TWI775077B - Feedstock identification method and device - Google Patents

Abstract

The embodiments of the present specification provide a method and device for identification of livestock animals, wherein the method for identification of livestock animals includes: acquiring a first biometric image and a second biometric image of the livestock to be identified, and converting the first biometric image Perform similarity calculation with the benchmark biometric images of multiple animals in the database, take the benchmark biometric image that meets the preset conditions in the calculation result as the target image, and read the biometric code segment in the database. The identity coding sequence, wherein the biometric coding segment corresponds to the target image; extract the biometric information to be identified of the second biometric image, and map the biometric information to be identified with the category coding segment in the identity coding sequence. The reference biometric information is compared; according to the comparison result and the similarity calculation result, the target identity coding sequence is queried to determine the identity code of the animal to be identified.

Description

Feedstock identification method and device

The embodiments of this specification relate to the technical field of identification, and in particular, to a method for identification of livestock animals. One or more embodiments of the present specification simultaneously relate to a feedstock identification device, a computing device, and a computer-readable storage medium.

With the acceleration of the pace of social development, the work pressure and life pressure faced by everyone as a participant in social development are also increasing. More and more people like to keep pets. Keeping pets will enrich themselves and help their physical and mental health a lot, especially for some elderly people who live alone with their children who are struggling in other places, and have pets to accompany them. It will make the life of the elderly happier, and for some well-trained pets, when the elderly have some accidents, such as sickness, the pet will also play a role in the alarm. The pet is equivalent to a family member for the owner. A very important presence, so a lot of pet-oriented services have arisen.

In view of this, the embodiments of the present specification provide a kind of animal identification method. One or more embodiments of the present specification simultaneously relate to a feeder identification device, a computing device, and a computer-readable storage medium, so as to solve the technical defects existing in the prior art.

According to a first aspect of the embodiments of the present specification, there is provided a method for identifying livestock animals, comprising: acquiring a first biometric image and a second biometric image of the livestock to be identified, and calculating the first biometric characteristics The similarity between the image and the benchmark biometric images of multiple animals in the database; the benchmark biometric image that satisfies the preset conditions in the similarity calculation result is used as the target image, and the reading contains the same as the target image. The identity coding sequence of the corresponding biometric coding segment; extract the biometric information to be identified of the second biometric image, and map the biometric information to be identified with the reference biometric feature of the category coding segment in the identity coding sequence The information is compared; according to the number of identically aligned coding segments contained in the identity coding sequence and the similarity calculation result, the target identity coding sequence is queried, and the target identity coding sequence is used as the identity coding sequence of the animal to be identified. identification code.

Optionally, the biometric information to be identified includes category information to be identified, and the category encoding segment includes a category encoding segment; correspondingly, the biometric information to be identified and the category encoding segment in the identity encoding sequence The mapped reference biometric information is compared, including: reading in the database in advance according to each reference biometric information and class The mapping relationship table constructed by the mapping relationship of different coding segments; query the reference type information mapped by the type coding segment in the mapping relationship table; compare the type information to be identified with the reference type information mapped by the type coding segment Comparison.

Optionally, the biometric information to be identified further includes category information to be identified, and the category encoding segment further includes a category encoding section; correspondingly, the reference for mapping the category information to be identified and the category encoding section After the sub-step of aligning the category information is executed, it also includes: screening the identity coding sequences including the category coding segments that are consistent with the alignment as a first identity coding sequence set according to the comparison result; querying the first identity coding sequence in the mapping relationship table The reference variety information mapped by the variety coding segment in each identity coding sequence included in the identity coding sequence set; and the to-be-identified variety information is compared with the reference variety information mapped by the variety coding segment.

Optionally, the biometric information to be identified further includes gender information to be identified, and the category coding segment further includes a gender coding segment; correspondingly, the reference for mapping the species information to be identified and the species coding segment After the step of aligning the variety information is performed, it also includes: screening, according to the comparison result, the identity coding sequences comprising the aligned variety coding segments in the first set of identity coding sequences as the second set of identity coding sequences; Query the reference gender information of the gender coding segment mapped in each identity coding sequence included in the second identity coding sequence set in the mapping relationship table; map the gender information to be identified with the reference gender of the gender coding segment information to compare.

Optionally, the target identity coding sequence is queried according to the number of identically aligned coding segments contained in the identity coding sequence and the similarity calculation result, and the target identity coding sequence is used as the animal to be identified. The identification code includes: according to the number of aligned coding segments contained in each identity encoding sequence in the second identity encoding sequence set and the similarity calculation result, query the second identity encoding sequence set in the second identity encoding sequence set. The target identity coding sequence is used as the identity code of the animal to be identified.

Optionally, the querying the target identity coding sequence according to the number of aligned coding segments contained in the identity coding sequence and the similarity calculation result includes: mapping the category coding segment in the identity coding sequence with the The number of biometric information whose biometric information to be identified is aligned is determined as the number of aligned coding segments contained in the identity coding sequence; the identity coding sequence containing the most aligned coding segments is screened and based on The screening results construct a third identity coding sequence set; determine whether the number of identity coding sequences in the third identity coding sequence set is equal to 1; If so, determine the identity coding sequence in the third identity coding sequence set as the target identity coding sequence; The identity coding sequence of the biometric coding segment corresponding to the target image with the highest similarity is determined as the target identity coding sequence.

Optionally, obtaining the first biometric image and the second biometric image of the animal to be identified, and calculating the first biometric image and the benchmark biometric images of the multiple animals in the database. After the similarity step is performed, the reference biometric image that meets the preset conditions in the similarity calculation result is used as the target image, and the identity coding sequence step that contains the biometric coding segment corresponding to the target image is read. Before executing, it also includes: judging whether the similarity between the reference biometric image of any animal and the first biometric image meets the preset similarity threshold in the similarity calculation result; if not, executing The step of taking the reference biometric image that satisfies the preset condition in the similarity calculation result as the target image, and reading the identity code sequence including the biometric code segment corresponding to the target image.

Optionally, the identity coding sequence further includes an attribute coding segment; correspondingly, extracting the biometric information to be identified from the second biometric image, and combining the biometric information to be identified with the identity coding sequence. After the step of comparing the benchmark biometric information mapped by the middle class coding segment is performed, the target identity coding sequence is queried according to the number of aligned coding segments contained in the identity coding sequence and the similarity calculation result. before performing the step of using the target identity coding sequence as the identity code of the animal to be identified, further comprising: according to the comparison result, screening the identity coding sequence comprising the consistent category coding segment as the fourth identity code sequence set; extract the attribute information to be identified of the first biometric image or the second biometric image, and compare the attribute information to be identified with the information contained in each identity coding sequence in the fourth identity coding sequence set. The benchmark attribute information of the attribute code segment mapping is compared.

Optionally, the attribute information to be identified includes age information to be identified, and the attribute encoding segment includes an age encoding segment; correspondingly, the extraction of the first biometric image or the second biometric image is performed. To-be-identified attribute information, comparing the to-be-identified attribute information with the reference attribute information mapped by the attribute coding segments included in each identity coding sequence in the fourth set of identity coding sequences, including: comparing the first biometric map Input the pre-trained age recognition model as or the second biometric image, and obtain the output age recognition result as the age information to be recognized; determine whether the reference age information mapped by the age coding segment belongs to the age to be recognized The age range included in the information; if yes, it is determined that the comparison result of comparing the age information to be identified with the reference age information mapped by the age coding segment is consistent.

Optionally, the to-be-identified attribute information further includes to-be-identified location attribute information, and the attribute encoding segment includes a location encoding segment; correspondingly, the determining combines the to-be-identified age information with the age. The comparison result of the comparison of the reference age information of the coding segment mapping is that after the step of comparison is consistent, it also includes: screening the identity coding sequence including the age coding segment that is consistent with the comparison as the fifth identity coding sequence set according to the comparison result; Detect whether the first biometric image or the second biometric image carries location attribute information; if so, extract the location attribute information as the to-be-identified location attribute information, and use the to-be-identified location attribute information The information is compared with the reference position attribute information of the position coding segment mapping included in each identity coding sequence in the fifth set of identity coding sequences.

