a one-to-one correspondence between the shooting picture sequences in the above-mentioned shooting picture sequence set and the camera devices in the camera device set. The above-mentioned camera device set includes: at least one camera device placed in the activity area corresponding to the above-mentioned target cat, and a target camera device placed in the cat litter box. The shooting picture sequence can be each shooting picture in the shooting video. The target camera device placed in the cat litter box can be a device for photographing the cat litter box.

[0027] Step 102: Generate at least one behavior information and corresponding at least one behavior frequency information for the target cat according to the target shot picture sequence.

[0028] In some embodiments, the execution subject may generate at least one behavior information and corresponding at least one behavior frequency information for the target cat according to the target shot picture sequence. The at least one behavior information may be behavior information of at least one behavior performed by the target cat in the litter box area. For example, the at least one behavior information may include but is not limited to at least one of the following: excretion behavior information, strolling behavior information, lying down behavior information. There is a one-to-one correspondence between the behavior frequency information in at least one behavior frequency information and the behavior information in at least one behavior information. The behavior frequency information may characterize the number of occurrences of the behavior corresponding to the behavior information.

[0029] In some optional implementations of some embodiments, the generating of at least one behavior information and corresponding at least one behavior frequency information for the target cat according to the target shot picture sequence may include the following steps:

[0030] The first step is to obtain a preset picture group number set. There is a geometric relationship between the picture group numbers in the above picture group number set. For example, the picture group number may be the number of captured pictures included in each captured picture group after division. For example, the picture group number set may be [1, 2, 4].

[0031] In the second step, for each picture group number in the picture group number set, the following first generation step is performed:

[0032] Sub-step 1: divide each shot in the target shot sequence according to the number of shot groups to generate a shot group sequence, wherein the number of shot included in each shot group is the number of shot groups.

[0033] As an example, the execution subject may determine a number of adjacent captured picture subsequences in each picture group as a group of captured picture groups to obtain a captured picture group sequence.

[0034] Sub-step 2: for each shot picture group in the shot picture group sequence, perform the following second generation step:

[0035] The first sub-step is to combine the shooting pictures according to the picture sequence corresponding to the shooting pictures in the shooting picture group to generate a combined picture.

[0036] As an example, the execution subject may combine the pictures according to the picture order of each picture in the picture group using a preset picture arrangement template to generate a combined picture. The picture arrangement template may represent a template for how to arrange pictures.

[0037] The second sub-step is to obtain the first convolution layer group corresponding to the above-mentioned combined picture. The first convolution layers in the above-mentioned first convolution layer group are connected in series, and the dimensions of the convolution kernels corresponding to the first convolution layers are set according to the dimensions of the picture corresponding to the above-mentioned combined picture. The first convolution layer can be a convolutional neural network layer. The dimensions of the convolution kernels corresponding to the first convolution layers are successively smaller.

[0038] For example, the dimension corresponding to the combined picture is 500*500. Then the corresponding first convolution layers include: a first convolution layer based on a 100*100 convolution kernel, a first convolution layer based on a 50*50 convolution kernel, and a first convolution layer based on a 20*20 convolution kernel.

[0039] The third sub-step is to input the combined picture into the first convolutional layer group to generate combined picture feature information, wherein the combined picture feature information can represent the semantic features of the picture content of the combined picture.

[0040] In the third step, for each combined picture feature information sequence in the obtained combined picture feature information sequence set, an initial behavior information group and a corresponding initial behavior probability information group are generated based on the above combined picture feature information sequence using a behavior detection model based on multi-head behavior output. Among them, the behavior detection model based on multi-head behavior output can be a behavior detection model whose output model is multiple outputs. There is a one-to-one correspondence between the output layer in each output layer and the detection behavior information in each detection behavior information. For example, the behavior detection model based on multi-head behavior output can be a multi-layer series of convolutional layers + a fully connected layer as multiple outputs.

[0041] The fourth step is to generate at least one behavior information and at least one corresponding behavior frequency information according to the obtained initial behavior information set and initial behavior probability information set, wherein the behavior information in the at least one behavior information and the behavior frequency information in the at least one behavior frequency information have a one-to-one correspondence.

[0042] As an example, first, deduplication processing is performed on each initial behavior information in the above-mentioned initial behavior information set to generate at least one initial behavior information as at least one behavior information. Then, for each behavior information in the at least one behavior information, the first step is to determine the initial behavior information set in the initial behavior information set that is the same as the above-mentioned behavior information. The second step is to remove the initial behavior information whose corresponding initial behavior probability information is less than a predetermined value from the above-mentioned initial behavior information set to obtain the initial behavior information set after removal. The third step is to determine the number of information corresponding to the initial behavior information in the above-mentioned initial behavior information set after removal as the behavior frequency information corresponding to the above-mentioned behavior information.

