WO2022052189A1

WO2022052189A1 - Method and device for acquiring external features of animal, and computer device

Info

Publication number: WO2022052189A1
Application number: PCT/CN2020/118607
Authority: WO
Inventors: 陶品; 颜嘉辰; 史元春; 韩亮; 刘志强
Original assignee: 清华大学; 青海大学
Priority date: 2020-09-09
Filing date: 2020-09-29
Publication date: 2022-03-17
Also published as: CN112040132A

Abstract

Provided in the present application are a method and device for acquiring the external features of an animal, and a computer device. A common image capturing device is utilized to acquire a target image of an animal to be measured and photographing posture information of the image capturing device when photographing the target image, feature extraction is performed with respect to the target image to produce information of multiple key feature points of said animal, and the horizontal photographing distance between the center point of the image capturing device and said animal is determined on the basis of location information of a first key feature point and of photographing posture information, thus acquiring, on the basis of the horizontal photographing distance and location information of a second key feature point corresponding to body measurements to be measured, corresponding external feature values to be measured, preventing possible injuries that may be inflicted on said animal or on a measuring user during manual measurement, reducing equipment costs and professional requirements on the photographing user and a photographing venue, and increasing the universality of the method set forth in the present application for acquiring the external features of an animal.

Description

Method, Apparatus and Computer Equipment for Obtaining External Characteristics of Animals

This application claims the priority of the Chinese patent application with the application number 202010940011.3 and the invention titled "Method, Apparatus and Computer Equipment for Obtaining External Characteristics of Animals", which was filed with the China Patent Office on September 9, 2020, the entire contents of which are incorporated by reference in in this application.

technical field

The present application relates to the technical field of image processing, and in particular, to a method, device and computer equipment for acquiring external features of animals.

Background technique

In the field of animal husbandry, in order to realize the scientific and effective feeding and management of animals, it is usually necessary to measure the external characteristics of animals such as body weight and body size (such as height, length, chest circumference, body oblique length, etc.) to determine the growth and nutritional status of the corresponding animals. . However, because animals often do not obey human commands quietly, this brings great difficulties to the measurement of the external characteristics of animals, especially for larger animals. or cause injury to the measuring personnel.

In order to avoid the above injuries, it is currently proposed to scan the animal by means of a laser detector, and obtain the external characteristic values that the animal needs to measure according to the 3D reconstruction results of the scan. However, the measurement can be completed only after the measurement is completed, and the required measurement equipment is expensive and has high requirements on the measurement environment, which limits the application and popularization of this measurement method.

SUMMARY OF THE INVENTION

In view of this, in order to avoid the threat of injury that may be caused by manual measurement of the animal to be tested or the measurement user, an embodiment of the present application provides a method for acquiring external characteristics of an animal, the method comprising:

Obtain the target image of the animal to be tested, and the shooting attitude information when the image collector shoots the target image;

Perform feature extraction on the target image to obtain a plurality of key feature point information of the animal to be tested contained in the target image, where the key feature point information includes position information of the corresponding key feature point in the target image;

According to the shooting attitude information and the position information of the first key feature point, the horizontal shooting distance between the center point of the image collector and the animal to be measured is obtained, and the first key feature point refers to the to-be-measured animal. Among the multiple key feature points of the animal, the key feature points whose distance from the ground is less than the first threshold;

According to the horizontal shooting distance and the position information of the second key feature point corresponding to the external feature to be tested, the external feature value to be tested corresponding to the animal to be tested is obtained.

Optionally, the external features to be measured include at least one body size to be measured and/or at least one attribute to be measured of the animal to be measured, the horizontal shooting distance and the corresponding external features to be measured. The position information of the second key feature point to obtain the corresponding external feature value to be tested of the animal to be tested, including

According to the horizontal shooting distance and the position information of the second key feature point corresponding to the body measure to be measured, obtain the body measure parameter to be measured corresponding to the animal to be measured;

According to the obtained at least one body size parameter to be measured, the corresponding attribute value to be measured of the animal to be measured is obtained.

Optionally, the shooting attitude information includes a shooting pitch angle and a shooting roll angle, and the center point of the image collector and the animal to be measured are obtained according to the shooting attitude information and the position information of the first key feature point. The horizontal shooting distance between, including:

Using the position information of the first key feature point, determine the first distance between the first key feature point and the horizontal center line of the image collector;

Using the first distance and the shooting roll angle to obtain a first included angle of the first key feature point relative to the horizontal centerline;

Using the first included angle and the shooting pitch angle, the second included angle formed by the first key feature point and the first vertical point relative to the center point of the image collector is obtained, and the first vertical point Refers to the vertical projection point of the center point of the image collector relative to the ground plane on which the animal to be tested stands;

obtaining the shooting height of the image collector when shooting the target image;

Using the shooting height, the horizontal shooting distance between the center point of the image collector and the animal to be tested is obtained through the operation of the tangent function.

Optionally, according to the horizontal shooting distance and the position information of the second key feature point corresponding to the body size to be measured, obtain the corresponding body size parameters of the animal to be measured, including:

According to the respective position information of any two key feature points of the animal to be tested, the third angle formed by the corresponding two key feature points relative to the center point of the image collector is obtained;

Use the third included angle and the horizontal shooting distance to obtain the feature point distance between the corresponding two key feature points;

Using the obtained feature point distances between at least two second key feature points corresponding to each type of body size to be measured, the parameters of the body size to be measured corresponding to the animal to be measured are obtained.