Optionally, the target identity coding sequence is queried according to the number of identically aligned coding segments contained in the identity coding sequence and the similarity calculation result, and the target identity coding sequence is used as the animal to be identified. The identification code, including: according to the number of identically aligned coding segments contained in each identity coding sequence in the fifth set of identity coding sequences and the similarity calculation result, query the target identity coding sequence, and encode the target identity The sequence is used as the identification code of the animal to be identified.

According to a second aspect of the embodiments of the present specification, there is provided an identification device for livestock animals, including: a similarity calculation module configured to acquire a first biometric image and a second biometric image of the livestock to be identified , and calculate the similarity between the first biometric image and the reference biometric images of a plurality of animal species in the database; The reading module is configured to use the reference biometric image that satisfies the preset condition in the similarity calculation result as the target image, and read the identity coding sequence including the biometric coding segment corresponding to the target image; compare The module is configured to extract the to-be-identified biometric information of the second biometric image, and compare the to-be-identified biometric information with the reference biometric information mapped by the category coding segment in the identity coding sequence Query module, configured to query the target identity coding sequence according to the number of identically aligned coding segments contained in the identity coding sequence and the similarity calculation result, and use the target identity coding sequence as the to-be-identified sequence The identification code of the animal.

Optionally, the query module includes: a determination sub-module, configured to determine the number of biometric information that is consistent with the biometric information to be identified that is mapped by the category coding segment in the identity coding sequence. is the number of identically aligned coding segments contained in the identity coding sequence; the screening submodule is configured to screen the identity coding sequence with the largest number of identically aligned coding segments as the third set of identity coding sequences; the judging submodule group, configured to judge whether the number of identity coding sequences in the third set of identity coding sequences is equal to 1; if the execution result of the judgment submodule is yes, then run the first target identity coding sequence to determine the submodule; The first target identity coding sequence determination sub-module is configured to determine the identity coding sequence in the third set of identity coding sequences as the target identity coding sequence; if the execution result of the judgment submodule is no, then run the second target identity coding sequence to determine the submodule; the second target identity coding sequence to determine the submodule is configured to The identity coding sequence that includes the biometric coding segment corresponding to the target image with the highest similarity to the first biometric image in the third identity coding sequence set is determined as the target identity coding sequence.

According to a third aspect of the embodiments of the present specification, a computing device is provided, including: a memory and a processor; the memory is used for storing computer-executable instructions, and the processor is used for executing the computer-executable instructions: Obtain the first biometric image and the second biometric image of the animal to be identified, and calculate the similarity between the first biometric image and the reference biometric images of the multiple animals in the database; The reference biometric image that satisfies the preset condition in the degree calculation result is taken as the target image, and the identity coding sequence including the biometric coding segment corresponding to the target image is read; Identifying biometric information, comparing the biometric information to be identified with the reference biometric information mapped by the category coding segment in the identity coding sequence; according to the number of aligned coding segments contained in the identity coding sequence and The similarity calculation result is used to query the target identity coding sequence, and the target identity coding sequence is used as the identity of the animal to be identified. code.

According to a fourth aspect of the embodiments of the present specification, a computer-readable storage medium is provided, which stores computer-executable instructions, and when the instructions are executed by a processor, implements any one of the steps of the method for identifying livestock.

In an embodiment of the present specification, the similarity between the first biometric image and the reference biometric images of multiple livestock in the database is calculated by acquiring the first biometric image and the second biometric image of the animal to be identified. Calculate, take the reference biometric image that meets the preset condition in the calculation result as the target image and read the identity coding sequence containing the biometric coding segment in the database, wherein the biometric coding segment corresponds to the target image ; Extract the biometric information to be identified of the second biometric image, and compare the biometric information to be identified with the reference biometric information mapped by the category coding segment in the identity coding sequence; according to the comparison result and the similarity The calculation results query the target identification code sequence to determine the identification code of the animal to be identified.

The embodiments of this specification implement the method of image recognition to collect the biometric information of the animal to be identified in the image, and combine the biometric information to be identified with the identity of the animal pre-stored in the database The benchmark biometric information mapped by different coding segments in the coding sequence is compared to identify the animals to be identified, which improves the feasibility of the method for identification of the animals, and the animals are identified by taking images and identifying them through image recognition. The identity of the animal not only ensures the safety of the animal, but also can identify the identity of the animal anytime and anywhere, which improves the convenience.

302: Similarity calculation module

304: read module

306: Comparison Module

308: Query Module

400: Computing Equipment

410: Memory

420: Processor

430: Busbar

440: Access device

450:Database

460: Internet

[Fig. 1] is a process flow chart of a method for identification of livestock animals provided by an embodiment of this specification; [Fig. 2] is a process flow diagram of a method for identification of livestock animals provided by an embodiment of this specification applied to insurance items; [ Fig. 3 ] is a schematic diagram of a feeder identification device provided by an embodiment of the present specification; [ Fig. 4 ] is a structural block diagram of a computing device provided by an embodiment of the present specification.

In the following description, numerous specific details are set forth in order to provide a thorough understanding of this specification. However, this specification can be implemented in many other ways different from those described herein, and those skilled in the art can make similar promotions without departing from the connotation of this specification. Therefore, this specification is not limited by the specific implementation disclosed below.

The terminology used in one or more embodiments of this specification is for the purpose of describing a particular embodiment only and is not intended to limit the one or more embodiments of this specification. As used in this specification and in one or more embodiments and the appended claims, the singular forms "a," "the," and "the" are intended to include the plural forms as well, unless the context clearly dictates otherwise. It will also be understood that the term "and/or" as used in one or more embodiments of this specification refers to and includes any or all of one or more of the associated listed items. Combinations are possible.

It will be understood that although the terms first, second, etc. may be used in this specification to describe various information in one or more embodiments of this specification, these information should not be limited by these terms. These terms are only used to distinguish information of the same type from one another. For example, a first could be termed a second, and similarly, a second could be termed a first, without departing from the scope of one or more embodiments of this specification. Depending on the context, the word "if" as used herein can be interpreted as "at the time of" or "at the time of" or "in response to determining."

In this specification, a method for identification of livestock animals is provided. This specification also relates to a device for identification of livestock animals, a computing device, and a computer-readable storage medium, which will be described in detail in the following embodiments.

FIG. 1 shows a flow chart of a method for identifying livestock animals according to an embodiment of the present specification, including steps 102 to 108 .

Step 102: Obtain a first biometric image and a second biometric image of the animal to be identified, and calculate the similarity between the first biometric image and the reference biometric images of multiple animals in the database .

In the embodiments of this specification, domestication refers to feeding, domestication or raising; domestication includes animals and plants fed or raised by users, including animal pets (pet dogs, pet cats, pet pigs, etc.), plant pets, tea pet pets (Golden toad, Pixiu, etc.) and alternative pets (marmots, rabbits, hamsters, hedgehogs, bats, etc.), etc., in addition to the above-mentioned domesticated pets, domesticated animals also include poultry, such as chickens, ducks, etc.

The embodiments of this specification take the animal to be identified as a domestic pet as an example To explain, as more and more users choose to keep pets, more and more animal husbandry service agencies emerge as the times require. During the process of applying for insurance or medical diagnosis, if the pre-established file information of the animal is stored in the database, the animal should be identified at the animal care service agency to check its historical insurance record or historical medical treatment. Diagnosis records and other related information, in order to recommend insurance types for users or conduct accurate diagnosis and treatment based on relevant information.

In the method for identification of livestock animals provided by the embodiments of this specification, in the case that the file information of the livestock to be identified is pre-stored in the database, the livestock service agency can obtain the biometric image of the livestock to be identified by obtaining the biometric image of the livestock to be identified. and identify the animal to be identified through the acquired biometric image.