[0043] Optionally, the above-mentioned generating the initial behavior information group and the corresponding initial behavior probability information group according to the above-mentioned combined picture feature information sequence by using the behavior detection model based on multi-head behavior outputs may include the following steps:

[0044] In the first step, each target shot picture in the target shot picture sequence is input into the second convolution layer group to output picture feature information to obtain a picture feature information sequence. The second convolution layers in the second convolution layer group are connected in series. The dimensions of the convolution kernels corresponding to the second convolution layers are pre-set according to the picture dimensions corresponding to the target shot pictures. The picture dimensions can be the image size corresponding to the target shot pictures.

[0045] In the second step, the picture feature information sequence and the combined picture feature information sequence are input into the behavior detection model to generate the initial behavior information group and the corresponding initial behavior probability information group.

[0046] Step 103 , in response to determining that the at least one behavior information includes excretion behavior information, filter out a shot picture subsequence associated with the excretion behavior information from the target shot picture sequence.

[0047] In some embodiments, in response to determining that the at least one behavior information includes excretion behavior information, the execution subject may filter out a shot picture subsequence associated with the excretion behavior information from the target shot picture sequence.

[0048] As an example, first, the execution subject can detect the existence of a shooting picture subsequence corresponding to excrement in the target shooting picture sequence through the target image processing software as a candidate shooting picture subsequence, and then perform a predetermined time extension on the candidate shooting picture subsequence to generate a shooting picture subsequence.

[0049] Step 104, generating defecation time information, defecation shape information, defecation foreign body information, and defecation position information according to the above-mentioned shooting picture subsequence.

[0050] In some embodiments, the execution subject may generate excretion time information, excretion shape information, excretion foreign matter information, and excretion location information according to the above-mentioned shooting screen subsequence. Among them, the excretion time information may be the duration of the target cat's excretion behavior. The excretion foreign matter information may be information indicating whether there is blood stain on the excrement. The excretion shape information may indicate the shape of the excrement. The excretion location information may be the excretion location of the excrement.

[0051] As an example, the execution subject may determine the sequence duration corresponding to the candidate shot screen subsequence as the defecation time information. The target detection model is used to generate defecation shape information, defecation foreign body information, and defecation position information according to the shot screen subsequence. The target detection model may be a YOLOv5 model.

[0052] Step 105: Generate daily activity behavior information of the target cat according to at least one shot picture sequence.

[0053] In some embodiments, the execution subject may generate daily activity behavior information for the target cat according to at least one shooting picture sequence. The daily activity behavior information may be behavior information of the target cat's activity in the activity area. For example, the daily activity behavior information may include but is not limited to at least one of the following: cat meowing behavior information, cat running around behavior information, and licking behavior information.

[0054] As an example, first, the execution subject may input each shot picture sequence into a behavior detection model based on multi-head behavior output to generate an initial behavior information set. Then, each initial behavior information in the obtained at least one initial behavior information set is deduplicated to generate a deduplicated initial behavior information set as daily activity behavior information.

[0055] Step 106, input the above-mentioned at least one behavior information, the above-mentioned at least one behavior frequency information, the above-mentioned excretion time information, the above-mentioned excretion shape information, the above-mentioned excretion foreign body information, the above-mentioned excretion location information, and the above-mentioned daily activity behavior information into the logistic regression model and the tree model to generate a first urine retention detection result and a second urine retention detection result.

[0056] In some embodiments, the execution subject may input the at least one behavior information, the at least one behavior frequency information, the excretion time information, the excretion shape information, the excretion foreign body information, the excretion location information, and the daily activity behavior information into a logistic regression model and a tree model to generate a first urine retention detection result and a second urine retention detection result. The first urine retention detection result may include: a detection result without urine retention symptoms and a detection result with urine retention symptoms. The second urine retention detection result may include: a detection result without urine retention symptoms and a detection result with urine retention symptoms. The tree model may be a decision tree model.

[0057] Step 107: Generate an actual urine retention detection result based on the first urine retention detection result and the second urine retention detection result.

[0058] In some embodiments, the execution entity may generate an actual urine retention detection result based on the first urine retention detection result and the second urine retention detection result.

[0059] As an example, in response to determining that there is a detection result indicating the presence of a urinary retention symptom in the first urine retention detection result and the second urine retention detection result, an actual urine retention detection result indicating the presence of a urinary retention symptom is generated. In response to determining that there is no detection result indicating the presence of a urinary retention symptom in the first urine retention detection result and the second urine retention detection result, an actual urine retention detection result indicating that there is no urinary retention symptom is generated.