Optionally, if there is a first height difference between the operating body and the respective surfaces of the animal to be tested, the acquiring the shooting height of the image collector when shooting the target image includes:

calling the height and the first height difference of the operating body that captures the target image;

According to the height of the operating body and the first height difference, the shooting height of the image collector relative to the ground plane where the animal to be tested stands is obtained.

Optionally, performing feature extraction on the target image to obtain multiple key feature point information of the animal to be tested contained in the target image, including:

Inputting the target image into an image feature extraction model to obtain a plurality of key feature points of the animal to be tested contained in the target image;

outputting multiple key feature points of the animal to be tested contained in the target image;

In response to the position adjustment operation for the output at least one key feature point, multiple key feature point information of the animal to be tested is obtained.

Optionally, the method further includes:

Perform integrity verification on the obtained multiple key feature points of the animal to be tested;

If the verification fails, first prompt information is output, and the first prompt information is used to remind the operator to re-shoot the target image of the animal to be tested.

The present application also proposes a device for acquiring external features of animals, the device comprising:

an information acquisition module, used for acquiring a target image of the animal to be tested, and the shooting attitude information when the image collector shoots the target image;

A feature extraction module is used to perform feature extraction on the target image to obtain a plurality of key feature point information of the animal to be tested contained in the target image, and the key feature point information includes the corresponding key feature points in the target image. location information in the image;

The horizontal shooting distance obtaining module is used to obtain the horizontal shooting distance between the center point of the image collector and the animal to be measured according to the shooting attitude information and the position information of the first key feature point. The first key Feature points refer to the key feature points whose distance from the ground is less than the first threshold among the multiple key feature points of the animal to be tested;

The external feature value obtaining module to be tested is used for obtaining the external feature value to be tested corresponding to the animal to be tested according to the horizontal shooting distance and the position information of the second key feature point corresponding to the body size to be tested.

The present application also proposes a computer device, the computer device comprising:

Communication Interface;

A memory for storing a program that implements the above-mentioned method for obtaining an attribute value of an animal;

The processor is configured to load and execute the program, and implement each step of the above-mentioned method for obtaining an attribute value of an animal.

Optionally, the computer device is a mobile terminal or a server;

When the computer device is the mobile terminal, the computer device further includes:

An image collector for taking images of the animal to be tested;

a display screen for outputting the image of the animal to be tested captured by the image collector;

A variety of attitude sensors for sensing the shooting attitude information when the image collector shoots the target image of the animal to be tested;

When the computer device is a server, the communication interface is used to receive the target image of the animal to be tested shot and sent by the image collector, and the shooting posture information when the image collector shoots the target image; The corresponding external characteristic value of the tested animal is fed back to the preset mobile terminal for output.

Based on the above technical solutions, the present application provides a method, device and computer equipment for acquiring external features of animals. When external features such as weight and height of an animal to be measured need to be acquired, in order to avoid manual measurement of the animal to be tested or measurement of users The threat of injury that may be caused, this embodiment proposes to use image analysis to obtain the external characteristic value of the animal to be tested. The shooting attitude information of the target image is obtained by extracting the features of the target image to obtain a plurality of key feature point information of the animal to be tested, such as position information. After that, it can be determined according to the position information of the first key feature point and the above shooting attitude information The horizontal shooting distance between the center point of the image collector and the animal to be tested, so as to obtain the corresponding external feature value to be tested according to the horizontal shooting distance and the position information of the second key feature point corresponding to the external feature to be tested. Since the above-mentioned target image in this embodiment can be captured by an ordinary image collector, a professional depth camera must be used, which greatly reduces the equipment cost for acquiring the external features of the animal, and reduces the professional requirements for the user who captures the image, and The requirement of the site where the animal to be tested is located improves the universality of the method for acquiring the external features of the animal proposed in this application.

Description of drawings

In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the accompanying drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only It is an embodiment of the present application. For those of ordinary skill in the art, other drawings can also be obtained according to the provided drawings without any creative effort.

FIG. 1a is a schematic diagram of the hardware structure of an optional example of the computer device proposed by the application;

FIG. 1b is a schematic diagram of the hardware structure of another optional example of another computer device proposed by this application;

2 is a schematic diagram of an optional application scenario applicable to the method for obtaining external features of animals proposed by the present application;

3 is a schematic diagram of another optional application scenario applicable to the method for obtaining external features of animals proposed by the present application;

FIG. 4 is a schematic diagram of another optional application scenario applicable to the method for obtaining external features of animals proposed by the present application;

5 is a schematic flowchart of an optional example of the method for acquiring external features of animals proposed by the application;

6 is a schematic diagram of a shooting posture of an image collector for shooting an image of an animal to be tested in the method for acquiring external features of an animal proposed by the present application;

7 is a schematic diagram of another optional animal shooting scene suitable for the method for obtaining external features of animals proposed by the present application;

8 is a schematic flowchart of another optional example of the method for acquiring external features of animals proposed by the application;

Fig. 9a is an optional schematic diagram of the target image of the animal to be measured obtained in the animal external feature acquisition method proposed by the application;

9b is an optional schematic diagram of the obtained target image of the animal to be tested in the method for acquiring external features of animals proposed by the application;

10 is a schematic diagram of an optional scene in which the key feature points of the animal to be tested are adjusted in the method for acquiring external features of animals proposed by the application;

11 is a schematic diagram of another optional animal shooting scene suitable for the method for obtaining external features of animals proposed by the present application;

FIG. 12 is a schematic structural diagram of an optional example of the device for obtaining external features of animals proposed by the present application;

13 is a schematic structural diagram of another optional example of the device for acquiring external features of animals proposed by the application;

FIG. 14 is a schematic structural diagram of yet another optional example of the apparatus for acquiring external features of animals proposed in the present application.

detailed description

In order to improve the existing technical problems described in the background technology, it is proposed to use an image analysis method to obtain the external features of animals. Specifically, a depth camera is used to take pictures of the animal to be tested, to obtain the key feature point information on the image of the animal to be tested, and then according to the depth The distance measured by the camera is calculated to obtain the external feature value of at least one external feature of the animal to be tested. The external feature may include but is not limited to the weight, length, height, etc. of the animal to be tested. The cost of cameras is also high and difficult to popularize.

