Monitoring method, device and equipment of breeding place and medium

Abstract

The invention discloses a monitoring method and device for a breeding place, equipment and a storage medium. Comprising the following steps: acquiring a to-be-monitored sound signal of a breeding place, and extracting a target voiceprint feature in the to-be-monitored sound signal; based on a pre-established blacklist sample library, performing comparison processing on the target voiceprint features and voiceprint features in the blacklist sample library to obtain a comparison result; and according to a comparison result, determining an alarm result monitored by the farm. According to the invention, after the target voiceprint feature in the to-be-monitored sound signal is compared with the voiceprint feature in the blacklist sample library, whether the farm needs to give an alarm is judged, so that the situation that the monitoring equipment of the existing farm is shielded by other objects, so that a visual blind area is caused, and the monitoring efficiency is improved is avoided. Subsequent observation and rapid treatment of harmful organisms which are hidden in positions and influence livestock growth are achieved, the working efficiency is greatly improved, and resources are saved.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a monitoring method, device, equipment and storage medium for a breeding site. Background Art

[0002] With the continuous progress of the modern economy and the modernization of agriculture and animal husbandry, Internet of Things technology has received more and more attention in agricultural and animal husbandry production, especially in poultry farms.

[0003] Most of the existing methods of monitoring breeding farms are carried out manually, usually by installing surveillance cameras in the breeding farms, and then the breeding staff monitor in the monitoring room. However, this requires 24-hour manual supervision and a large number of cameras to cover all blind spots, which will inevitably consume a lot of manpower and material resources. In addition, due to the complexity of the breeding environment, it is difficult for the staff to timely discover the safety hazards faced by the breeding objects in the breeding farms through monitoring, making it impossible for the staff to process and feedback in real time, which will lead to low work efficiency and a large number of poultry losses, thereby causing unnecessary losses. Therefore, how to control monitoring to solve the problem that the existing breeding farm monitoring equipment is blocked by other objects, resulting in blind spots, thereby causing safety hazards faced by the breeding objects in the breeding farms, has become a technical problem that needs to be solved urgently by technical personnel in this field. Summary of the invention

[0004] Based on this, it is necessary to provide a monitoring method, device, equipment and storage medium for a breeding site in response to the above-mentioned technical problems, so as to solve the problem that the existing breeding site monitoring equipment is blocked by other objects, resulting in blind spots in the field of vision, thereby causing safety hazards to the breeding objects in the breeding site.

[0005] A first aspect of an embodiment of the present application provides a method for monitoring a breeding site, the method comprising: Acquire the sound signal to be monitored in the breeding place, and extract the target voiceprint feature in the sound signal to be monitored; Based on a pre-established blacklist sample library, the target voiceprint feature is compared with the voiceprint features in the blacklist sample library to obtain a comparison result; According to the comparison result, the alarm result of the breeding site monitoring is determined.

[0006] A second aspect of an embodiment of the present application provides a monitoring device for a breeding site, the monitoring device for a breeding site comprising: An acquisition module, used to acquire the sound signal to be monitored in the breeding site, and extract the target voiceprint feature in the sound signal to be monitored; A comparison module, for comparing the target voiceprint features with the voiceprint features in the blacklist sample library based on a pre-established blacklist sample library to obtain a comparison result; A determination module is used to determine the alarm result of the breeding site monitoring according to the comparison result.

[0007] In a third aspect, an embodiment of the present invention provides a computer device, comprising a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method for monitoring a breeding site as described in the first aspect when executing the computer program.

[0008] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for monitoring a breeding site as described in the first aspect is implemented.

[0009] In summary, the present invention provides a monitoring method, device, equipment and storage medium for a breeding site, which obtains the sound signal to be monitored in the breeding site, extracts the target voiceprint features in the sound signal to be monitored, and compares the target voiceprint features with the voiceprint features in the blacklist sample library based on a pre-established blacklist sample library to obtain a comparison result, and determines the alarm result of the breeding site monitoring according to the comparison result. In the present invention, by comparing the target voiceprint features in the sound signal to be monitored with the voiceprint features in the blacklist sample library, it is determined whether the breeding site needs to be alarmed, avoiding the blind spots in the field of vision caused by the existing breeding site monitoring equipment being blocked by other objects, and realizing the subsequent observation and rapid processing of some relatively hidden harmful organisms that affect livestock reproduction, greatly improving work efficiency and saving resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.