[0060] In some optional implementations of some embodiments, generating an actual urine retention detection result according to the first urine retention detection result and the second urine retention detection result may include the following steps:

[0061] In the first step, in response to determining that both the first urine retention detection result and the second urine retention detection result indicate the presence of a urine retention symptom, an actual urine retention detection result indicating the presence of a urine retention symptom is generated.

[0062] In the second step, in response to determining that both the first urine retention detection result and the second urine retention detection result indicate that there is no urine retention symptom, an actual urine retention detection result indicating that no urine retention symptom is detected is generated.

[0063] In the third step, in response to determining that the detection results between the first urine retention detection result and the second urine retention detection result are inconsistent, each target shot image in the target shot image sequence is input into the urine retention detection neural network to generate a third urine retention detection result. The urine retention detection neural network may be a pre-trained neural network model for generating urine retention detection results. Specifically, the urine retention detection neural network model may be an encoding and decoding model based on a convolutional neural network.

[0064] The fourth step is to generate an actual urine retention detection result based on the first urine retention detection result, the second urine retention detection result and the third urine retention detection result.

[0065] As an example, in response to determining that there is a detection result indicating the presence of a urinary retention symptom among the first urinary retention detection result, the second urinary retention detection result, and the third urinary retention detection result, an actual urinary retention detection result indicating the presence of a urinary retention symptom is generated. In response to determining that there is no detection result indicating the presence of a urinary retention symptom among the first urinary retention detection result, the second urinary retention detection result, and the third urinary retention detection result, an actual urinary retention detection result indicating that there is no urinary retention symptom is generated.

[0066] Optionally, the generating of the actual urine retention detection result according to the first urine retention detection result, the second urine retention detection result and the third urine retention detection result may include the following steps:

[0067] The first step is to determine at least one first result corresponding to the presence of urine retention symptoms and at least one second result corresponding to the absence of urine retention symptoms from the first urine retention detection result, the second urine retention detection result and the third urine retention detection result.

[0068] In the second step, in response to determining that the number of results corresponding to the at least one first result is greater than the number of results corresponding to the at least one second result, an actual urine retention detection result indicating that a urine retention symptom is detected is generated.

[0069] In a third step, in response to determining that the number of results corresponding to the at least one first result is less than the number of results corresponding to the at least one second result, an actual urine retention detection result indicating that no urine retention symptom is detected is generated.

[0070] Step 108, in response to determining that the actual urine retention test result indicates the presence of urine retention symptoms, the cat information corresponding to the target cat and the actual urine retention test result are sent to the user terminal of the target care object.

[0071] In some embodiments, in response to determining that the actual urine retention detection result indicates the presence of urine retention symptoms, the execution subject may send the cat information corresponding to the target cat and the actual urine retention detection result to a user terminal of the target care object.

[0072] In some optional implementations of some embodiments, after step 108, the steps further include:

[0073] In the first step, feature extraction is performed on each shot in the shot subsequence to generate shot feature information and obtain a shot feature information subsequence, wherein the shot feature information can represent the semantics of the picture content feature corresponding to the shot.

[0074] As an example, the execution entity may input each captured picture into an image feature extraction model to generate captured picture feature information.

[0075] In the second step, for each shot picture sequence in the at least one shot picture sequence, the following processing steps are performed:

[0076] Sub-step 1: removing the shot pictures whose corresponding picture contents do not contain the target cat from the shot picture sequence, to obtain a shot picture sub-sequence after the removal.

[0077] Sub-step 2: generating a sub-sequence of captured picture feature information for the above-mentioned removed captured picture sub-sequence.

[0078] The third step is to sort the obtained subsequences of the captured picture feature information according to the time sequence to obtain a sorted captured picture feature information sequence.

[0079] In the fourth step, the captured image feature information in the above-ordered captured image feature information sequence is input into a cat event detection generation model to generate at least one cat event information. The cat event detection generation model may be a neural network model for generating cat event information. In practice, the cat event information may be event information of a behavior event performed by a target cat. The cat event information may correspond to a behavior performed by a cat. For example, the cat event information may be an event of a cat eating. In practice, the cat event detection generation model may be a behavior detection model. For example, the cat event detection generation model may be a two-stream network or a 3D convolutional model using optical flow.

[0080] Step 5: Generate a cat urine retention detection reason for the target cat based on the at least one cat event information, wherein the cat urine retention detection reason may be the reason why the cat has urine retention symptoms.

[0081] As an example, the execution subject may generate a cat urine retention detection reason for the target cat according to the at least one cat event information by matching a cat event set with the urine retention reason.

[0082] Step 6: Send the above cat urine retention detection reason to the above user terminal.

[0083] Optionally, the generating of the cat urine retention detection reason for the target cat according to the at least one cat event information may include the following steps:

[0084] The first step is to determine the urinary retention influence probability information corresponding to each cat event information in the at least one cat event information, and obtain at least one urinary retention influence probability information. The urinary retention influence probability information can represent the probability of the cat event information causing the cat to have urinary retention symptoms. The urinary retention influence probability information can be a value between 0 and 1, and the larger the value, the greater the probability of the cat having urinary retention symptoms.