Based on this, the present application hopes to take pictures of the animals to be measured by using the common camera of the electronic equipment that has been popularized at present, and then analyze and process the captured images to obtain the external characteristics to be measured, so as to reduce the equipment cost and avoid the external appearance of the animals. In the process of feature acquisition, the animal to be tested and the measurement personnel are harmed, and it can be applied to various application scenarios. The electronic device may be a terminal with an image acquisition function, such as a smart phone, a tablet computer, a wearable device, etc. This application does not limit the device type of the electronic device, which can be determined according to the actual needs of the application scenario.

It should be noted that with regard to the specific implementation process of how to obtain the external characteristics of the animal to be tested by using the ordinary two-dimensional images captured in the present application, the following will combine the drawings in the embodiments of the present application to describe the technical solutions in the embodiments of the present application. For a clear and complete description, it is obvious that the described embodiments are only a part of the embodiments of the present application, but not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application. Also, for convenience of description, only parts related to the related invention are shown in the drawings. The embodiments in this application and the features in the embodiments may be combined with each other without conflict.

It should be understood that "system", "device", "unit" and/or "module" as used in this application is a method used to distinguish different components, elements, parts, parts or assemblies at different levels. However, other words may be replaced by other expressions if they serve the same purpose.

As shown in this application and in the claims, unless the context clearly dictates otherwise, the words "a", "an", "an" and/or "the" are not intended to be specific in the singular and may include the plural. Generally speaking, the terms "comprising" and "comprising" only imply that the clearly identified steps and elements are included, and these steps and elements do not constitute an exclusive list, and the method or apparatus may also include other steps or elements. The inclusion of an element by the phrase "comprising a..." does not preclude the presence of additional identical elements in the process, method, article of manufacture or apparatus that includes the element.

Wherein, in the description of the embodiments of the present application, unless otherwise stated, “/” means or means, for example, A/B can mean A or B; “and/or” in this document is only a description of the associated object The association relationship of , indicates that there can be three kinds of relationships, for example, A and/or B, can indicate that A exists alone, A and B exist at the same time, and B exists alone. In addition, in the description of the embodiments of the present application, "plurality" refers to two or more than two. The following terms "first" and "second" are only used for descriptive purposes, and should not be construed as indicating or implying relative importance or implying the number of indicated technical features. Thus, a feature defined as "first" or "second" may expressly or implicitly include one or more of that feature.

In addition, flowcharts are used in this application to illustrate operations performed by a system according to an embodiment of the application. It should be understood that the preceding or following operations are not necessarily performed in the exact order. Instead, the various steps can be processed in reverse order or simultaneously. At the same time, other actions can be added to these procedures, or a step or steps can be removed from these procedures.

1a, a schematic diagram of the hardware structure of a computer device provided by an embodiment of the present application is shown. The computer device may include, but is not limited to, a communication interface 11, a memory 12, and a processor 13, where:

In this embodiment of the present application, the number of the communication interface 11 , the memory 12 , and the processor 13 can be at least one, and the communication interface 11 , the memory 12 , and the processor 13 can communicate with each other through a communication bus.

Optionally, the communication interface 11 may include an interface of a communication module, such as a GSM module, a WIFI module, an interface for realizing data communication in a mobile communication network (such as a 5G and 6G network), etc. In practical applications, it can be used to realize communication with other The data interaction of the device, such as the transmission of the image of the animal to be tested, the obtained external characteristics of the animal to be tested, etc.; the communication interface 11 may also include interfaces such as a USB interface, a serial/parallel port, etc., for realizing the internal components of the computer equipment The data interaction between them, such as various intermediate data, input parameters, etc., which are generated or required in the implementation process of the animal external feature acquisition method proposed in this application, can be determined according to the needs of the actual application scenario, and this application does not go into detail. described.

The memory 12 may store a program composed of multiple instructions for implementing the method for acquiring external features of an animal provided by the embodiment of the present application, and the processor 13 calls and loads the program stored in the memory 12, thereby realizing the external animal external feature provided by the embodiment of the present application. For the feature acquisition method and the specific implementation process, reference may be made to, but not limited to, the descriptions of the corresponding embodiments below.

In the embodiment of the present application, the memory 12 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one magnetic disk storage device or other volatile solid-state storage device. The processor 13 may be a central processing unit (Central Processing Unit, CPU), an application-specific integrated circuit (ASIC), a digital signal processor (DSP), an application-specific integrated circuit (ASIC), an off-the-shelf programmable gate Array (FPGA) or other programmable logic devices, etc.

In a possible implementation manner, the memory 12 may include a program storage area and a data storage area, and the program storage area may store an operating system and an application program required for at least one function (such as an image display function) to realize the application proposed in the present application. The data storage area can store the data generated during the use of the computer equipment, such as the captured images of each frame of the animal to be tested, the input shooting parameters of the image collector (such as a camera), The height of the operator who uses the image pickup to capture the image, etc., depends on the situation.