[0011] Figure 1 This is a schematic diagram of an application environment of a monitoring method for a breeding site provided by an embodiment of the present invention; Figure 2 It is a flow chart of a method for monitoring a breeding site provided by one embodiment of the present invention; Figure 3 It is a partial flow diagram of a method for monitoring a breeding site provided by an embodiment of the present invention; Figure 4 It is a structural schematic diagram of a monitoring device for a breeding site provided by an embodiment of the present invention; Figure 5 It is a structural schematic diagram of a computer device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0012] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0013] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0014] It should also be understood that the term "and / or" used in the present description and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0015] As used in the present specification and the appended claims, the term "if" may be interpreted as "when" or "upon" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" may be interpreted as meaning "upon determination" or "in response to determining" or "upon detection of [described condition or event]" or "in response to detecting [described condition or event]", depending on the context.

[0016] In addition, in the description of the present specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0017] References to "one embodiment" or "some embodiments" etc. described in the present specification mean that one or more embodiments of the present invention include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0018] It should be understood that the order of execution of the steps in the following embodiments does not imply a precedence of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0019] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Artificial Intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results. AI basic technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing technology, operation / interaction systems, mechatronics, etc. AI software technologies mainly include computer vision technology, robotics technology, biometrics technology, speech processing technology, natural language processing technology, and machine learning / deep learning.

[0020] In order to illustrate the technical solution of the present invention, specific embodiments are provided below for illustration.

[0021] A monitoring method for a breeding site provided by an embodiment of the present invention can be applied in Figure 1 In an application environment, a client communicates with a server. The client includes, but is not limited to, a PDA, a desktop computer, a laptop computer, an ultra-mobile personal computer (UMPC), a netbook, a personal digital assistant (PDA), and other computer devices. The server can be implemented with an independent server or a server cluster consisting of multiple servers.

[0022] See also Figure 2, is a flow chart of a method for monitoring a breeding site provided by an embodiment of the present invention. The method for monitoring a breeding site can be applied to Figure 1 The server in the above connects to the corresponding client, such as Figure 2 As shown, the monitoring method of the breeding site may include the following steps.

[0023] S201: Acquire a sound signal to be monitored in a breeding site, and extract a target voiceprint feature from the sound signal to be monitored.

[0024] In step S201, the sound signal to be monitored in the breeding site is obtained through the sound receiver, and the sound receiver transmits the received sound signal to be monitored in the breeding site to the sound processing center through the network, and the sound processing center extracts the target voiceprint feature in the sound signal to be monitored.

[0025] It should be noted that, generally speaking, to obtain the sound signal to be monitored in the breeding site, one sound receiver can be set up, or multiple sound receivers can be set up. According to the performance parameters of the sound receiver, a suitable sound receiver can be selected under different environmental conditions to collect the sound signal to be monitored. No specific limitation is made here.

[0026] Optionally, extracting target voiceprint features from the sound signal to be monitored includes: Sending the sound signal to be monitored of the breeding site to a preset sound processing center to obtain the sound category of the sound signal to be monitored; The sound category is analyzed to determine the target voiceprint feature category in the sound signal to be monitored.

[0027] Specifically, the sound signal to be monitored in the breeding site is obtained, and the sound signal to be monitored may be the sound information of harmful organisms that affect livestock reproduction, or it may be the sound information of livestock in the breeding site. The current sound signal to be monitored in the breeding site is obtained through a sound receiver, and then the current sound signal to be monitored in the breeding site is sent to a preset sound processing center to obtain the sound category of the current sound signal to be monitored, and its sound category is analyzed to determine the target voiceprint feature category in the current sound signal to be monitored. In order to accurately identify the sound information of harmful organisms that affect livestock reproduction or the sound information of livestock in the breeding site, the preset sound processing center is connected to the monitoring center for communication. The sound processing center is used to identify and judge the sound information of harmful organisms that affect livestock reproduction or the sound information of livestock in the breeding site to determine whether an alarm is needed.

[0028] When monitoring a breeding site, the embodiment of the present application can identify different sound signals through a single deployed sound processing center, thereby improving the fertility of livestock in the breeding site and effectively achieving the purpose of saving energy.

[0029] It should be noted that, after setting up a sound processing center in an office environment such as a monitoring room, the sound processing center can provide voiceprint feature extraction, voiceprint feature storage and voiceprint feature comparison, and the sound processing center can use a centralized server for processing voice signals, so that the sound processing center can process all sound information in its coverage area. And the sound processing center can be connected to multiple sound receivers, that is, it can simultaneously receive the sound signals to be monitored sent by multiple sound receivers.

[0030] Specifically, after receiving the sound signal to be monitored sent by the sound receiver, the sound processing center performs sound recognition on the sound signal to be monitored, and analyzes the sound category of the recognized sound signal to be monitored, thereby extracting the target voiceprint feature category in the sound signal to be monitored.