[0085] As an example, the execution subject can determine the urine retention influence probability information corresponding to each cat event information in the at least one cat event information by representing the association table between the cat event information and the urine retention influence probability information, and obtain at least one urine retention influence probability information. The association table represents the correspondence between at least one cat event information and the urine retention influence probability information. Each correspondence in the association table can be pre-generated. In practice, the association table includes: the correspondence between the cat event information group and the influence probability information, and the correspondence between the cat event information and the urine retention influence probability information.

[0086] The second step is to generate a cat urine retention detection reason based on the at least one urine retention impact probability information.

[0087] As an example, first, the execution subject can filter out cat event information whose corresponding urinary retention influence probability information is in the front target number from at least one cat event information to obtain a cat event information set. Then, by matching the cat event set with the urinary retention cause, the cat urinary retention detection cause corresponding to the cat event information set is determined.

[0088] Optionally, the generating of the cat urine retention detection reason for the target cat according to the at least one cat event information may include the following steps:

[0089] The first step is to sort each cat event information in the at least one cat event information to generate a cat event information sequence.

[0090] The second step is to determine the urinary retention influence probability information corresponding to each cat event information in the at least one cat event information, and obtain at least one urinary retention influence probability information.

[0091] The third step is to determine the influence probability information corresponding to each cat event information group in at least one cat event information group in the cat event information sequence, and obtain at least one influence probability information. The number of cat event information included in the cat event information group can be various numerical values. There is an adjacent relationship between the various cat event information in the cat event information group. For example, the cat event information sequence is {cat event information A, cat event information B, cat event information C, cat event information D, cat event information E}. At least one cat event information group may include: {cat event information A, cat event information B}, {cat event information A, cat event information B, cat event information C}, {cat event information A, cat event information B, cat event information C, cat event information D}, {cat event information B, cat event information C, cat event information D}, {cat event information B, cat event information C, cat event information D, cat event information E}, {cat event information B, cat event information C}, {cat event information B, cat event information C, cat event information D}, {cat event information C, cat event information D}, {cat event information C, cat event information D, cat event information E}.

[0092] As an example, the above-mentioned execution subject can determine the influence probability information corresponding to each cat event information group in at least one cat event information group in the cat event information sequence by representing the association table between cat event information and urinary retention influence probability information, and obtain at least one influence probability information.

[0093] Step 4: Select a cat event information group whose corresponding impact probability information is greater than a target value from the at least one cat event information group, and use it as a target cat event information group to obtain at least one target cat event information group. The target value may be a preset value. For example, the target value may be 0.7.

[0094] The fifth step is to determine the event information group weight corresponding to each target cat event information group in the at least one target cat event information group according to at least one influence probability information corresponding to the at least one target cat event information group, and obtain at least one event information group weight.

[0095] As an example, the execution entity may perform probability normalization processing on each piece of influence probability information in the at least one influence probability information to generate at least one event information group weight.

[0096] Step 6, for each target cat event information group in the at least one target cat event information group, perform the following determination steps:

[0097] Sub-step 1: determine whether there is cat event information whose corresponding urine retention influence probability information is greater than a preset probability among the cat event information included in the target cat event information group.

[0098] Sub-step 2, in response to determining existence, determining existence of cat event information corresponding to urine retention influence probability information greater than a preset probability as candidate cat event information, as at least one candidate cat event information.

[0099] Sub-step 3: determining the average information of the urinary retention impact probability corresponding to the at least one candidate cat event information as the first average information.

[0100] Sub-step 4: determining the average information of the urinary retention influence probability corresponding to each cat event information included in the target cat event information group as the second average information.

[0101] In the seventh step, the first average information in the at least one first average information obtained is correspondingly multiplied with the event information group weight in the at least one event information group weight to generate first multiplied information, thereby obtaining at least one first multiplied information.

[0102] In the eighth step, the second average information in the obtained at least one second average information is correspondingly multiplied with the event information group weight in the at least one event information group weight to generate second multiplied information, thereby obtaining at least one second multiplied information.

[0103] The ninth step is to determine the priority information between the first multiplication information and the second multiplication information, wherein the priority information indicates that the priority corresponding to the first multiplication information is higher than the priority corresponding to the second multiplication information.

[0104] The tenth step is to sort the target cat event information groups in the at least one target cat event information group in descending order of importance based on the at least one first multiplication information, the at least one second multiplication information and the priority information to obtain a target cat event information group sequence.

[0105] In the eleventh step, the detection reason corresponding to the first target cat event information group in the target cat event information group sequence is determined as the cat urine retention detection reason.