It should be understood that the structure of the computer device shown in FIG. 1a does not constitute a limitation on the computer device in the embodiments of the present application. In practical applications, the computer device may include more or less components than those shown in FIG. 1a , This application does not list them all here.

In some embodiments, the above-mentioned computer equipment may be a mobile terminal with a certain image processing capability, such as a smart phone, a tablet computer, etc. In this case, as shown in FIG. 1b, the computer equipment may further include:

The image collector 14 is used for capturing images of the animal to be tested. The image collector may be a built-in camera of a mobile terminal, such as a common optical camera, and the structure and working principle of the image collector will not be described in detail in this application.

The display screen 15 is used for outputting the image of the animal to be tested captured by the image collector 14 .

In practical applications, the display screen 15 may be a touch display screen, which includes a touch sensing unit capable of sensing touch events on the touch display panel, and a user using the mobile terminal can perform operations such as zooming in, zooming out, Touch operations such as the shooting focus are selected to obtain a target image of the animal to be tested that meets the requirements. The specific implementation process is not described in detail in this embodiment.

Of course, the above-mentioned display screen 15 can also be a non-touch display screen, and the user can use the input device of the mobile terminal to realize the input operation of the content presented on the display screen 15, which includes but is not limited to the above-mentioned various methods of displaying images. operation, depending on the situation.

Various attitude sensors 16 are used for sensing the shooting attitude information when the image collector shoots the target image of the animal to be tested. The attitude sensor 16 may specifically include, but is not limited to, a gyroscope sensor, various angle sensors, position sensors, distance sensors, acceleration sensors, touch sensors, temperature sensors, etc., and may be determined according to the use requirements of the mobile terminal itself. It can be seen that the present application can utilize the gesture sensor of the mobile terminal itself to meet the requirements for acquiring external features of animals, and the specific implementation process can refer to the descriptions in the corresponding parts of the embodiments below, which will not be described in detail here. Wherein, in the case where the computer device is a mobile terminal, the above-mentioned communication interface 11 may include various memory interfaces, earphone interfaces, antenna communication interfaces, etc., which may be determined according to the actual external device communication requirements and internal component communication requirements, and this embodiment does not Detailed description.

It should be understood that the structure of the mobile terminal shown in FIG. 1b does not constitute a limitation on the mobile terminal in this embodiment of the present application. In practical applications, the mobile terminal may include more or less components than those shown in FIG. 1b, or Combining certain components, such as power management module, battery, charge management module, vibration mechanism, indicator light, audio module, speaker (ie horn), microphone, receiver (ie earpiece), mobile communication module, antenna and other input devices, output Equipment, this application does not list them all here.

In still other embodiments, the above-mentioned computer device may also be a server. In this case, the user only needs to upload the target image (which may be a picture or video, etc.) and related parameters of the animal to be tested to the server, and the server will follow the application in this application. The proposed method for acquiring the external features of animals can obtain the external feature values of the corresponding animals to be tested. The equipment that does not need to shoot images of the animals to be tested has high image processing capabilities, which reduces the requirements for the equipment. For example, ordinary cameras, camera etc.

Therefore, in the application scenario where the method for obtaining external features of animals proposed in the present application is realized through data interaction between the image collector and the server, the above-mentioned communication interface 11 is used to receive the data of the animal to be tested captured and sent by the image collector. The target image and the shooting attitude information when the image collector shoots the target image, and at the same time, the obtained external characteristic values of the animal to be tested corresponding to the test are fed back to the preset mobile terminal for output. Based on the above analysis, in the practical application of various landing scenarios, combined with the above analysis, in some embodiments, various image acquisition devices such as various cameras as shown in FIG. After the image of the animal to be tested is obtained, it can be wirelessly transmitted to a computer device (which can be a server, or, if the image of the animal to be tested is collected by a camera deployed on a livestock farm, as shown in Figure 3, the computer device here It can also be an electronic device with data processing capability, but is not limited to this), the computer device obtains the corresponding external feature value of the animal to be tested according to the method for acquiring external features of animals described in the following method embodiments.

In still other embodiments, as shown in the application scenario shown in FIG. 4 , for the application scenario of the method for acquiring external features of animals proposed in the present application, the user can also hold a mobile terminal with a certain image processing capability and a built-in image collector, and shoot to be The image of the animal to be tested is obtained, and the mobile terminal executes the method for acquiring external features of the animal proposed in the present application to obtain and output relevant information about the external features of the animal to be tested.

It can be seen that the computer equipment suitable for the method and apparatus for acquiring external features of animals proposed in this application can be a server or a mobile terminal with an integrated image collector and a certain image processing capability. The program corresponding to the method for acquiring external features of animals provided by the embodiments implements the methods for acquiring external features of animals provided by the embodiments of the present application. The program can be stored in the memory of the computing device and invoked by the processor to realize program functions. The specific implementation process is not Do limit.

From the perspective of a computer device, the method for acquiring external features of an animal provided by the embodiments of the present application will be introduced. The method steps described below can be implemented by executing a corresponding program on the computer device.

Referring to FIG. 5 , a schematic flowchart of an optional example of the method for acquiring external features of animals provided by the embodiment of the present application is shown. The method can be applied to computer equipment. It can be seen from the above analysis that the device type of the computer equipment in the present application is different. For limitation, it can be determined according to a specific application scenario, and the application scenario can refer to, but is not limited to, several scenarios for obtaining external features of animals listed above, and this application only takes this as an example for illustrative description. As shown in FIG. 5 , the method for acquiring external features of animals proposed in this embodiment may include but is not limited to the following steps:

Step S51, acquiring the target image of the animal to be tested, and the shooting posture information when the image collector shoots the target image;

In combination with the description of the technical concept of the present application in the above embodiments, the image capture of the animal to be tested can be realized by a mobile terminal with an image collector, that is, the operator (such as animal management personnel, measurement personnel, etc.) can hold the mobile terminal and start the mobile terminal. The image acquisition device (such as a camera) of the terminal points the image acquisition direction to the current direction of the animal to be tested, so that the animal to be tested appears in the preview box, click the shooting button to obtain the target image of the animal to be tested.