[0031] It should be noted that in some application scenarios, due to obstruction by other objects, there are blind spots in the field of vision and traditional monitoring equipment cannot be used. However, the embodiment of the present application uses a sound processing center to screen and judge the sound signals to be monitored by the sound receiver, thereby improving the ability to recognize sound information, avoiding potential safety hazards faced by breeding objects in breeding facilities, and is better suitable for working environments with more blind spots in the field of vision in monitoring equipment.

[0032] S202: Based on a pre-established blacklist sample library, the target voiceprint feature is compared with the voiceprint features in the blacklist sample library to obtain a comparison result.

[0033] In step S202, a blacklist sample library is first established in advance, and the current target voiceprint features are compared with the voiceprint features in the blacklist sample library. The comparison result is obtained according to the comparison score and the set threshold. After judging the current comparison result, step S203 is executed to determine the alarm result of the breeding site monitoring. If an alarm is required, the sound processing center notifies the on-duty personnel. If an alarm is not required, the sound processing center will not notify the on-duty personnel.

[0034] In the embodiment of the present invention, several sections of speech data affecting livestock reproduction are collected in advance as training speech samples, and different speech data correspond to different user identities. The training speech samples affecting livestock reproduction are trained using the GMM model (Gaussian Mixture Model), and a voiceprint feature category model for the training speech samples affecting livestock reproduction is obtained, and a preset blacklist sample library is composed of the obtained different voiceprint feature category models. The extracted target voiceprint features are compared with the voiceprint feature parameters corresponding to each voiceprint feature category model in the established preset blacklist sample library to obtain a comparison result. In the specific implementation process, taking into account that the voice of each individual object may change within a certain period of time, in order to improve the comparison effect, the comparison degree can be pre-set as a predetermined threshold. When judging whether the comparison degree between the extracted target voiceprint feature and the voiceprint feature parameters corresponding to each voiceprint feature category model in the established preset blacklist sample library exceeds the predetermined threshold, the current comparison result is determined based on the judgment and comparison. For example, if the matching degree between the current target voiceprint feature and the voiceprint feature parameter A in the preset blacklist sample library reaches 90% or more and exceeds the predetermined threshold, it is determined that the current target voiceprint feature corresponds to the voiceprint feature in the preset blacklist sample library.

[0035] It should be noted that the voice data that affects livestock reproduction may be the sounds made by livestock when they are frightened, or the sounds made by livestock's natural enemies, etc. This application does not make any specific limitations here.

[0036] Optionally, the target voiceprint feature is compared with the voiceprint feature in the preset blacklist sample library to obtain a comparison result, including: Acquire the target voiceprint features in the sound signal to be monitored and all voiceprint features in the blacklist sample library; Comparing and scoring the similarities between the target voiceprint feature and all the voiceprint features to obtain a target score set; In the case where there is at least one score value greater than or equal to a preset threshold in the target score set, obtaining the comparison result as a successful comparison; When there is no score value greater than or equal to the preset threshold in the target score set, the comparison result is obtained as unsuccessful comparison.

[0037] Optionally, in this embodiment, the target voiceprint feature vector of the first target voiceprint feature of the customer and all voiceprint feature vectors in the blacklist sample library can be extracted through a deep neural network, and the similarity between the target voiceprint feature vector and all voiceprint feature vectors can be scored to obtain a target score set. For example, if there are 5 voiceprint data in the blacklist sample library, 5 voiceprint feature vectors will be obtained. The first voiceprint feature vector, the second voiceprint feature vector, the third voiceprint feature vector, the fourth voiceprint feature vector, and the fifth voiceprint feature vector are scored for similarity respectively to obtain a similarity score of 80 between the first voiceprint feature vector and the target voiceprint feature vector. By analogy, a score set of [80, 55, 38, 71, 98] is obtained. The threshold can be selected as 95. If there is at least one score value 98 greater than or equal to the threshold value 95 in the target score set, the comparison result is a successful comparison. If the score set is [80, 52, 48, 68, 90], and there is no score value greater than or equal to the threshold value 95 in the target score set, the comparison result is an unsuccessful comparison.

[0038] Optionally, the target voiceprint feature and all voiceprint features are compared and scored for similarity to obtain a target score set, including: Determining whether the target voiceprint feature and all the voiceprint features belong to the same category; If the target voiceprint feature and all the voiceprint features belong to the same category, compare and score the similarities between the target voiceprint feature and all the voiceprint features to obtain an original score set; Each score value in the original score set is sorted from small to large, and multiplied by 100 to obtain a target score set.