[0106] The above-mentioned optional implementation methods and optional contents in some embodiments, as an inventive point of the present disclosure, solve the second technical problem mentioned in the background technology: "After a cat develops urinary retention symptoms, how to accurately determine the corresponding cause of the urinary retention symptoms is also crucial." Based on this, first of all, the present disclosure uses the urinary retention impact probability information corresponding to the cat event information and the impact probability information corresponding to the cat event information group to determine in a multi-granular manner the probability of the occurrence of urinary retention symptoms when different events occur crosswise. Therefore, when determining the maximum probability of the occurrence of urinary retention symptoms, the corresponding detection cause of the corresponding cat event information group is the most reasonable and sufficiently accurate as the detection cause of the cat urinary retention.

[0107] The above-mentioned embodiments of the present disclosure have the following beneficial effects: the urine retention detection result sending method of some embodiments of the present disclosure can accurately and efficiently generate the urine retention detection result corresponding to the target cat. Specifically, the reason why the relevant urine retention detection result is not accurate enough is that the relevant personnel need to observe the behavior information of the cat for a long time, resulting in low efficiency of manual detection and inaccurate detection results. Based on this, the urine retention detection result sending method of some embodiments of the present disclosure first obtains a shooting picture sequence set corresponding to the target cat in the target time period, wherein the shooting picture sequence in the above-mentioned shooting picture sequence set has a corresponding relationship with the camera in the camera set, and the above-mentioned camera set includes: at least one camera placed in the activity area corresponding to the above-mentioned target cat, and a target camera placed in the cat litter box. Here, by setting the target camera and at least one camera, the global behavior monitoring of the target cat can be realized, and according to the obtained shooting picture sequence set, the real-time detection of the urine retention symptoms of the target cat can be realized. Then, according to the target shooting picture sequence, at least one behavior information and corresponding at least one behavior frequency information for the above-mentioned target cat can be accurately generated. Among them, the above-mentioned target shooting picture sequence is a shooting picture sequence corresponding to the above-mentioned target camera. Here, the obtained at least one behavior information and the corresponding behavior frequency information are used as detection indicators for determining the subsequent urinary retention symptoms, so as to accurately generate the urinary retention detection results in the subsequent time. Then, in response to determining that the above-mentioned at least one behavior information includes excretion behavior information, a shooting picture subsequence associated with the excretion behavior information is screened out from the above-mentioned target shooting picture sequence, so as to further determine the urinary retention symptoms from the perspective of excretion behavior, and ensure the accuracy of the subsequent urinary retention detection results. Then, according to the above-mentioned shooting picture subsequence, the excretion time information, excretion shape information, excretion foreign body information, and excretion position information can be accurately generated as detection indicators for subsequent urinary retention symptoms, so as to accurately generate urinary retention detection results in the subsequent time. Next, according to at least one shooting picture sequence, the daily activity behavior information for the above-mentioned target cat can be accurately generated. Among them, there is a corresponding relationship between the above-mentioned at least one shooting picture sequence and at least one camera device. Here, the urinary retention symptoms are further determined by the dimension of daily activity behavior to ensure the accuracy of subsequent detection. Furthermore, by inputting the above-mentioned at least one behavior information, the above-mentioned at least one behavior frequency information, the above-mentioned excretion time information, the above-mentioned excretion shape information, the above-mentioned excretion foreign body information, the above-mentioned excretion location information, and the above-mentioned daily activity behavior information into the logistic regression model and the tree model, the first urine retention detection result and the second urine retention detection result can be accurately generated. Here, the corresponding urine retention detection result is accurately generated by a dual-model method, avoiding the occurrence of a situation where a single model may not detect accurately enough. Furthermore, based on the above-mentioned first urine retention detection result and the above-mentioned second urine retention detection result, the actual urine retention detection result can be accurately generated.Finally, in response to determining that the actual urine retention detection result indicates the presence of urine retention symptoms, the cat information corresponding to the target cat and the actual urine retention detection result are sent to the user terminal of the target care object to achieve timely treatment of the urine retention symptoms of the target cat. In summary, through the generation of indicators related to urine retention symptoms from various aspects and the use of dual models, the urine retention detection result corresponding to the target cat can be accurately and efficiently generated.

[0108] Further references Figure 2 As an implementation of the methods shown in the above figures, the present disclosure provides some embodiments of a device for sending urine retention detection results. These device embodiments are similar to Figure 1 Corresponding to the method embodiments shown, the urine retention detection result sending device can be specifically applied to various electronic devices.