Of course, in some embodiments, the camera deployed in the livestock breeding ground can also capture the image of each animal to be tested, and then transmit the image to a computer device with data processing capability through wireless communication, as described above. The terminal or server, from which the computer equipment obtains the target image of the animal to be tested, etc. The specific acquisition method of the target image of the animal to be tested in this application is not limited, and can be determined according to the actual application scenario, and is not limited to the above listed in this application. Method to realize.

It should be noted that in the process of image capturing of the animal to be tested, multiple frames of images of the animal to be tested may be continuously captured (such as continuous multiple frames of pictures or videos. For video shooting, computer equipment can extract the corresponding images of the animal to be tested from it. Continuous multi-frame pictures, the realization process will not be described in detail), but because the application requires the feature points of the whole body of the animal to be tested, and the captured image needs to contain a complete image of the animal to be tested, the above-mentioned target image in this embodiment can be Refers to the target image that contains the complete animal to be tested. For example, in the above scenario where the operator holds the mobile terminal to capture the image, it can be determined that the entire body of the animal to be tested appears in the image acquisition preview frame, that is, the preview image displayed on the display screen of the mobile terminal contains the complete image to be tested. When there is an animal, confirm the shooting again, and determine the image captured at this time as the target image of the animal to be tested; for other scenarios, you can also directly select a frame of images containing the complete animal to be tested from the multiple frames of images of the animal to be tested For the target image, this application will not describe them in detail here.

Among them, in the process of taking the image of the animal to be tested, since the relative height between the image collector and the animal to be tested is uncertain, it is often affected by the height of the animal to be tested, the operator (ie the user who takes the image) and the height of the animal to be tested. The height difference on the ground where the animal is located, the distance between the operator and the animal to be measured, the height of the operator, the installation height of the image collector (mainly refers to the scene of the camera in the livestock breeding place) and other factors, in order to shoot To complete the target image of the animal to be tested, it may be necessary to adjust the shooting angle, shooting height, focal length and other parameters of the image collector. Therefore, the shooting parameters of the image collector have a certain relationship with the relative positional relationship between the operator and the animal to be tested in the current shooting scene. In this way, when the application needs to predict the horizontal distance between the image collector and the animal to be tested It can be realized in combination with the relevant shooting parameters of the image collector at the time of shooting.

Through experiments or analysis, it is known that in the embodiment of the present application, when the image collector captures the target image of the animal to be tested, the shooting parameters such as the shooting posture information of the image collector may affect the image acquisition when the target image is captured. The horizontal distance between the camera and the animal to be tested. Therefore, in this embodiment, it is necessary to obtain the shooting attitude information when the image collector shoots the target image, which may specifically include the shooting pitch angle, shooting roll angle, shooting zoom ratio, etc. of the image collector, and the content contained in this application is not limited. , as the case may be.

Among them, the shooting pitch angle (Pitch) of the image collector can also be called the tilt angle, as shown in FIG. The rotation angle of the horizontal centerline of the image collector) can change the pitch angle of the image collector. In practical applications, the shooting pitch angle is usually determined according to the horizontal distance between the animal to be measured and the image collector, as well as the height of the animal to be measured itself, so as to ensure that the image collected by the image collector can contain complete Of course, in this process, the image collector can be zoomed to adjust the display size of the animal to be measured in the entire image preview frame to meet the shooting requirements. Specifically, the shooting pitch can be determined according to the needs of the actual shooting scene. angle, which is not described in detail in this application.

The shooting roll angle (roll) can refer to the rotation angle of the image collector around the Z axis (which usually refers to the acquisition direction of the image collector, that is, the direction of the lens), as shown in Figure 6, in the actual shooting process, for the image acquisition For the mobile terminal of the image collector, the image collector can rotate around the axis perpendicular to the screen of the mobile terminal; in the same way, the heading angle Yaw of the image collector can refer to the rotation angle of the image collector around the Y axis, that is, the image collector rotates around the Y axis. The angle by which the electronic device is rotated in the upward direction.

In practical applications, the shooting attitude information such as the shooting pitch angle and shooting roll angle described above can be determined by the parameters sensed by the mobile terminal with the image collector or the various sensors of the independent image collector. The sensors may include gyroscopes, various angle sensors, position sensors, etc. The type and number of sensors may be determined according to actual detection requirements, but are not limited to the detection methods described in this embodiment.

Step S52, perform feature extraction on the target image to obtain multiple key feature point information of the animal to be tested contained in the target image;

In the embodiment of the present application, the key feature point information includes the position information of the corresponding key feature point in the target image. Since the key feature point may be a certain pixel point of the animal to be tested in the target image, the corresponding position information can be obtained from the The three-dimensional coordinates of the pixel in the image coordinate system (such as the color values of R, G, and B in the RGB coordinate system, but not limited to this) are determined, or the coordinates of the pixel in the image coordinate system are converted to The three-dimensional coordinates in the camera coordinate system, etc., may depend on the situation, and the present application does not limit the content included in the information of each key feature point.