The animal to be identified in the embodiments of this specification is the animal that needs to be identified, and the first biometric image and the second biometric image are the biometric images of the animal to be identified , collected when the animal identification request is received or uploaded by the user who sent the animal identification request; the reference biometric image is the biological feature image of the animal collected when the animal file is established.

Specifically, the biometric image includes a first biometric image and a second biometric image, wherein the first biometric image includes a nose print feature image of the animal to be identified. , the second biometric image includes a full-body image or a facial image of the animal to be identified; after receiving the identification request sent by the user for the animal to be identified and obtained After taking the first biometric image of the animal to be identified, the similarity calculation is performed between the first biometric image and the reference biometric images of a plurality of pre-stored animals in the database; , the animal to be identified in the embodiments of this specification refers to a domesticated pet, and the first biometric image includes a nose print, palm print image, etc. of the animal to be identified; obtain the first biometric map of the animal to be identified After the imaging, the similarity calculation is performed between the first biometric image and the reference biometric images of the multiple animals in the database, and the identification of the animals to be identified is performed according to the similarity calculation result.

In an embodiment provided in this specification, after obtaining the similarity calculation result between the first biometric image and the reference biometric images of the multiple animals in the database, it is also necessary to determine whether there is any arbitrary similarity in the similarity calculation result. The similarity between the reference biometric image of a domesticated animal and the first biometric image satisfies a preset similarity threshold; if not, the reference biometric image that satisfies the preset condition in the similarity calculation result is used as the target image, and read the identity coding sequence including the biometric coding segment corresponding to the target image.

Specifically, if it is determined according to the judgment result that the similarity between any reference biometric image and the first biometric image is greater than a preset similarity threshold, it can be determined that the animal to be identified and the reference biometric feature In the image, the animal to which the reference biometric feature belongs is consistent, then the identity code sequence of the animal corresponding to the reference biometric image can be used as the identification code of the animal to be identified; the first biometric map If the similarity of the images is less than or equal to the preset similarity threshold, it is necessary to select n reference biometric images with high similarity that meet the preset conditions as the target image according to the similarity calculation result, where n is a positive integer, and n Determined according to the actual situation, which is not limited here. After the target image is determined, the biometric information to be identified in the second biometric image is compared with the reference biometric information mapped by the identity feature coding segment corresponding to the target image. .

Take the scenario where the user insures the animal to be identified (pet pet1) as an example, the first biometric image in this scene is the nose print feature image, after the insurance service agency obtains the nose print feature image of the pet pet1, the nose print feature image is Calculate the similarity between the pattern feature image and the benchmark biometric images of multiple pets in the database. If the similarity between the benchmark biometric image of pet2 in the database and the nose print feature image of pet pet1 is 97% , greater than the preset similarity threshold of 95%, then the identity code sequence of pet pet2 is determined as the identity code of pet pet1; If the similarity is less than or equal to the preset similarity threshold of 95%, the similarity calculation results are sorted in descending order, and the top n benchmark biometric images that meet the preset conditions in the sorting result are selected as the target image. image, and use the target image to identify pet pet1.

By calculating the similarity between the first biometric image and the reference biometric images of a plurality of animal species in the database, and selecting some target images as the comparison objects for subsequent processing according to the calculation results, the comparison scope is narrowed. It is beneficial to improve the efficiency and accuracy of identification.

Step 104, the benchmark that satisfies the preset condition in the similarity calculation result The biometric image is used as the target image, and the identity coding sequence containing the biometric coding segment corresponding to the target image is read.

The identity code sequence described in the embodiments of this specification is the identity code of the animal to which the reference biometric information in the target image belongs, and the biometric code segment is the code segment mapped to the reference biometric information in the target image, and the identity code sequence In addition to the biometric coding sequence, the identity coding sequence may also contain other coding segments, such as category coding segments.

Specifically, on the basis of obtaining the similarity calculation result between the first biometric image and the reference biometric images of the multiple animals in the database, if any one of the reference biometric images is determined according to the similarity calculation result with all the reference biometric images If the similarities of the first biometric images are all less than or equal to the preset similarity threshold, the similarity calculation results are sorted in descending order, and the top n benchmark creatures that meet the preset conditions in the sorting results are selected. The feature image is used as the target image, and the identity coding sequence containing the biometric coding segment corresponding to the target image is read.

Still take the scene where the user insures the animal to be identified (pet pet1) as an example, the first biometric image in this scene is the nose print feature image, if the benchmark of any pet in the database is determined according to the similarity calculation result The similarity between the biometric image and the pet pet1's nose print feature image is less than or equal to the preset similarity threshold of 95%, then the similarity calculation results are sorted in descending order, and the sorting results meet the predetermined similarity. Set the top n reference biometric images of the condition as the target image, and read n identity coding sequences that respectively contain the biometric coding segments corresponding to the target image; After reading the n identity coding sequences, the identity of the animal to be identified is determined by comparing the biometric information to be identified in the second biometric image with the reference biometric information mapped to the category coding segment in the identity feature coding sequence Identifier.

Step 106 , extract the biometric information to be recognized of the second biometric image, and compare the biometric information to be recognized with the reference biometric information mapped by the category coding segment in the identity coding sequence.

The second biometric image described in the embodiments of the present specification includes a whole body image of the animal, and the biometric information to be identified is the information related to the category of the animal to be identified contained in the second biometric image of the animal to be identified. The biological feature information contained in the target image is the reference biological feature information, and the reference biological feature information has a mapping relationship with the category coding segment.

After determining the target image and reading the identity coding sequence including the biometric coding segment corresponding to the target image, by mapping the biometric information to be identified in the second biometric image with the category coding segment in the identity coding sequence The benchmark biometric information is compared to determine the identification code of the animal to be identified.

In an embodiment provided in this specification, the biometric information to be identified includes species information to be identified of the animal to be identified, and the category coding segment includes a species coding segment; the biometric information to be identified is combined with the The reference biometric information mapped to the category coding segment in the identity coding sequence is compared. After the comparison is completed, the identity code of the animal can be queried according to the comparison result and the similarity calculation result. The specific comparison process can be done as follows The method is realized by: reading the mapping relationship table constructed in advance according to the mapping relationship between each reference biometric information and the category coding segment in the database; querying the reference category information mapped by the category coding segment in the mapping relationship table; The category information to be identified is compared with the reference category information mapped by the category code segment.

Specifically, the mapping relationship table is pre-established by the server and stored in the database, and the mapping relationship table includes the mapping relationship between each reference biometric information and the category code segment, and the type code segment belongs to the animal. The coding segment of the species map is composed of two characters, and the characters include letters, numbers or special characters. In the embodiment of this description, the species to which the livestock belongs include reptiles, mammals, fish, amphibians, insects, etc. .

After extracting the species information to be identified of the animal to be identified in the second biometric image, query the reference species information mapped by species coding segments in the n identity coding sequences in the mapping relationship table, and compare the species information to be identified Whether it is consistent with the reference species information to determine the identification code of the animal to be identified.

Taking the pet dog pet D1 as the animal to be identified as an example, read 10 identity coding sequences A1, A2, . The species of pet dog pet D1 is mammals. After determining the species of pet dog pet D1, it is necessary to compare the species information to be identified (mammal species) of pet dog pet D1 with the reference species information mapped to the species coding segment in the 10 identity coding sequences , the examples of this specification are obtained by The identity coding sequence A1 in the 10 identity coding sequences is used as an example to illustrate. The species coding segment contained in the identity coding sequence A1 is O2. If the reference species information mapped by O2 is queried in the mapping relationship table as mammals, the species information to be identified will be The results obtained by comparing (mammals) with the reference species information (mammals) are consistent.

The specific implementation of the comparison between the category information to be identified and the reference category information mapped to the category encoding segment included in the identification encoding sequence other than the identification encoding sequence A1 is similar, and the specific implementation of the comparison with the benchmark category information in the identification encoding sequence A1 is Yes, it will not be repeated here.

Further, on the basis of comparing the species information of the livestock, the accuracy of the identification result of the to-be-identified livestock can also be improved by comparing the variety information.