[0109] like Figure 2As shown, a urine retention detection result sending device 200 includes: an acquisition unit 201, a first generation unit 202, a screening unit 203, a second generation unit 204, a third generation unit 205, a fourth generation unit 206, a fifth generation unit 207 and a sending unit 208. Among them, the acquisition unit 201 is configured to acquire a set of captured picture sequences corresponding to the target cat in the target time period, wherein the captured picture sequences in the above-mentioned captured picture sequence set correspond to the camera devices in the camera device set, and the above-mentioned camera device set includes: at least one camera device placed in the activity area corresponding to the above-mentioned target cat, and a target camera device placed in the cat litter box; the first generation unit 202 is configured to generate at least one behavior information and corresponding at least one behavior frequency information for the above-mentioned target cat according to the target captured picture sequence, wherein the above-mentioned target captured picture sequence is a captured picture sequence corresponding to the above-mentioned target camera device; the screening unit 203 is configured to, in response to determining that the above-mentioned at least one behavior information includes excretion behavior information, screen out a captured picture subsequence associated with the excretion behavior information from the above-mentioned target captured picture sequence; the second generation unit 204 is configured to generate excretion time information, excretion shape information, excretion frequency information, and excretion frequency information according to the above-mentioned captured picture subsequence. foreign body information, excretion location information; a third generating unit 205, configured to generate daily activity behavior information for the above-mentioned target cat according to at least one shooting picture sequence, wherein there is a corresponding relationship between the above-mentioned at least one shooting picture sequence and at least one camera device; a fourth generating unit 206, configured to input the above-mentioned at least one behavior information, the above-mentioned at least one behavior frequency information, the above-mentioned excretion time information, the above-mentioned excretion shape information, the above-mentioned excretion foreign body information, the above-mentioned excretion location information, and the above-mentioned daily activity behavior information into a logistic regression model and a tree model to generate a first urine retention detection result and a second urine retention detection result; a fifth generating unit 207, configured to generate an actual urine retention detection result according to the above-mentioned first urine retention detection result and the above-mentioned second urine retention detection result; a sending unit 208, configured to send the cat information corresponding to the above-mentioned target cat and the above-mentioned actual urine retention detection result to the user terminal of the target care object in response to determining that the above-mentioned actual urine retention detection result indicates the presence of urine retention symptoms.

[0110] It can be understood that the units recorded in the urine retention detection result sending device 200 are similar to those in the reference Figure 1 Therefore, the operations, features and beneficial effects described above for the method are also applicable to the urine retention detection result sending device 200 and the units contained therein, and will not be described in detail here.

[0111] Reference below Figure 3 , which shows a structural schematic diagram of an electronic device (eg, an electronic device) 300 suitable for implementing some embodiments of the present disclosure. Figure 3The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0112] like Figure 3 As shown, the electronic device 300 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 301, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from a storage device 308 into a random access memory (RAM) 303. In the RAM 303, various programs and data required for the operation of the electronic device 300 are also stored. The processing device 301, the ROM 302, and the RAM 303 are connected to each other via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0113] Typically, the following devices may be connected to the I / O interface 305: input devices 306 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; output devices 307 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 308 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 309. The communication devices 309 may allow the electronic device 300 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 3 The electronic device 300 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead. Figure 3 Each block shown in the figure may represent one device, or may represent multiple devices as required.

[0114] In particular, according to some embodiments of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, some embodiments of the present disclosure include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In some such embodiments, the computer program can be downloaded and installed from the network through the communication device 309, or installed from the storage device 308, or installed from the ROM 302. When the computer program is executed by the processing device 301, the above-mentioned functions defined in the method of some embodiments of the present disclosure are executed.

[0115] It should be noted that the computer-readable medium in some embodiments of the present disclosure may be a computer-readable signal medium or a computer-readable storage medium or any combination of the two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In some embodiments of the present disclosure, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In some embodiments of the present disclosure, the computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0116] In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (HyperText Transfer Protocol), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0117] The computer-readable medium may be included in the electronic device; or it may exist independently without being installed in the electronic device. The computer-readable medium carries one or more programs. When the one or more programs are executed by the electronic device, the electronic device: obtains a set of captured picture sequences corresponding to the target cat in the target time period, wherein the captured picture sequences in the captured picture sequence set correspond to the cameras in the camera set, and the camera set includes: at least one camera placed in the activity area corresponding to the target cat, and a target camera placed in the litter box; based on the target captured picture sequence, generates at least one behavior information and corresponding at least one behavior frequency information for the target cat, wherein the target captured picture sequence is a captured picture sequence corresponding to the target camera; in response to determining that the at least one behavior information includes excretion behavior information, filters out a captured picture subsequence associated with the excretion behavior information from the target captured picture sequence; based on the above-mentioned camera A picture subsequence is used to generate defecation time information, defecation shape information, defecation foreign body information, and defecation position information; according to at least one shooting picture sequence, daily activity behavior information for the target cat is generated, wherein there is a corresponding relationship between the at least one shooting picture sequence and the at least one camera device; the at least one behavior information, the at least one behavior frequency information, the defecation time information, the defecation shape information, the defecation foreign body information, the defecation position information, and the daily activity behavior information are input into a logistic regression model and a tree model to generate a first urine retention detection result and a second urine retention detection result; according to the first urine retention detection result and the second urine retention detection result, an actual urine retention detection result is generated; in response to determining that the actual urine retention detection result indicates the presence of urine retention symptoms, the cat information corresponding to the target cat and the actual urine retention detection result are sent to the user terminal of the target care object.