The key feature points of the animal to be tested may refer to feature points that can represent different external features of the animal to be tested, usually located on the body surface of the animal to be tested, that is, the edge feature points belonging to the animal to be tested in the target image (It refers to the edge feature points of the three-dimensional image), which can be specifically determined according to the category of the animal to be measured and the category of external features to be measured (such as the type of body size parameters to be measured, the category of attributes to be measured) and other information. To obtain each external feature value of the animal to be tested, it is necessary to know the position of the animal to be tested, so that the feature point corresponding to the position is recorded as a key feature point.

It can be seen that for animals to be tested of different categories and sizes, the positions and numbers of the corresponding key feature points may be different. This application can be pre-determined according to the above method and combined with the actual measurement requirements. In the target image of the animal to be tested, when extracting the key feature points of the animal to be tested, it is possible to know which position of the animal to be tested needs to be extracted directly according to the preset correspondence between each animal to be tested and each of the key feature points. The feature points are key feature points, and the specific implementation process is not limited.

Exemplarily, in combination with the application scenarios shown in Figures 2 to 4 above, if the animal to be tested is a yak, it is necessary to measure the yak's height, body length and other external characteristic values (here can refer to the body size parameter), in order to obtain the yak. The body size parameter of height needs to know the distance between the highest point in the vertical direction of the yak (that is, the highest point on the yak's back) and the lowest point (that is, the contact point between the yak's hoof and the ground). Therefore, after obtaining the target image of the yak, In the target image, the feature points with a higher position among the feature points on the back of the yak can be recorded as key feature points, and the feature points where the yak hooves are in contact with the ground can be recorded as key feature points. It should be understood that in order to improve the reliability of image analysis , in determining various key feature points, several feature points adjacent to each other can be selected as key feature points, and then further screened to avoid omission and reduce the detection accuracy.

Similarly, in order to obtain the body size parameter of the yak's body length, according to the above analysis method, one or more feature points with the most protruding yak's mouth in the obtained yak's target image can be determined as key feature points, and the yak's tail can be determined as the key feature point. The most prominent one or more feature points are determined as key feature points, and the distance between the most distant key feature points among the two types of key feature points is calculated in turn to obtain the body length of the yak. The process of acquiring other body size parameters of the yak to be measured and the process of acquiring each of the body size parameters to be measured of other animals to be tested are similar, and will not be described in detail in this application.

In some embodiments, the present application can use a preset feature extraction algorithm to process the target image of the animal to be tested to obtain key feature points and location information of the animal to be tested in the target image. The test animal trains a corresponding feature extraction model. In this way, after obtaining the target image of the animal to be tested, the target image is directly input into the feature extraction model for processing, and the key feature point information of the animal to be tested is output. The specific implementation process is not limited and can be determined according to actual application requirements.

Step S53, obtaining the horizontal shooting distance between the center point of the image collector and the animal to be measured according to the shooting attitude information and the position information of the first key feature point;

It should be noted that the first key feature point in this embodiment refers to the key feature point whose distance from the ground is less than the first threshold among the multiple key feature points of the animal to be tested. For the feature points on the yak's hooves that touch the ground and/or their adjacent feature points, the location and the number of the first key feature points can be determined according to the category of the animal to be tested. It should be understood that the distance between the first key feature point and the ground may be zero or a small value, therefore, the first threshold is usually a relatively small value, but the specific value of the first threshold is not limited in this application.

Combined with the above description of the shooting attitude information, during the shooting process of the animal to be tested, the adjusted shooting attitude information of the image collector is often based on the horizontal shooting distance between the center point of the image collector and the animal to be tested, and the Factors such as the height of the measured animal are determined, and reverse reasoning. In the embodiment of the present application, the shooting attitude information and the position information of the first key feature point can be used to obtain the shooting target image through calculation. The level between the animal to be tested and the center point of the image collector Shooting distance, in the process of acquiring the horizontal shooting distance, can be realized according to actual needs in combination with corresponding auxiliary parameters, such as the height of the operator to shoot the image of the animal to be measured by holding an electronic device with an image collector, and the independence of shooting the image of the animal to be tested. The specific acquisition process of the horizontal shooting distance is not limited in this application.

Wherein, referring to the schematic diagram of the animal shooting scene shown in FIG. 7 , assuming that the image of the animal to be tested is shot by the operator holding the mobile terminal, the horizontal shooting distance calculated according to the above method may refer to the first distance between the center point of the image collector and the animal to be tested. The horizontal distance GP between the key feature points represents the ground distance between the image collector and the animal to be measured, not the distance in the X-axis direction of the image collector coordinate system.

Step S54, according to the horizontal shooting distance and the position information of the second key feature point corresponding to the external feature to be tested, obtain the external feature value to be tested corresponding to the animal to be tested;

Following the description of the corresponding part above, for different external features of the animal to be tested, the position information of the key feature points located at different positions of the animal to be tested may be different. The key feature point is recorded as the second key feature point, and the second key feature point may include the above-mentioned first key feature point and other key feature points, which may be specifically determined according to the measurement requirements of each external feature to be measured. This embodiment of the present application I won't go into detail here.

Based on this, for any external characteristic value to be measured of the animal to be measured, such as the body size parameter to be measured, the attribute value to be measured (such as weight), according to the preset correspondence, a second external characteristic value corresponding to the external characteristic value to be measured is obtained. After the position information of the key feature points, since the number of the second key feature points corresponding to one external feature value to be measured is often multiple, such as two, three or even more, this embodiment can The position information of the second key feature point is calculated to obtain the corresponding external feature value to be measured.