Specifically, the biometric information to be identified further includes category information to be identified, and the category encoding segment further includes a category encoding section; the category information to be identified is compared with the reference category information mapped by the category encoding section, and After the comparison result is obtained, the identity coding sequence of the species coding segment mapped by the aligned reference species information is taken as a new set, and the species information to be identified is compared with each identity coding sequence in the new identity coding sequence set. Comparing the reference variety information contained in the variety coding segment mapping can be achieved through the following steps: screening the identity coding sequences including the aligned species coding segments according to the comparison result as the first identity coding sequence set; in the mapping In the relation table, query the reference varieties of the variety coding segment mapping in each identity coding sequence contained in the first set of identity coding sequences information; compare the to-be-identified variety information with the reference variety information mapped by the variety coding segment.

Specifically, after the category information to be identified is compared with the reference category information, the identity encoding sequence containing the category encoding segment mapped by the benchmark category information that is consistent with the comparison is selected as the first identity encoding sequence set according to the comparison result, and the The reference variety information mapped by the variety coding segment included in the identity coding sequence in the first identity coding sequence set is compared with the variety information to be identified to narrow the range of comparison, and the variety coding segment described in the embodiments of this specification is a breeding The coding segment of the type mapping to which the object belongs, which consists of three characters, and the characters include letters, numbers or special characters.

After extracting the to-be-identified species information of the animal to be identified in the second biometric image, query the mapping relationship table for the reference species information mapped by the species-encoding segments included in each identity-encoding sequence in the first identity-encoding sequence set, and compare Determine whether the information of the species to be identified is consistent with the information of the reference species to determine the identification code of the animal to be identified.

Following the above example, the animal to be identified is the pet dog pet D1, then the breed of the pet dog includes Akita Inu, Shiba Inu and Doberman Pinscher, etc., according to the comparison result, the identity coding sequence composition of the type coding segment that is consistent with the comparison is screened. The first identity code sequence set S1={A1, A2, A5, A8, A10}; extracting the type information to be identified of the second biometric image, it can be determined that the breed of the domesticated animal (pet dog pet D1) is Akita For dogs, after determining the breed of pet dog pet D1, it is necessary to compare the breed information (Akita Inu) of pet dog pet D1 to be identified with the breeds contained in the 5 identity coding sequences in the first set of identity coding sequences. The reference variety information mapped by the coding segment is compared, and the embodiment of this specification takes the identity coding sequence A2 in the first identity coding sequence set as an example to illustrate, the variety coding segment included in the identity coding sequence A2 is 334, which is shown in the mapping relationship table. If the benchmark breed information mapped by query 334 is Doberman, then the result obtained by comparing the to-be-identified breed information (Akita) with the benchmark breed information (Doberman) is that the comparison is inconsistent.

The specific implementation of the comparison between the biometric information to be identified and the reference variety information mapped by the variety coding segment included in the other identity coding sequences in the first identity coding sequence set is similar to the specific implementation of the comparison with the reference variety information in the identity coding sequence A2. That is, it will not be repeated here.

Furthermore, on the basis of comparing the species information and variety information of the animal to be identified, the accuracy of the identification result of the animal to be identified can also be improved by comparing the gender information.

Specifically, the biometric information to be identified includes gender information to be identified, and the category coding segment includes a gender coding segment; the species information to be identified is compared with the reference species information mapped by the species coding segment, and a comparison is obtained. After the results are compared, the identity coding sequence of the variety coding segment mapped with the consistent reference variety information is used as a new set, and the gender information to be identified and the identity coding sequence contained in each identity coding sequence in the new set of identity coding sequences are used as a new set. The comparison of the reference gender information of the gender coding segment mapping can be achieved through the following steps: according to the comparison result, the identity coding sequence containing the aligned variety coding segment is selected from the first identity coding sequence set as the second identity. A collection of coding sequences; Query the reference gender information of the gender coding segment mapped in each identity coding sequence included in the second identity coding sequence set in the mapping relationship table; map the gender information to be identified with the reference gender of the gender coding segment information to compare.

Specifically, after the to-be-identified variety information is compared with the reference variety information, the identity coding sequence containing the variety coding segment mapped by the benchmark variety information that is consistent in the comparison is screened according to the comparison result as the second identity coding sequence set, and the The reference gender information mapped by the gender coding segment included in the identity coding sequence in the second set of identity coding sequences is compared with the gender information to be identified to narrow the comparison range, and the gender coding segment described in the embodiments of this specification is a The coding segment of the gender mapping to which the animal belongs is composed of one character. The character X indicates that the gender of the animal is female, the character Y indicates that the gender of the animal is male, and the character Z indicates other.

After extracting the to-be-identified gender information of the animal to be identified in the second biometric image, query the mapping relationship table for the reference gender information mapped by the gender-encoding segments included in each identity-encoding sequence in the second identity-encoding sequence set, and compare Determine whether the gender information to be identified is consistent with the reference gender information to determine the identification code of the animal to be identified.

Following the above example, the animal to be identified is the pet dog pet D1, then its gender includes male and female, and the second identity coding sequence set S1 comprising the identity coding sequence of the consistent breed coding segment is screened according to the comparison result. ={A1, A5, A8, A10}; Extract the type information to be identified of the second biometric image, and determine the pet The gender of the pet dog pet D1 is male. After determining the breed of the pet dog pet D1, it is necessary to compare the gender information (male) of the pet dog pet D1 to be identified and the gender coding segments contained in the 4 identity coding sequences in the second identity coding sequence set The mapped reference gender information is compared, and the specific comparison process is similar to the above-mentioned comparison process between the category information to be identified and the reference category information, and will not be repeated here.

According to the range from large to small, the species information, variety information and gender information are compared with the reference information in turn, and after each comparison is completed, a new set of identity coding sequences is generated according to the comparison results, which is conducive to narrowing the next time. The comparison range further improves the identification efficiency of livestock.

In addition to comparing the category information (category information, breed information and gender information) of the animal to be identified, the accuracy of the identification result of the animal to be identified can also be improved by comparing the attribute information.

In an embodiment provided in this specification, the identity feature encoding sequence includes an attribute encoding segment in addition to a biometric encoding segment and a category encoding segment; a reference biometric feature that maps the biometric information to be identified and the category encoding segment After the information is compared and the comparison result is obtained, the identity coding sequence containing the category coding segment mapped by the consistent reference biometric information is used as a new set, and the attribute information to be identified is compared with the new identity coding sequence. Comparing the reference attribute information of the attribute coding segment mappings contained in each identity coding sequence in the set can be realized by the following steps: according to the comparison result, the identity coding sequence containing the class coding segment that is consistent in the alignment is selected as the fourth identity coding sequence collection; extracting the first biometric image or the second biometric image The to-be-identified attribute information is compared with the reference attribute information mapped to the attribute-encoding segment included in each identity-encoding sequence in the fourth identity-encoding sequence set.

Specifically, after extracting the to-be-identified attribute information of the animal to be identified in the first biometric image or the second biometric image, query the mapping relationship table for the attribute codes contained in each identity code sequence in the fourth set of identity code sequences The reference attribute information of the segment mapping is compared, and the identification code of the animal to be identified is determined by comparing whether the attribute information to be identified is consistent with the reference attribute information.

In an embodiment provided in this specification, the to-be-identified attribute information includes to-be-identified age information, and the attribute encoding segment includes an age encoding segment; the attribute information to be identified and the attribute encoding segment in the identity encoding sequence are mapped The benchmark attribute information is compared, which can be achieved through the following steps: inputting the first biometric image or the second biometric image into a pre-trained age recognition model, and obtaining the output age recognition result as the pending age recognition result. Identify age information; determine whether the reference age information mapped by the age coding segment belongs to the age range included in the age information to be identified; if so, determine the reference age for mapping the age information to be identified and the age coding segment The result of the comparison of the information is the same as the comparison.