[0118] Computer program code for performing the operations of some embodiments of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0119] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0120] The units described in some embodiments of the present disclosure may be implemented by software or hardware. The units described may also be arranged in a processor, for example, may be described as: a processor includes an acquisition unit, a first generation unit, a screening unit, a second generation unit, a third generation unit, a fourth generation unit, a fifth generation unit and a sending unit. The names of these units do not constitute limitations on the units themselves in certain circumstances, for example, the acquisition unit may also be described as a "unit for acquiring a sequence set of photographed images corresponding to a target cat in a target time period".

[0121] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0122] The above descriptions are only some preferred embodiments of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present disclosure is not limited to the technical solutions formed by a specific combination of the above-mentioned technical features, but should also cover other technical solutions formed by any combination of the above-mentioned technical features or their equivalent features without departing from the above-mentioned inventive concept. For example, the above-mentioned features are replaced with the technical features with similar functions disclosed in the embodiments of the present disclosure (but not limited to) and the technical solutions formed.

Claims

1. A method for sending urine retention detection results, comprising: Acquire a set of shot picture sequences corresponding to a target cat in a target time period, wherein a shot picture sequence in the shot picture sequence set corresponds to a camera in a camera set, and the camera set includes: at least one camera placed in an activity area corresponding to the target cat and a target camera placed in a cat litter box; Generate at least one behavior information and corresponding at least one behavior frequency information for the target cat according to a target shot picture sequence, wherein the target shot picture sequence is a shot picture sequence corresponding to the target camera device; In response to determining that the at least one behavior information includes excretion behavior information, filtering out a shot picture subsequence associated with the excretion behavior information from the target shot picture sequence; Generate defecation time information, defecation shape information, defecation foreign body information, and defecation position information according to the captured picture subsequence; Generate daily activity behavior information for the target cat according to at least one shooting picture sequence, wherein there is a corresponding relationship between the at least one shooting picture sequence and at least one camera device; Inputting the at least one behavior information, the at least one behavior frequency information, the excretion time information, the excretion shape information, the excretion foreign body information, the excretion location information, and the daily activity behavior information into a logistic regression model and a tree model to generate a first urine retention detection result and a second urine retention detection result; generating an actual urine retention detection result according to the first urine retention detection result and the second urine retention detection result; In response to determining that the actual urine retention detection result indicates the presence of urine retention symptoms, cat information corresponding to the target cat and the actual urine retention detection result are sent to a user terminal of the target care object.

2. The method according to claim 1, wherein: The step of generating an actual urine retention detection result according to the first urine retention detection result and the second urine retention detection result comprises: In response to determining that both the first urine retention detection result and the second urine retention detection result indicate the presence of a urine retention symptom, generating an actual urine retention detection result indicating the presence of a urine retention symptom; In response to determining that both the first urine retention test result and the second urine retention test result indicate that there is no urine retention symptom, generating an actual urine retention test result indicating that no urine retention symptom is detected; In response to determining that the detection results between the first urine retention detection result and the second urine retention detection result are inconsistent, inputting each target shot image in the target shot picture sequence into a urine retention detection neural network to generate a third urine retention detection result; An actual urine retention detection result is generated according to the first urine retention detection result, the second urine retention detection result, and the third urine retention detection result.

3. The method according to claim 1, wherein: The method further comprises: Extracting features from each shot in the shot subsequence to generate shot feature information, thereby obtaining a shot feature information subsequence; For each shot picture sequence in the at least one shot picture sequence, the following processing steps are performed: Remove the shot pictures in which the target cat does not exist in the corresponding picture content from the shot picture sequence to obtain a shot picture subsequence after removal; generating a subsequence of captured picture feature information for the removed captured picture subsequence; sorting the obtained subsequences of the captured picture feature information for the removed captured picture subsequence according to the time sequence relationship to obtain a sorted captured picture feature information sequence; Inputting the photographic picture feature information in the sorted photographic picture feature information sequence into a cat event detection generation model to generate at least one cat event information; Generate a cat urine retention detection reason for the target cat according to the at least one cat event information; The cat urine retention detection reason is sent to the user terminal.