[0039] Optionally, in this embodiment, it is necessary to determine whether the target voiceprint feature and all voiceprint features belong to the same category. If the target voiceprint feature and all voiceprint features belong to the same category, the similarity between the target voiceprint feature vector and all voiceprint feature vectors is scored to obtain an original score set, and each score value in the original score set is sorted from small to large, and multiplied by 100 to obtain a target score set. For example, if there are 5 voiceprint data in the voiceprint database, 5 voiceprint feature vectors will be obtained, and the first voiceprint feature vector, the second voiceprint feature vector, the third voiceprint feature vector, the fourth voiceprint feature vector, and the fifth voiceprint feature vector are scored for similarity respectively, and the similarity score between the first voiceprint feature vector and the target voiceprint feature vector is 0.8, thereby deriving an original score set [0.80, 0.55, 0.38, 0.71, 0.98], and each score value in the original score set is sorted from small to large, and multiplied by 100 to obtain a target score set [38, 55, 71, 80, 98]. Therefore, if different types of sounds are compared, the score will be very low, and if the same type of sounds are compared, the score will be very high, so as long as the appropriate threshold is set, they can be correctly distinguished.

[0040] It should be noted that the blacklist sample library does not necessarily contain 5 voiceprint data, but may contain 10, 50 or 100 voiceprint data, and the more voiceprint data there are in the blacklist sample library, the more accurate the comparison result will be. That is, the voiceprint data in the blacklist sample library is set according to the actual situation, and this application does not make any specific limitations here.

[0041] In this embodiment, the target voiceprint features are compared with the voiceprint features in the blacklist sample library through a pre-established blacklist sample library to obtain a comparison result, which can more quickly and accurately determine whether the current target voiceprint features are sounds that affect livestock reproduction, thereby achieving the purpose of clearly dividing the target voiceprint features in the monitored sound signal into categories and timely warnings, thereby solving the technical problems of potential safety hazards faced by breeding objects and improving the accuracy of monitoring.

[0042] S203: Determine the alarm result of the breeding site monitoring according to the comparison result.

[0043] In step S203, by comparing the target voiceprint features with the voiceprint features in the blacklist sample library, a corresponding comparison result is obtained, and according to the corresponding comparison result, it is determined whether the monitoring center needs to alarm, so as to determine the alarm result of the breeding site monitoring. When the voice-controlled alarm of the monitoring center receives sound information that affects livestock reproduction, the received sound information is sent to the sound processing center. After determining that the received sound belongs to the sound category that affects livestock reproduction, the voice-controlled alarm can send an alarm signal to notify the on-duty personnel to avoid the occurrence of serious dangers.

[0044] Optionally, the alarm results of the breeding site monitoring are determined by comparing the results, including: If the comparison result is successful, an alarm message is sent to the monitoring center to determine that the alarm result of the breeding site monitoring is an alarm; If the comparison result is that the comparison is unsuccessful, no alarm information is sent to the monitoring center, and the alarm result of the breeding site monitoring is determined to be no alarm.

[0045] Optionally, the comparison result may be the degree of similarity between the target voiceprint feature vector corresponding to the sound signal to be monitored and all voiceprint feature vectors in the preset blacklist sample library. If the similarity represented by the comparison result is greater than or equal to a preset threshold, it can be determined that the sound signal to be monitored and the sound signal on the established blacklist are voices emitted by the same object, and are sound signals that affect livestock reproduction; correspondingly, if the similarity represented by the comparison result is less than a preset threshold, it can be determined that the sound signal to be monitored and the sound signal on the established blacklist are voices emitted by different objects, and are sound signals of breeding objects in a breeding facility.

[0046] In this embodiment, if the comparison result is successful, that is, there is a sound signal that affects livestock reproduction, then an alarm message is sent to the monitoring center, and the alarm result of the breeding site monitoring is determined to be an alarm. If the comparison result is unsuccessful, that is, there is a sound signal that affects livestock reproduction, and it belongs to the sound signal of the breeding object in the breeding site, then no alarm message is sent to the monitoring center, and the alarm result of the breeding site monitoring is determined to be no alarm.

[0047] In the monitoring method for a breeding farm provided in an embodiment of the present application, by comparing the target voiceprint features with the voiceprint features in the blacklist sample library, a corresponding comparison result is obtained, and the alarm result corresponding to the monitoring of the breeding farm is determined according to the comparison result. Compared with traditional video surveillance, voiceprint monitoring is not blocked by vision. Different blacklist sound signals can be added for different application scenarios, and compared with traditional video surveillance, the entire breeding farm can be monitored without specifying a location. The monitoring method for the breeding farm thus achieves the purpose of a large monitoring range, wide application and simple deployment.