Specifically, the first biometric image or the second biometric image is input into the pre-trained age recognition model, the age range of the animal to be recognized outputted by the model is obtained, and each individual in the fourth identity code sequence set is determined. Whether the reference age information mapped by the age-encoding segment included in the share-encoding sequence belongs to the age interval, and if so, the comparison is consistent; Birth year (first 4 digits) month (5th to 6th) day (7th to 8th).

Following the above example, the animal to be identified is pet dog pet D1, the age interval of pet dog pet D1 output by the age prediction model is 3 to 4 years old, and the age coding segment mapping included in the identity coding sequence A10 in the fourth identity coding sequence The reference age information is 4 years old, the result of comparing the age information to be identified with the reference age information mapped to the age coding segment included in the identity coding sequence A10 is consistent; the age information to be identified is consistent with the fourth set of identity coding sequences The specific implementation of the comparison between the reference age information mapped to the age coding segments included in the other identity coding sequences is similar, and the specific implementation of the comparison with the reference age information in the identity coding sequence A10 may be referred to, which will not be repeated here.

In addition to comparing the category information (type information, breed information and gender information) and age information of the animal to be identified, the location attribute information can also be compared to improve the accuracy of the identification result of the animal to be identified.

In an embodiment provided in this specification, the to-be-identified attribute information further includes to-be-identified location attribute information, and the attribute encoding segment includes a location encoding segment; a reference age that maps the to-be-identified age information to the age encoding segment After the information is compared and the comparison result is obtained, the identity coding sequence containing the age coding segment mapped with the consistent reference age information is taken as a new set, and the position information to be identified is compared with each of the new set. Comparing the reference location information of the location coding segment mappings included in the identity coding sequences can be achieved through the following steps: screening the identity coding sequences including the age coding segments that are consistent with the alignment as the fifth identity coding sequence set according to the comparison result; Detect whether the first biometric image or the second biometric image carries location attribute information; if so, extract the location attribute information as the to-be-identified location attribute information, and use the to-be-identified location attribute information The information is compared with the reference position attribute information of the position coding segment mapping included in each identity coding sequence in the fifth set of identity coding sequences.

Specifically, after the age information to be identified is compared with the reference age information, the identity coding sequence containing the age coding segment mapped by the reference age information that is consistent in the comparison is selected as the fifth identity coding sequence set according to the comparison result. The reference location information mapped by the location encoding segment included in the identity encoding sequence in the fifth identity encoding sequence set is compared with the location information to be identified, so as to narrow the comparison range. And, the position coding segment described in the embodiment of this specification is made up of 6 digits, and the first two are the province (city) where the animal is located, the middle two are the city (district) where the animal is located, and the last two are the county where the animal is located. .

In the embodiment of this specification, the location attribute information can be obtained by querying the image attribute information of the first biometric image or the second biometric image; and the location information carried in the image attribute information is mostly location coordinates. During the process, the location coordinates need to be converted into specific location information; after detecting the location information carried in the first biometric image or the second biometric image, Extract the position information, and compare the position information as the position information to be identified with the reference position information of the position coding segment mapping contained in each identity coding sequence in the fifth identity coding sequence set, and the specific comparison process is similar to the aforementioned comparison process, It is not repeated here.

Step 108, query the target identity coding sequence according to the number of identically aligned coding segments contained in the identity coding sequence and the similarity calculation result, and use the target identity coding sequence as the identification of the animal to be identified. code.

Specifically, the coding segments in the identity coding sequence are respectively mapped to one kind of reference biometric information, and after the comparison is completed, the identity coding sequence with the largest number of coding segments mapped by the aligned reference biometric information is determined as the target identity coding sequence, If there are two or more identically aligned reference biometric information mappings in two or more identity coding sequences with the largest and equal number of coding segments, the identity of the biometric coding segment corresponding to the target image with greater similarity will be included. The coding sequence is determined to be the target identity coding sequence.

In an embodiment provided in this specification, the querying of the target identity coding sequence according to the number of identically aligned coding segments contained in the identity coding sequence and the similarity calculation result can be specifically implemented by the following steps: The number of biometric information mapped by the category coding segment in the identity coding sequence that is consistent with the biometric information to be identified is determined as the number of aligned coding segments contained in the identity coding sequence; The identity coding sequence with the largest number of coding segments is constructed and a third identity coding sequence set is constructed based on the screening result; the number of identity coding sequences in the third identity coding sequence set is judged Whether it is equal to 1; if so, determine the identity coding sequence in the third identity coding sequence set as the target identity coding sequence; The identity coding sequence of the biometric coding segment corresponding to the target image with the highest similarity of the biometric image is determined as the target identity coding sequence.

Specifically, on the basis of comparing the gender information to be identified with the reference gender information mapped by the gender coding segment and obtaining the comparison result, the number of coding segments that are consistent according to the alignment contained in the identity coding sequence and The similarity calculation result queries the target identity coding sequence, and uses the target identity coding sequence as the identity code of the animal to be identified, that is, according to the ratio of each identity coding sequence contained in the second set of identity coding sequences. To the consistent number of coding segments and the similarity calculation result, query the target identity coding sequence in the second set of identity coding sequences, and use the target identity coding sequence as the identification of the animal to be identified code; after comparing the position attribute information to be identified with the reference position attribute information of the position coding segment mapping contained in each identity coding sequence in the fifth set of identity coding sequences and obtaining a comparison result, the said identity coding sequence The number of consistent coding segments included in the comparison and the similarity calculation result query the target identity coding sequence, and the target identity coding sequence is used as the identity code of the animal to be identified, that is, according to the fifth identity The number of aligned coding segments contained in each identity coding sequence in the set of coding sequences and the result of the similarity calculation query the target identity coding sequence, and The target identity coding sequence is used as the identification code of the animal to be identified; the final identified identification code of the animal to be identified is 340827, 20141019, 02, 334, X, 100008 as an example, wherein, The first 6 digits are the first 6 digits of the owner's ID card, which can be used to indicate the location of the animal; the 7th to 14th digits are the date of birth of the animal, which can be used to indicate the age of the animal; and the 1st to 14th digits are attribute information codes. The 15th to 16th digits indicate the species of the livestock; the 17th to 19th digits represent the species of the livestock; the 20th position represents the sex of the livestock; Biometric coding segment of the animal; after determining the identity code of the animal to be identified, the insurance service agency can use the identification code to query the historical insurance record and historical medical diagnosis information of the animal to be identified to recommend the appropriate animal to be identified. 's risk.

The embodiment of this specification adopts the method of image recognition to identify the biometric information of the animal in the image, and compares the biometric information with the fields of the electronic ID card of the animal pre-stored in the database to obtain The identification of the animals to be identified improves the feasibility of the method of identification of the animals, and the identification of the animals by means of photography not only ensures the safety of the animals, but also improves the convenience of identifying the animals' identities. sex.

The following is a further description of the method for the identification of domesticated animals by taking the application of the method for identification of domesticated animals provided in this specification in an insurance project as an example with reference to FIG. 2 . Wherein, FIG. 2 shows a processing flow of applying a method for identification of livestock animals provided by an embodiment of the present specification to an insurance item The specific steps include step 202 to step 218.

Step 202: Obtain the first biometric image and the second biometric image of the pet dog pet D1, and calculate the similarity between the first biometric image and the reference biometric images of multiple pet dogs in the database.

Specifically, when receiving the insurance request sent by the user for the pet dog pet D1, it is necessary to obtain the biometric image of the pet dog pet D1 to identify it, so as to determine the historical insurance application record and the historical medical treatment of the pet dog pet D1. According to the historical insurance record and historical medical treatment information, the appropriate insurance type is determined for the pet dog pet D1.

Step 204: Determine whether the similarity calculation result satisfies a preset similarity threshold.

Specifically, it is determined whether the similarity between the reference biometric image of the pet dog and the first biometric image meets the preset similarity threshold in the similarity calculation result; if so, go to step 206; if not, go to step 206 Then step 208 is executed.

Step 206, taking the identity coding sequence including the biometric coding segment corresponding to the reference biometric image as the target identity coding sequence.