It should be understood that this application needs to obtain the real external feature value of the animal to be tested, not the external feature value in the image. Therefore, in the operation process, the coordinate system can be converted to obtain the corresponding position of the real animal to be tested. The position information of the second key feature point, so that the obtained external feature value to be measured is the actual external feature value of the animal to be tested. The actual external characteristic value of the animal to be tested can be obtained by processing such as coordinate conversion, and this application will not describe the conversion process in detail.

In some embodiments, combined with the above analysis, the external characteristics of the animal to be tested may include at least one of the body measurements of the animal to be measured (such as the above-mentioned height, length, body oblique length, chest circumference, etc.), and/or at least An attribute to be measured (such as weight). In practical applications, the attribute value of the animal to be measured is usually calculated based on the corresponding body size parameters. Therefore, in the actual measurement, the body size parameters of the animal to be measured can be obtained first, and then further based on this. Get the property value to be tested.

Based on this, the above-mentioned step S54 may specifically include: obtaining the body size parameters to be measured corresponding to the animal to be measured according to the horizontal shooting distance and the position information of the second key feature point corresponding to the body size to be measured, and according to the obtained at least one The body size parameter to be measured is obtained, and the corresponding attribute value to be measured of the animal to be measured is obtained.

In this optional embodiment, in combination with the description of the correspondence between the external feature value of the animal to be measured and the key feature points, the external feature value of the body size parameter to be measured is the corresponding relationship between the external feature value and the key feature points. Similarly, it is possible to pre-determine the key feature points of the positions required to obtain various parameters of the body measure to be measured, that is, to establish the correspondence between the various body measures to be measured and the key feature points of different positions of the animals to be measured, so that, In the actual measurement application of the body size parameters of the animal to be tested, each of the body size parameters of the animal to be tested can be obtained first according to the above method.

Afterwards, since there is often a certain operational relationship between the body size parameters of the object to be measured and the properties to be measured (such as expressing the operational relationship in a formula, etc.), this can be determined based on actual measurement experience. In this way, in this embodiment, the above method is used to obtain After at least one body size parameter of the animal to be measured, the value of the corresponding attribute to be measured of the animal to be measured can be further estimated according to the predetermined calculation relationship, such as the weight of the animal to be measured, etc. The method of obtaining the operation relationship between the body size parameters of the object to be measured and the attributes to be measured, and the expression method of the operation relationship are not limited, and can be determined according to actual needs.

It should be understood that, for different properties of the animal to be tested, the body size parameters to be measured required to obtain the value of the property to be measured may be different, which can be determined according to actual measurement experience or experiments. The position and number of the second key feature points of the animal to be measured are often different. Therefore, in practical applications, it can be determined according to the properties to be measured of the animal to be measured this time. Obtain the body size parameter to be measured required for the attribute value to be measured, and then determine the second key feature point and other parameters required for calculating the body size parameter to be measured. After that, the second key feature can be obtained according to the above analysis method position information and other parameters of the point.

To sum up, in the embodiment of the present application, when it is necessary to obtain the external characteristics of the animal to be tested, such as weight, height, body length, etc., in order to avoid the threat of injury that may be caused by manual measurement of the animal to be tested or measurement of the user, This embodiment proposes to use image analysis to obtain the external features of the animal to be tested. Specifically, an image collector is used to obtain the target image of the animal to be tested, and the shooting attitude information when the image collector captures the target image. Feature extraction is performed on the target image to obtain a plurality of key feature point information of the animal to be tested, such as position information. After that, the center point of the image collector and the to-be-to-be-measured center point can be determined according to the position information of the first key feature point and the above shooting attitude information. The horizontal shooting distance between the animals is measured, so as to obtain the corresponding external feature to be measured according to the horizontal shooting distance and the position information of the second key feature point corresponding to the body size to be measured. Since the above-mentioned target image in this embodiment can be captured by an ordinary image collector, a professional depth camera must be used, which greatly reduces the equipment cost for acquiring the external features of the animal, and reduces the professional requirements for the user who captures the image, and The requirement of the site where the animal to be tested is located improves the universality of the method for acquiring the external features of the animal proposed in this application.

Referring to FIG. 8 , a schematic flowchart of another optional example of the method for acquiring external features of animals provided by the embodiment of the present application is shown. This embodiment may be an optional refinement implementation of the method for acquiring external features of animals proposed in the above embodiments. However, it is not limited to the detailed implementation manner described in this embodiment. As shown in FIG. 8 , the method may include:

Step S81, acquiring the target image of the animal to be tested, and the shooting pitch angle and shooting roll angle when the image collector shoots the target image;

For the specific implementation process of step S81, reference may be made to the descriptions in the corresponding parts of the foregoing embodiments, and details are not repeated here.

Among them, regarding the shooting pitch angle and shooting roll angle, combined with the description of their meaning and the acquisition method in the above embodiment, if the image collector shoots the target image, there is a certain shooting pitch angle, as shown in FIG. 7, which will lead to The horizontal center line (ie X axis) of the image collector is no longer in a parallel relationship with the horizon, but in an intersecting relationship; if there is a certain shooting roll angle, the image of the animal to be tested in the captured image is inclined, as shown in Figure 9a Instead of the horizontal state shown by the image of the animal to be tested obtained by horizontal shooting as shown in FIG. 9b, the parameter calculation can be carried out in a corresponding way according to the actual state of the animal to be tested in the target image. Perform tilt calibration on the target image to obtain a calibrated image as shown in FIG. 9b , and then perform subsequent calculations, which are not limited in this application, and may be determined according to circumstances.