[0048] In summary, the present invention provides a monitoring method, device, equipment and storage medium for a breeding site, which obtains the sound signal to be monitored in the breeding site, extracts the target voiceprint features in the sound signal to be monitored, and compares the target voiceprint features with the voiceprint features in the blacklist sample library based on a pre-established blacklist sample library to obtain a comparison result, and determines the alarm result of the breeding site monitoring according to the comparison result. In the present invention, by comparing the target voiceprint features in the sound signal to be monitored with the voiceprint features in the blacklist sample library, it is determined whether the breeding site needs to be alarmed, avoiding the blind spots in the field of vision caused by the existing breeding site monitoring equipment being blocked by other objects, and realizing the subsequent observation and rapid processing of some relatively hidden harmful organisms that affect livestock reproduction, greatly improving work efficiency and saving resources.

[0049] Based on Figure 1 In the specific implementation, step S202 can be implemented by the following steps: Figure 3 As shown: S301, collecting at least two sound samples for training the influencing livestock fertility for sample training.

[0050] In an embodiment of the present invention, a user can record voice data affecting livestock reproduction through a sound receiver on a client, and a server obtains the voice data affecting livestock reproduction input by the user through communication with the client as a training voice sample. Specifically, at least two different voice data affecting livestock reproduction need to be recorded as training voice samples.

[0051] S302, pre-processing each sound sample of the training to influence livestock reproduction, and extracting voiceprint features of each sound sample of the training to influence livestock reproduction after the pre-processing.

[0052] In an embodiment of the present invention, before extracting the sound samples of each training affecting livestock reproduction, it is necessary to preprocess the sound samples of each training affecting livestock reproduction. Due to the aliasing, high-order harmonic distortion, high frequency and other factors brought by the equipment for collecting voice data, the quality of the voice signal is affected. Through preprocessing, a more uniform and smooth voice signal can be obtained, and low-frequency interference in the voice data can be removed, providing high-quality feature parameters for the extraction of voiceprint features, thereby improving the quality of voice processing.

[0053] Specifically, the preprocessing of the sound samples of the training that affect livestock reproduction is divided into four steps: sampling quantization, zero drift removal, pre-emphasis and windowing. Among them, sampling quantization processing refers to filtering the voice signal of the sound sample of the training that affects livestock reproduction with a sharp cutoff filter so that its Nyquist frequency FN is 4KHZ; zero drift removal refers to calculating the average value of the quantized amplitude sequence, and subtracting the average value from each amplitude value to obtain an amplitude sequence with an average value of 0 after zero drift removal; pre-emphasis refers to setting the pre-emphasis coefficient in the transfer function of the digital filter, and obtaining an amplitude sequence with equivalent high, medium and low frequency amplitudes of the voice signal through the digital filter; windowing refers to using the Hamming window function to perform Hamming window processing on each voice signal frame.

[0054] In the embodiment of the present invention, the so-called voiceprint is a sound wave spectrum carrying speech information displayed by an electroacoustic instrument. The voiceprint features include one or more of acoustic features, lexical features, prosodic features and language features; this embodiment implements the training of different voiceprint models based on the voiceprint features of different training sound samples affecting livestock fertility, so it is necessary to extract the voiceprint features of different training sound samples affecting livestock fertility from the training sound samples in advance and save them accordingly; specifically, extracting corresponding voiceprint features from different training sound samples affecting livestock fertility means that the voiceprint extracted and selected from the sound has acoustic, lexical, prosodic or language features with strong separability and high stability. The extraction of voiceprint features can be achieved through the extraction methods such as kaldi's mfcc or pytorch's fbank, and the embodiment of the present invention will not be described in detail.

[0055] It should be noted that the voiceprint features extracted from the sound samples used for training to influence livestock fertility may be one or more.

[0056] S303, normalizing the voiceprint features of each training sound sample affecting livestock reproduction, and determining different voiceprint feature categories of each training sound sample affecting livestock reproduction.

[0057] In the specific implementation, in order to determine the type of voiceprint features of the received training sound samples affecting livestock fertility, this example needs to classify the training sound samples affecting livestock fertility according to the preset classification standard of livestock fertility sounds to obtain the classification result. That is, by normalizing the voiceprint features of each training sound sample affecting livestock fertility, different voiceprint feature categories of each training sound sample affecting livestock fertility are determined. This step requires enough training sound samples affecting livestock fertility, so as to improve the accuracy of the normalized voiceprint features.