Step 208 , taking the reference biometric image that satisfies the preset condition in the similarity calculation result as the target image, and read the identity coding sequence including the biometric coding segment corresponding to the target image.

Specifically, if the similarity between any reference biometric image and the first biometric image is less than or equal to a preset similarity threshold, it is necessary to select n similarity comparisons that satisfy the preset conditions according to the similarity calculation result. A large baseline biometric image is used as the target image, where n is a positive integer.

Step 210 , extract the biometric information to be identified of the second biometric image, and compare the biometric information to be identified with the reference biometric information mapped by the category coding segment in the identity coding sequence.

Specifically, the biometric information to be identified includes category information to be identified, type information to be identified, and gender information to be identified; the category encoding segment includes a category encoding segment, a variety encoding segment, and a gender encoding segment.

Step 212 , according to the comparison result, screen the identity coding sequences including the identically aligned category coding segments as the identity coding sequence set.

Step 214: Extract the attribute information to be identified of the first biometric image or the second biometric image, and encode the attribute information to be identified with the attributes contained in each identity coding sequence in the set of identity coding sequences. The benchmark attribute information of the segment map is compared.

Specifically, the to-be-identified attribute information includes to-be-identified age information and to-be-identified location information; the attribute encoding segment includes an age encoding segment and a location encoding segment.

Step 216, query the target identity coding sequence according to the number of identically aligned coding segments contained in the identity coding sequence and the similarity calculation result, and use the target identity coding sequence as the identification of the pet dog pet D1 code.

Specifically, the coding segments in the identity coding sequence are respectively mapped to one kind of reference biometric information, and after the comparison is completed, the identity coding sequence with the largest number of coding segments mapped by the aligned reference biometric information is determined as the target identity coding sequence, If there are two or more identity-coding sequences that contain consistent alignments, the number of coding segments mapped to the reference biometric information is the largest and If they are equal, the identity coding sequence including the biometric coding segment corresponding to the target image with a greater degree of similarity is determined as the target identity coding sequence, and the target identity coding sequence is used as the dog's ID code.

Step 218 , query the historical insurance application records and historical medical treatment records of the pet dog pet D1 by using the identification code, and determine the insurance type for the pet dog pet D1 according to the historical insurance application records and historical medical treatment records.

The embodiment of this specification adopts the method of image recognition to identify the biometric information of the pet dog in the image, and maps the biometric information to different coding segments of the electronic ID card of the pet dog pre-stored in the database. The benchmark biometric information is compared to identify the pet dog, which improves the feasibility of the pet dog identification method, and the identification of the pet dog through the method of photography not only ensures the safety of the domestic animals, but also improves the pet dog's identity. The ease of identification of the dog's identity.

Corresponding to the above method embodiments, the present specification also provides an embodiment of a feeder identification device, and FIG. 3 shows a schematic structural diagram of a feeder identification device provided by an embodiment of the present specification. As shown in FIG. 3 , the device includes: a similarity calculation module 302, configured to acquire a first biometric image and a second biometric image of the animal to be identified, and calculate the first biometric image Similarity with the reference biometric images of multiple animals in the database; the reading module 304 is configured to use the reference biometric image that meets the preset condition in the similarity calculation result as the target image, and read an identity coding sequence comprising the biometric coding segment corresponding to the target image; The comparison module 306 is configured to extract the to-be-recognized biometric information of the second biometric image, and compare the to-be-identified biometric information with the reference biometric information mapped to the category coding segment in the identity coding sequence. Alignment; the query module 308 is configured to query the target identity coding sequence according to the number of identically aligned coding segments contained in the identity coding sequence and the similarity calculation result, and use the target identity coding sequence as the target identity coding sequence. Describe the identification code of the animal to be identified.

Optionally, the biometric information to be identified includes category information to be identified, and the category code segment includes a category code segment; correspondingly, the comparison module 306 includes: a mapping relationship table reading sub-module, which is It is configured to read the mapping relationship table constructed in advance according to the mapping relationship between each reference biometric information and the category coding segment in the database; the reference category information query sub-module is configured to query the category in the mapping relationship table The reference category information of the coding segment mapping; the category information comparison sub-module is configured to compare the to-be-identified category information with the reference category information of the category coding segment mapping.

Optionally, the biometric information to be identified further includes species information to be identified, and the category encoding segment further includes a species encoding segment; correspondingly, the comparison module 306 further includes: a first identity encoding sequence screener a module, configured to screen the identity coding sequences comprising the identically aligned species coding segments as the first set of identity coding sequences according to the comparison result; A reference variety information query sub-module, configured to query the mapping relationship table for the reference variety information mapped by the variety coding segment in each identity coding sequence included in the first identity coding sequence set; the variety information comparison submodule The group is configured to compare the to-be-identified variety information with the reference variety information mapped by the variety code segment.

Optionally, the biometric information to be identified further includes gender information to be identified, and the category encoding segment further includes a gender encoding segment; correspondingly, the comparison module 306 further includes: a second identity encoding sequence screener The module is configured to screen, according to the comparison result, the identity coding sequences containing the identically aligned variety coding segments in the first set of identity coding sequences as the second set of identity coding sequences; the reference gender information query sub-module is selected by is configured to query the mapping relationship table for the reference gender information mapped by the gender coding segments in each identity coding sequence included in the second set of identity coding sequences; the gender information comparison sub-module is configured to The identification gender information is compared with the reference gender information mapped by the gender coding segment.

Optionally, the query module 308 includes: a first query sub-module, which is configured according to the number of aligned coding segments contained in each identity coding sequence in the second set of identity coding sequences and the similarity According to the result of the degree calculation, the target identity coding sequence is queried in the second set of identity coding sequences, and the target identity coding sequence is used as the identity code of the animal to be identified.

Optionally, the query module 308 includes: a determination sub-module configured to encode the category in the identity coding sequence The number of the biometric information mapped by the code segment that is consistent with the biometric information to be identified is determined as the number of aligned coding segments contained in the identity coding sequence; The identity coding sequence with the largest number of consistent coding segments is used as the third identity coding sequence set; the judgment submodule is configured to judge whether the number of identity coding sequences in the third identity coding sequence set is equal to 1; if the judgment is The execution result of the submodule is yes, then run the first target identity coding sequence to determine the submodule; the first target identity coding sequence to determine the submodule is configured to The coding sequence is determined to be the target identity coding sequence; if the execution result of the judgment submodule is no, then run the second target identity coding sequence to determine the submodule; the second target identity coding sequence determines the submodule, It is configured to determine an identity coding sequence that includes a biometric coding segment corresponding to a target image with the highest similarity to the first biometric image in the third set of identity coding sequences as the target identity coding sequence.

Optionally, the animal identification device further includes: a judging module configured to determine whether there is any reference biometric image of the animal and the first biometric map in the similarity calculation result. The similarity of the images satisfies the preset similarity threshold; if the execution result of the judgment module is no, the read module is run.

Optionally, the identity coding sequence further includes an attribute coding segment; correspondingly, the animal identification device further includes: an identity coding sequence screening module, configured to screen the categories that contain the same alignment according to the comparison result The identity coding sequence of the coding segment is used as the fourth identity coding sequence set; the attribute information comparison module is configured to extract the to-be-identified attribute information of the first biometric image or the second biometric image, The to-be-identified attribute information is compared with the reference attribute information mapped to the attribute code segment included in each identity code sequence in the fourth set of identity code sequences.

Optionally, the attribute information to be identified includes age information to be identified, and the attribute coding section includes an age coding section; correspondingly, the attribute information comparison module includes: a submodule for determining age information to be identified, which is It is configured to input the first biometric image or the second biometric image into a pre-trained age recognition model, and obtain the output age recognition result as the age information to be recognized; the age information judgment submodule is It is configured to judge whether the reference age information mapped by the age coding segment belongs to the age range included in the age information to be identified; if the execution result of the age information judgment submodule is yes, run the comparison result to determine the submodule group; the comparison result determination submodule is configured to determine to compare the age information to be identified with the reference age information mapped to the age coding segment The alignment results of the alignment are consistent.