4. The method according to claim 1, wherein: The step of generating at least one behavior information and corresponding at least one behavior frequency information for the target cat according to the target shot picture sequence includes: Acquire a preset picture group number set, wherein there is a geometric relationship between the picture group numbers in the picture group number set; For each picture group number in the picture group number set, the following first generating step is performed: According to the number of picture groups, each shot picture in the target shot picture sequence is divided to generate a shot picture group sequence, wherein the number of shot pictures included in each shot picture group is the number of picture groups; For each shot group in the shot group sequence, the following second generating step is performed: Combining the captured pictures according to the picture sequence corresponding to the captured pictures in the captured picture group to generate a combined picture; Acquire a first convolution layer group corresponding to the combined picture, wherein each first convolution layer in the first convolution layer group is connected in series, and a dimension of a convolution kernel corresponding to each first convolution layer is set according to a picture dimension corresponding to the combined picture; Inputting the combined picture into the first convolutional layer group to generate combined picture feature information; For each combined picture feature information sequence in the obtained combined picture feature information sequence set, an initial behavior information group and a corresponding initial behavior probability information group are generated according to the combined picture feature information sequence using a behavior detection model based on multi-head behavior outputs; and based on the obtained initial behavior information group set and initial behavior probability information group set, at least one behavior information and at least one corresponding behavior frequency information are generated.

5. The method according to claim 4, wherein: The generating of the initial behavior information group and the corresponding initial behavior probability information group based on the combined picture feature information sequence and the behavior detection model based on multi-head behavior outputs comprises: Inputting each target shot picture in the target shot picture sequence into a second convolutional layer group to output picture feature information to obtain a picture feature information sequence, wherein each second convolutional layer in the second convolutional layer group is connected in series, and the dimension of the convolution kernel corresponding to each second convolutional layer is set according to the picture dimension corresponding to the target shot picture; The picture feature information sequence and the combined picture feature information sequence are input into the behavior detection model to generate the initial behavior information group and the corresponding initial behavior probability information group.

6. The method according to claim 2, wherein: The generating an actual urine retention detection result according to the first urine retention detection result, the second urine retention detection result and the third urine retention detection result comprises: Determine at least one first result corresponding to the presence of a urinary retention symptom and at least one second result corresponding to the absence of a urinary retention symptom from the first urinary retention detection result, the second urinary retention detection result, and the third urinary retention detection result; In response to determining that the number of results corresponding to the at least one first result is greater than the number of results corresponding to the at least one second result, generating an actual urine retention detection result indicating that a urine retention symptom is detected; In response to determining that the number of results corresponding to the at least one first result is less than the number of results corresponding to the at least one second result, an actual urine retention detection result indicating that no urine retention symptom is detected is generated.

7. The method according to claim 3, wherein: The step of generating a cat urine retention detection reason for the target cat according to the at least one cat event information comprises: Determine the urine retention influence probability information corresponding to each cat event information in the at least one cat event information to obtain at least one urine retention influence probability information; A cat urine retention detection reason is generated according to the at least one urine retention influence probability information.

8. A device for sending urine retention detection results, comprising: An acquisition unit is configured to acquire a set of shot picture sequences corresponding to a target cat in a target time period, wherein a shot picture sequence in the shot picture sequence set corresponds to a camera in a camera set, and the camera set includes: at least one camera placed in an activity area corresponding to the target cat, and a target camera placed in a cat litter box; A first generating unit is configured to generate at least one behavior information and corresponding at least one behavior frequency information for the target cat according to a target shot picture sequence, wherein the target shot picture sequence is a shot picture sequence corresponding to the target camera device; a screening unit, configured to, in response to determining that the at least one behavior information includes excretion behavior information, screen out a shot picture subsequence associated with the excretion behavior information from the target shot picture sequence; A second generating unit is configured to generate excretion time information, excretion shape information, excretion foreign body information, and excretion position information according to the captured picture subsequence; A third generating unit is configured to generate daily activity behavior information for the target cat according to at least one shooting picture sequence, wherein there is a corresponding relationship between the at least one shooting picture sequence and at least one camera device; a fourth generating unit configured to input the at least one behavior information, the at least one behavior frequency information, the excretion time information, the excretion shape information, the excretion foreign body information, the excretion location information, and the daily activity behavior information into a logistic regression model and a tree model to generate a first urine retention detection result and a second urine retention detection result; a fifth generating unit, configured to generate an actual urine retention detection result according to the first urine retention detection result and the second urine retention detection result; The sending unit is configured to send the cat information corresponding to the target cat and the actual urine retention detection result to the user terminal of the target care object in response to determining that the actual urine retention detection result indicates the presence of urine retention symptoms.

9. An electronic device, comprising: one or more processors; a storage device having one or more programs stored thereon, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 7.

10. A computer readable medium having a computer program stored thereon, wherein: When the program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images