[0058] It should be noted that the preset classification standard for sounds affecting livestock reproduction in the embodiment of the present application divides sound information into sound information including thunder sound information, rain sound information, sound information emitted by livestock's natural enemies, machine engine sound information, sound information emitted when livestock are frightened, and other sound information that affects livestock reproduction.

[0059] Furthermore, the embodiments of the present application can set different classification standards for sounds that affect livestock reproduction according to different breeding sites and regions, so as to obtain different classification results, and can more effectively, directly and specifically determine the different voiceprint feature categories of each training sound sample that affects livestock reproduction, thereby ensuring the accuracy of different voiceprint feature categories in the blacklist sample library.

[0060] Optionally, normalizing the voiceprint features of each training sound sample affecting livestock fertility to determine different voiceprint feature categories of each training sound sample affecting livestock fertility includes: Comparing the voiceprint features of every two training sound samples affecting livestock reproduction, and determining whether the similarity of the voiceprint features of every two training sound samples affecting livestock reproduction is greater than a preset similarity threshold; If the similarity of the voiceprint features of every two training sound samples affecting livestock fertility is greater than a preset similarity threshold, it is determined that the voiceprint features of every two training sound samples affecting livestock fertility belong to the same category; Until all voiceprint features of the sound samples for training to influence livestock reproduction are gone, different voiceprint feature categories of the sound samples for training to influence livestock reproduction are obtained.

[0061] In the embodiment of the present invention, the voiceprint features of the same training sound sample affecting livestock fertility are trained with a GMM model (Gaussian Mixture Model) to establish a corresponding voiceprint model. More specifically, the speech signal frames corresponding to all the voiceprint features belonging to the same training sound sample affecting livestock fertility can be determined first, and then all the voice signal frames belonging to the same training sound sample affecting livestock fertility can be trained to obtain the corresponding GMM model to compare the voiceprint features of every two training sound samples affecting livestock fertility, and determine whether the similarity of the voiceprint features of every two training sound samples affecting livestock fertility is greater than a preset similarity threshold. If the similarity of the voiceprint features of every two training sound samples affecting livestock fertility is greater than the preset similarity threshold, it is determined that the voiceprint features of every two training sound samples affecting livestock fertility belong to the same category, until all the voiceprint features of the training sound samples affecting livestock fertility are gone, thereby calculating the similarity of the voiceprint features of each training sound sample affecting livestock fertility in the blacklist sample library, and obtaining different voiceprint feature categories of each training sound sample affecting livestock fertility. In one embodiment, if there are multiple voiceprint features of the sound samples for training to affect livestock fertility in the blacklist sample library, a K-means clustering algorithm is used to calculate the voiceprint similarity between the voiceprint feature of the sound sample for training to affect livestock fertility with the highest priority and a preset similarity threshold.

[0062] S304, constructing the blacklist sample library according to different voiceprint feature categories of each training sound sample affecting livestock reproduction.

[0063] Furthermore, in this embodiment, after obtaining different voiceprint feature categories of each training sound sample affecting livestock reproduction, a blacklist sample library is constructed through the different voiceprint feature categories of each training sound sample affecting livestock reproduction.

[0064] In the embodiment of the present invention, enough sound samples that affect the reproduction of livestock are obtained as much as possible to build a blacklist sample library. This will enable more accurate comparison results to be obtained when the target voiceprint features are compared with the voiceprint features in the blacklist sample library. Based on the comparison results, it is determined whether the breeding site needs to be alarmed, which greatly improves work efficiency and saves resources to achieve the purpose of monitoring.

[0065] See also Figure 4 , Figure 4 Schematic diagram of the structure of a monitoring device for a breeding site provided by an embodiment of the present invention. In this embodiment, the terminal includes various units for executing Figure 2 For details, please refer to the steps in the corresponding embodiment. Figure 2 as well as Figure 2For the convenience of explanation, only the parts related to this embodiment are shown. Figure 4 The monitoring device 40 of the breeding site includes: an acquisition module 41, a comparison module 42, and a determination module 43.

[0066] An acquisition module is used to acquire the sound signal to be monitored in the breeding site and extract the target voiceprint feature in the sound signal to be monitored; A comparison module is used to compare the target voiceprint features with the voiceprint features in the blacklist sample library based on a pre-established blacklist sample library to obtain a comparison result; The determination module is used to determine the alarm result of the breeding site monitoring according to the comparison result.