Optionally, the to-be-identified attribute information further includes to-be-identified location attribute information, and the attribute encoding segment includes a location encoding segment; correspondingly, the attribute information comparison module further includes: an identity encoding sequence set determination submodule The group is configured to screen the identity coding sequences containing the age coding segments that are consistent in the comparison as the fifth identity coding sequence set according to the comparison result; the location attribute information detection sub-module is configured to detect the first biometric image. Or whether the position attribute information is carried in the second biometric image; if the execution result of the position attribute information detection sub-module is yes, then run the position attribute information comparison sub-module; the position attribute information comparison A sub-module, configured to extract the location attribute information as the to-be-identified location attribute information, and map the to-be-identified location attribute information to the location code segments included in each identity code sequence in the fifth set of identity code sequences The reference location attribute information of .

Optionally, the query module 308 includes: a second target identity coding sequence query sub-module, configured to be based on the number of aligned coding segments contained in each identity coding sequence in the fifth set of identity coding sequences And the similarity calculation result is used to query the target identity coding sequence, and the target identity coding sequence is used as the identity code of the animal to be identified.

The above is a schematic diagram of a feeder identification device of the present embodiment Program. It should be noted that the technical solution of the livestock identification device and the technical solution of the above-mentioned livestock identification method belong to the same concept, and the details that are not described in detail in the technical solution of the livestock identification device can be referred to above. A description of the technical solution of the identification method.

FIG. 4 shows a structural block diagram of a computing device 400 provided according to an embodiment of the present specification. Components of the computing device 400 include, but are not limited to, the memory 410 and the processor 420 . The processor 420 and the memory 410 are connected through the bus bar 430, and the database 450 is used for storing data.

Computing device 400 also includes access device 440 that enables computing device 400 to communicate via one or more networks 460 . Examples of such networks include a public switched telephone network (PSTN), a local area network (LAN), a wide area network (WAN), a personal area network (PAN), or a combination of communication networks such as the Internet. Access device 440 may include one or more of any type of network interface (eg, network interface card (NIC)), wired or wireless, such as IEEE 802.11 Wireless Local Area Network (WLAN) wireless interface, World Interoperability for Microwave Access (Wi-MAX) interface, Ethernet interface, Universal Serial Bus (USB) interface, cellular network interface, Bluetooth interface, Near Field Communication (NFC) interface, etc.

In one embodiment of the present specification, the above-described components of computing device 400 and other components not shown in FIG. 4 may also be connected to each other, such as through bus bars. It should be understood that the structural block diagram of the computing device shown in FIG. 4 is only for the purpose of example, rather than limiting the scope of this specification. Those skilled in the art can add or replace other components as required.

Computing device 400 may be any type of stationary or mobile computing device. devices, including mobile computers or mobile computing devices (eg, tablet computers, personal digital assistants, laptops, notebook computers, small notebooks, etc.), mobile phones (eg, smartphones), wearable computing devices (eg, , smart watches, smart glasses, etc.) or other types of mobile devices, or stationary computing devices such as desktop computers or PCs. Computing device 400 may also be a mobile or stationary server.

The memory 410 is used for storing computer-executable instructions, and the processor 420 is used for executing the following computer-executable instructions: acquiring the first biometric image and the second biometric image of the animal to be identified, and calculating all the Describe the similarity between the first biometric image and the benchmark biometric images of multiple animals in the database; take the benchmark biometric image that satisfies the preset conditions in the similarity calculation result as the target image, and read the image containing the The identity coding sequence of the biometric coding segment corresponding to the target image; extracting the biometric information to be identified of the second biometric image, and combining the biometric information to be identified with the category coding segment in the identity coding sequence The mapped reference biometric information is compared; according to the number of identically aligned coding segments contained in the identity coding sequence and the similarity calculation result, the target identity coding sequence is queried, and the target identity coding sequence is used as the The identification code of the animal to be identified.

The above is a schematic solution of a computing device according to this embodiment. It should be noted that the technical solution of the computing device and the technical solution of the above-mentioned method for identification of livestock animals belong to the same concept, and the technical solution of the computing device is not identical. For the details of the detailed description, reference can be made to the description of the technical solution of the above-mentioned method for identification of livestock.

An embodiment of the present specification further provides a computer-readable storage medium, which stores computer instructions, and when the instructions are executed by a processor, is used to implement the steps of the method for identifying livestock.

The above is a schematic solution of a computer-readable storage medium of this embodiment. It should be noted that the technical scheme of the storage medium and the technical scheme of the above-mentioned method for identification of livestock animals belong to the same concept, and the details that are not described in detail in the technical scheme of the storage medium can be referred to the technical scheme of the above-mentioned method for identification of pigeons. description of.

The foregoing describes specific embodiments of the present specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in a different order than in the embodiments and still achieve desirable results. Additionally, the processes depicted in the figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

The computer instructions include computer program code, which may be in source code form, object code form, an executable file, some intermediate form, or the like. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, pen drive, portable hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory) Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software body distribution media, etc. It should be noted that the content contained in the computer-readable media may be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction, for example, in some jurisdictions, according to legislation and patent practice, the computer-readable media Electric carrier signals and telecommunication signals are not included.

It should be noted that, for the convenience of description, the foregoing method embodiments are all expressed as a series of action combinations, but those skilled in the art should know that the embodiments of this specification are not limited by the described action sequences. Limitation, because certain steps may be performed in other orders or simultaneously according to embodiments of the present specification. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily all necessary for the embodiments of the specification.

In the above-mentioned embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference may be made to the relevant descriptions of other embodiments.

The preferred embodiments of the present specification disclosed above are provided only to aid in the elaboration of the present specification. Alternative embodiments are not intended to exhaust all details, nor do they limit the invention to only the described embodiments. Obviously, many modifications and changes can be made in accordance with the contents of the embodiments of the present specification. These embodiments are selected and described in this specification to better explain the principles and practical applications of the embodiments of this specification, so that those skilled in the art can well understand and utilize this specification. This specification is limited only by the scope of the patent application and its full scope and equivalents.

Claims (15)

A method for identifying livestock animals, the method is implemented through a computing device, the computing device includes a memory and a processor, the memory is used for storing computer-executable instructions, when the processor executes the computer-executable instructions, the following steps are performed , comprising: acquiring the first biometric image and the second biometric image of the animal to be identified, and calculating the similarity between the first biometric image and the benchmark biometric images of multiple animals in the database; Taking the reference biometric image that satisfies the preset condition in the similarity calculation result as the target image, read the identity coding sequence including the biometric coding segment corresponding to the target image; extract the pending biometric image of the second biometric image. Identify the biometric information, and compare the biometric information to be identified with the reference biometric information mapped to the category coding segment in the identity coding sequence; according to the number of aligned coding segments contained in the identity coding sequence and the similarity The calculation results query the target identity code sequence, and use the target identity code sequence as the identity code of the animal to be identified. As claimed in claim 1, the biometric information to be identified includes category information to be identified, and the category code segment includes a category code segment; correspondingly, the biometric information to be identified and the identity code are combined. Align the reference biometric information mapped to the category coding segment in the sequence, including: reading in the database in advance according to each reference biometric information and the category The mapping relationship table constructed by the mapping relationship of different coding segments; query the reference type information mapped by this type of coding segment in the mapping relationship table; compare the type information to be identified with the reference type information mapped by this type of coding segment, wherein , the species information to be identified includes information on the following species: reptiles, mammals, fish, amphibians and insects. According to the animal species identification method according to claim 2, the biometric information to be identified further includes species information to be identified, and the category encoding segment further includes a species encoding segment; correspondingly, the category information to be identified and the species After performing the sub-step of aligning the reference category information of the coding segment mapping, it also includes: screening, according to the comparison result, the identity coding sequence containing the type coding segment that is consistent with the alignment as the first identity coding sequence set; querying the mapping relationship table The reference variety information mapped by the variety coding segment in each identity coding sequence included in the first identity coding sequence set; and the to-be-identified variety information is compared with the reference variety information mapped by the variety coding segment. As claimed in claim 3, the biometric information to be identified further includes gender information to be identified, and the category code segment further includes a gender code segment; correspondingly, the to-be-identified species information is associated with the species After the step of comparing the benchmark variety information mapped by the coding segment is performed, it also includes: According to the comparison result, the first identity coding sequence set contains the identity coding sequence that contains the consistent variety coding segment as the second identity coding sequence set; the second identity coding sequence set is queried in the mapping relationship table which contains The reference gender information mapped by the gender coding segment in each identity coding sequence of ; compare the gender information to be identified with the reference gender information mapped by the gender coding segment. As claimed in claim 4, the method for identifying animal species, wherein the target identity coding sequence is queried according to the number of aligned coding segments contained in the identity coding sequence and the similarity calculation result, and the target identity coding sequence is used as The identity code of the animal to be identified includes: according to the number of aligned coding segments contained in each identity coding sequence in the second identity coding sequence set and the similarity calculation result, in the second identity coding sequence set The target identity coding sequence is queried, and the target identity coding sequence is used as the identity code of the animal to be identified. As claimed in claim 1, the method for identifying the animal species, wherein querying the target identity coding sequence according to the number of identically aligned coding segments contained in the identity coding sequence and the similarity calculation result, includes: The number of biometric information mapped by the category coding segment that is consistent with the biometric information to be identified is determined as the number of aligned coding segments contained in the identity coding sequence; the identity with the largest number of aligned coding segments is screened. coding sequences and constructing a third set of identity coding sequences based on the screening results; Determine whether the number of identity coding sequences in the third identity coding sequence set is equal to 1; if so, determine the identity coding sequence in the third identity coding sequence set as the target identity coding sequence; if not, then determine the third identity coding sequence as the target identity coding sequence The identity coding sequence including the biometric coding segment corresponding to the target image with the highest similarity of the first biometric image in the identity coding sequence set is determined as the target identity coding sequence. The animal identification method according to claim 1, wherein the first biometric image and the second biometric image of the animal to be identified are obtained, and the first biometric image and a plurality of animals in the database are calculated. After the similarity step of the reference biometric image of the object is performed, the reference biometric image that satisfies the preset condition in the similarity calculation result is used as the target image, and the biometric coding segment containing the corresponding target image is read. Before the identity coding sequence step is performed, it also includes: judging whether the similarity between the reference biometric image of any animal and the first biometric image meets the preset similarity threshold in the similarity calculation result; if not , the step of taking the reference biometric image that satisfies the preset condition in the similarity calculation result as the target image, and reading the identity code sequence including the biometric code segment corresponding to the target image is performed. As claimed in claim 1, the identity code sequence further includes an attribute code segment; correspondingly, the biometric information to be recognized of the second biometric image is extracted, and the biometric information to be recognized is extracted. with the identity code sequence After the step of aligning the reference biometric information of the category coding segment mapping is performed, the target identity coding sequence is queried according to the number of aligned coding segments contained in the identity coding sequence and the similarity calculation result, and the target identity coding sequence is queried. Before performing the step of using the coding sequence as the identification code of the animal to be identified, it also includes: screening, according to the comparison result, the identity coding sequence comprising the category coding segment that is consistent with the alignment as the fourth set of identity coding sequences; extracting the first biological feature The to-be-identified attribute information of the image or the second biometric image, and the to-be-identified attribute information is compared with the reference attribute information mapped to the attribute code segment included in each identity code sequence in the fourth set of identity code sequences. The animal identification method according to claim 8, wherein the attribute information to be identified includes age information to be identified, and the attribute encoding segment includes an age encoding segment; correspondingly, the first biometric image or the second biometric image is extracted. The to-be-identified attribute information of the biometric image, comparing the to-be-identified attribute information with the reference attribute information mapped to the attribute coding segments included in each identity coding sequence in the fourth set of identity coding sequences, including: the first biological The feature image or the second biometric image is input into the pre-trained age recognition model, and the output age recognition result is obtained as the age information to be recognized; it is judged whether the reference age information mapped by the age code segment belongs to the age information to be recognized. Included age range; if so, determine to map the age information to be identified with the age code segment The comparison results of the comparison of the baseline age information of the radiation are consistent. According to the animal species identification method according to claim 9, the attribute information to be identified further includes the attribute information of the position to be identified, and the attribute coding segment includes the position coding segment; The comparison result of the comparison of the reference age information of the coding segment mapping is that after the step of comparison is consistent, it also includes: screening the identity coding sequence including the age coding segment that is consistent with the comparison as the fifth identity coding sequence set according to the comparison result; Detect whether the first biometric image or the second biometric image carries location attribute information; if so, extract the location attribute information as the to-be-identified location attribute information, and combine the to-be-identified location attribute information with the fifth The reference position attribute information of the position coding segment mapping included in each identity coding sequence in the identity coding sequence set is compared. The method for identifying animal species according to claim 10, wherein the target identity coding sequence is queried according to the number of aligned coding segments contained in the identity coding sequence and the similarity calculation result, and the target identity coding sequence is used as The identity code of the animal to be identified includes: querying the target identity code sequence according to the number of aligned coding segments contained in each identity code sequence in the fifth identity code sequence set and the similarity calculation result, and assigning the target identity code sequence to the target identity code sequence. The identity coding sequence is used as the identity code of the animal to be identified. An identification device for feeding animals, comprising: a similarity calculation module, configured to obtain a first biometric image and a second biometric image of the livestock to be identified, and calculate the first biometric image and database The similarity of the reference biometric images of the plurality of animals in The identity coding sequence of the biometric coding segment corresponding to the image; the comparison module is configured to extract the biometric information to be identified of the second biometric image, and the to-be-identified biometric information and the category in the identity coding sequence The reference biometric information of the coding segment mapping is compared; the query module is configured to query the target identity coding sequence according to the number of aligned coding segments contained in the identity coding sequence and the similarity calculation result, and assign the target identity coding sequence to the target identity coding sequence. The identity coding sequence is used as the identity code of the animal to be identified. The animal identification device according to claim 12, wherein the query module includes: a determination sub-module, configured to map the category coding segment in the identity coding sequence and compare it with the biometric information to be identified. The number of biometric information is determined as the number of identically aligned coding segments contained in the identity coding sequence; the screening sub-module is configured to screen the identity coding sequence that contains the largest number of identically aligned coding segments as the third identity coding sequence Set; judgment sub-module, configured to judge the third identity code sequence set Whether the number of identity coding sequences in the middle is equal to 1; if the execution result of the judgment submodule is yes, then run the first target identity coding sequence to determine the submodule; the first target identity coding sequence to determine the submodule is configured as Determine the identity coding sequence in the third identity coding sequence set as the target identity coding sequence; if the execution result of the judgment submodule is no, then run the second target identity coding sequence to determine the submodule; the second target The identity coding sequence determination sub-module is configured to determine the identity coding sequence of the biometric coding segment corresponding to the target image with the highest similarity to the first biometric image in the third identity coding sequence set as the identity coding sequence. Target identity encoding sequence. A computing device, comprising: a memory and a processor; the memory is used for storing computer-executable instructions, and the processor is used for executing the computer-executable instructions: acquiring a first biometric image of a feed to be identified and a second biometric image feature image, and calculate the similarity between the first biometric image and the benchmark biometric images of multiple livestock in the database; take the benchmark biometric image that satisfies the preset conditions in the similarity calculation result as the target image image, read the identity coding sequence containing the biometric coding segment corresponding to the target image; extract the biometric information to be identified of the second biometric image, and combine the biometric information to be identified with the category in the identity coding sequence Code segment map The target identity coding sequence is queried according to the number of identically aligned coding segments contained in the identity coding sequence and the similarity calculation result, and the target identity coding sequence is used as the animal to be identified. 's identification code. A computer-readable storage medium storing computer-executable instructions, when the instructions are executed by a processor, implements the steps of the method for identifying animal species in any one of claim 1 to 11.

Images