[0067] Optionally, the acquisition module 41 is specifically used for: Sending the sound signal to be monitored in the breeding site to a preset sound processing center to obtain the sound category of the sound signal to be monitored; Analyze the sound category and determine the target voiceprint feature category in the sound signal to be monitored.

[0068] Optionally, the comparison module 42 is specifically used for: Collect at least two samples of sounds that affect livestock fertility for sample training; Preprocessing each sound sample of the training to influence livestock reproduction, and extracting voiceprint features of each sound sample of the training to influence livestock reproduction after preprocessing; Normalizing the voiceprint features of each training sound sample affecting livestock reproduction, and determining different voiceprint feature categories of each training sound sample affecting livestock reproduction; A blacklist sample library is constructed based on different voiceprint feature categories of each training sound sample that affects livestock fertility.

[0069] Optionally, the comparison module 42 is further used for: Comparing the voiceprint features of every two training sound samples affecting livestock fertility, and determining whether the similarity of the voiceprint features of every two training sound samples affecting livestock fertility is greater than a preset similarity threshold; If the similarity of the voiceprint features of every two training sound samples affecting livestock fertility is greater than a preset similarity threshold, it is determined that the voiceprint features of every two training sound samples affecting livestock fertility belong to the same category; Until all voiceprint features of the sound samples of each training affecting livestock reproduction are gone, different voiceprint feature categories of the sound samples of each training affecting livestock reproduction are obtained.

[0070] Optionally, the comparison module 42 is further used for: Obtain the target voiceprint features in the sound signal to be monitored and all voiceprint features in the blacklist sample library; Compare and score the similarity between the target voiceprint feature and all voiceprint features to obtain the target score set; When there is at least one score value greater than or equal to a preset threshold in the target score set, the comparison result is obtained as a successful comparison; When there is no score value greater than or equal to the preset threshold in the target score set, the comparison result is that the comparison is unsuccessful.

[0071] Optionally, the comparison module 42 is further used for: Determine whether the target voiceprint feature and all voiceprint features belong to the same category; If the target voiceprint feature and all voiceprint features belong to the same category, the similarity between the target voiceprint feature and all voiceprint features is compared and scored to obtain an original score set; Sort each score value in the original score set from small to large and multiply them by 100 to obtain the target score set.

[0072] Optionally, the determination module 43 is specifically configured to: If the comparison result is successful, an alarm message is sent to the monitoring center to confirm that the alarm result of the breeding site monitoring is an alarm; If the comparison result is unsuccessful, no alarm information will be sent to the monitoring center, and the alarm result of the breeding site monitoring is determined to be no alarm.

[0073] It should be noted that the information interaction, execution process and other contents between the above-mentioned units are based on the same concept as the embodiment of the method of the present invention. Their specific functions and technical effects can be found in the method embodiment part and will not be repeated here.

[0074] Figure 5 Schematic diagram of the structure of a computer device provided by an embodiment of the present invention. Figure 5 As shown, the computer device of this embodiment includes: at least one processor ( Figure 5 Only one is shown), a memory, and a computer program stored in the memory and executable on at least one processor, wherein when the processor executes the computer program, the steps in any of the above-mentioned monitoring method embodiments for each breeding site are implemented.

[0075] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program: Acquire the sound signal to be monitored in the breeding place, and extract the target voiceprint feature in the sound signal to be monitored; Based on the pre-established blacklist sample library, the target voiceprint features are compared with the voiceprint features in the blacklist sample library to obtain a comparison result; Based on the comparison results, the alarm results of the breeding site monitoring are determined.

[0076] The computer device may include, but is not limited to, a processor and a memory. Those skilled in the art will appreciate that Figure 5 This is merely an example of a computer device and does not constitute a limitation on the computer device. The computer device may include more or fewer components than those shown in the figure, or a combination of certain components, or different components. For example, it may also include a network interface, a display screen, and an input device.

[0077] In one embodiment, a computer-readable storage medium is provided. When the instructions in the computer-readable storage medium are executed by a processor in a computer device, the computer device can execute the steps of any embodiment of a method for monitoring a breeding site disclosed in the present invention, which will not be repeated here. The computer-readable storage medium can be non-volatile or volatile.

[0078] The processor may be a CPU, or other general-purpose processors, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0079] The memory includes a readable storage medium, an internal memory, etc., wherein the internal memory may be the memory of a computer device, and the internal memory provides an environment for the operation of the operating system and computer-readable instructions in the readable storage medium. The readable storage medium may be a hard disk of a computer device, and in other embodiments, it may also be an external storage device of the computer device, for example, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the computer device. Further, the memory may also include both an internal storage unit of the computer device and an external storage device. The memory is used to store an operating system, an application program, a boot loader (BootLoader), data, and other programs, such as the program code of a computer program, etc. The memory may also be used to temporarily store data that has been output or is to be output.

[0080] It can be understood by those skilled in the art that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0081] The technical business in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the system is divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here. If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium.

[0082] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, it should be understood by those skilled in the art that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A method for monitoring a breeding site, characterized in that: include: Acquire a sound signal to be monitored at a breeding site, and extract a target voiceprint feature from the sound signal to be monitored; Based on a pre-established blacklist sample library, the target voiceprint feature is compared with the voiceprint features in the blacklist sample library to obtain a comparison result; According to the comparison result, the alarm result of the breeding site monitoring is determined.

2. The method for monitoring a breeding site according to claim 1, characterized in that: The pre-established blacklist sample library includes: Collect at least two samples of sounds that affect livestock fertility for sample training; Preprocessing each of the sound samples of the training to influence livestock reproduction, and extracting voiceprint features of each of the preprocessed sound samples of the training to influence livestock reproduction; Normalizing the voiceprint features of the sound samples of the training to influence livestock reproduction, and determining different voiceprint feature categories of the sound samples of the training to influence livestock reproduction; The blacklist sample library is constructed according to different voiceprint feature categories of the sound samples that are trained to affect livestock fertility.

3. The method for monitoring a breeding site according to claim 2, characterized in that: The normalizing of the voiceprint features of the sound samples of the training to influence livestock reproduction to determine different voiceprint feature categories of the sound samples of the training to influence livestock reproduction includes: Comparing the voiceprint features of every two training sound samples affecting livestock reproduction, and determining whether the similarity of the voiceprint features of every two training sound samples affecting livestock reproduction is greater than a preset similarity threshold; If the similarity of the voiceprint features of every two training sound samples affecting livestock fertility is greater than a preset similarity threshold, it is determined that the voiceprint features of every two training sound samples affecting livestock fertility belong to the same category; Until all voiceprint features of the sound samples for training to influence livestock reproduction are gone, different voiceprint feature categories of the sound samples for training to influence livestock reproduction are obtained.

4. The method for monitoring a breeding site according to claim 1, characterized in that: The comparing the target voiceprint feature with the voiceprint features in the blacklist sample library to obtain a comparison result includes: Acquire the target voiceprint features in the sound signal to be monitored and all voiceprint features in the blacklist sample library; Comparing and scoring the similarities between the target voiceprint feature and all the voiceprint features to obtain a target score set; In the case where there is at least one score value greater than or equal to a preset threshold in the target score set, obtaining the comparison result as a successful comparison; When there is no score value greater than or equal to the preset threshold in the target score set, the comparison result is obtained as unsuccessful comparison.

5. The method for monitoring a breeding site according to claim 4, characterized in that: The similarity between the target voiceprint feature and all the voiceprint features is compared and scored to obtain a target score set, including: Determining whether the target voiceprint feature and all the voiceprint features belong to the same category; If the target voiceprint feature and all the voiceprint features belong to the same category, compare and score the similarities between the target voiceprint feature and all the voiceprint features to obtain an original score set; Each score value in the original score set is sorted from small to large, and multiplied by 100 to obtain a target score set.

6. The method for monitoring a breeding site according to claim 1, characterized in that: Determining the alarm result of the breeding site monitoring according to the comparison result includes: If the comparison result is successful, an alarm message is sent to the monitoring center to determine that the alarm result of the breeding site monitoring is an alarm; If the comparison result is that the comparison is unsuccessful, no alarm information is sent to the monitoring center, and the alarm result of the breeding site monitoring is determined to be no alarm.

7. The method for monitoring a breeding site according to claim 1, characterized in that: The extracting the target voiceprint feature from the sound signal to be monitored includes: Sending the sound signal to be monitored of the breeding site to a preset sound processing center to determine the sound category of the sound signal to be monitored; The sound category is analyzed to extract the target voiceprint feature category in the sound signal to be monitored.

8. A monitoring device for a breeding site, characterized in that: include: An acquisition module, used to acquire the sound signal to be monitored in the breeding site, and extract the target voiceprint feature in the sound signal to be monitored; A comparison module, for comparing the target voiceprint features with the voiceprint features in the blacklist sample library based on a pre-established blacklist sample library to obtain a comparison result; A determination module is used to determine the alarm result of the breeding site monitoring according to the comparison result.

9. A computer device, characterized in that: The computer device includes a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the method for monitoring a breeding site as described in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method for monitoring a breeding site as described in any one of claims 1 to 7 is implemented.

Images