During the shooting process of the animal to be tested by the image collector, the complete image of the animal to be tested needs to be captured due to the need for feature point information of different positions of the animal to be tested. The image preview interface of the image collector or computer equipment can display the current shooting lens, The captured preview image of the image to be tested, the shooting user can adjust the shooting angle, distance, etc. of the image collector according to actual needs, so that the animal to be tested is in the frontmost central position in the preview image, that is, the animal to be tested is in focus The center position, and the complete animal to be tested is photographed, and then the shooting button is pressed to obtain the target image of the animal to be tested, but it is not limited to this implementation.

In still other embodiments, for each frame of the acquired preview image, the computer device can use an edge detection algorithm to acquire contour information (ie, edge information) of the animal to be tested in the preview image, and determine whether the acquired contour information is Meet the shooting requirements, such as whether it contains the complete contour information of the animal to be tested, if so, determine the frame preview image as the target image, if not, instruct the image collector to adjust the shooting according to the matching degree between the acquired contour information and the shooting requirements Attitude information, such as adjusting the shooting angle, shooting distance, zoom magnification (that is, using the digital zoom or optical zoom function of the image collector and its computer equipment to adjust to change the size of the animal to be tested in the entire preview image, and the specific zoom adjustment process) This embodiment does not describe in detail), etc., so that the image collector with the adjusted post-shooting attitude information, the profile information of the animal to be tested contained in the re-obtained preview image meets the shooting requirements, and the specific implementation process is not detailed in this embodiment. described.

It should be noted that this application does not limit the specific content of the edge detection algorithm on which the contour information of the animal to be measured is obtained, such as target detection algorithm based on semantic segmentation, edge detection algorithm based on search, edge detection algorithm based on zero crossing, etc. , which can be flexibly selected according to the actual application requirements, which are not described in detail in this application.

Step S82, input the target image into the image feature extraction model to obtain multiple key feature points of the animal to be tested contained in the target image;

In this embodiment, the image feature extraction model can be obtained by training the sample images of each sample animal based on a machine learning algorithm, and is used to extract the key feature points of any sample animal. The machine learning algorithm can be flexible according to the needs of the actual application scenario. Selection, such as a deep neural network, etc., this application does not describe in detail the type of machine learning algorithm on which the image feature extraction model is trained, and the specific training process of the model.

In some embodiments proposed in this application, in order to improve the reliability of the method for acquiring external features of animals, it is necessary to acquire complete key feature points of the animal to be tested. Therefore, when extracting multiple key feature points of the animal to be tested contained in the target image After that, integrity verification can be performed on the obtained multiple key feature points. If the verification passes, the subsequent steps can be continued; if the verification fails, the first prompt message can be output to remind the operator (ie, the shooting user) to shoot again. The target image of the animal to be tested.

Specifically, in a possible implementation manner, the present embodiment may obtain the multiplicity of data extracted this time according to the confidence of each key feature point, the topological position relationship formed by the respective position information of the multiple key feature points, and the like. The overall score of each key feature point, if the overall score reaches the preset score, it can be considered that the key feature points on the animal to be tested are completely extracted this time; otherwise, the overall score does not reach the preset score, it can be considered that the mentioned The multiple key feature points of the test are not complete enough to calculate the body size parameters of the animal to be tested. In this case, the corresponding prompt information can be output to remind the shooting user to re-collect the image of the animal to be tested, and According to the above method, feature extraction and verification are performed on the collected target image again to ensure that the mentioned multiple key feature points of the animal to be tested are complete.

It should be noted that the implementation method of how to verify whether the mentioned multiple key feature points of the animal to be tested are complete is not limited to the implementation method described above in this embodiment, and can be flexibly adjusted according to actual application requirements. I won't go into detail here.

Step S83, outputting multiple key feature points of the animal to be tested contained in the target image;

Step S84, in response to the position adjustment operation for the output at least one key feature point, obtain a plurality of key feature point information of the animal to be tested;

In the practical application of this embodiment, for the key feature points of the animal to be tested extracted from the target image above, the identification may not be accurate enough. In order to ensure the reliability of subsequent parameter calculation using the position information of the key feature points , the present application proposes to calibrate the identified key feature points, but is not limited to the manual calibration solution described in this embodiment.

According to the above analysis, in the calibration implementation process of the key feature points proposed in this embodiment, the identified key feature points of the animal to be tested may be output first. Referring to the output schematic diagram of the key feature points of the animal to be tested shown in FIG. 10 , In the output target image containing the animal to be tested, the extracted key feature points of the animal to be tested can be represented by black dots, and the user can intuitively see the extracted key feature points through the display interface of the mobile terminal Where is the position on the animal to be tested? According to experience, if it thinks that the position of one or more key feature points currently proposed is not accurate enough, it can select the key feature point and drag it to a more accurate position. As shown in Figure 10, the user can click the key feature point with inaccurate position with his finger, move the position of the key feature point in the direction of the arrow until it reaches the position that the user thinks is more accurate, release the finger, and realize the key feature point. Position adjustment.

In the process of adjusting the position of the above key feature points, for the electronic device, it can respond to the position adjustment operation for at least one output key feature point, and obtain the corresponding key feature point information according to the position information of the adjusted key feature point. , the content contained in the key feature point information is not limited in this application, which may be determined according to the situation.

It should be noted that the implementation of the position of the key feature points extracted by manual calibration proposed in this application is not limited to the adjustment methods described in the above embodiments, and input devices (such as a mouse and a keyboard) of an electronic device can also be used. The corresponding function keys (such as the direction keys of the keyboard, the preset click operation of the mouse, etc.) are realized, which can be determined according to the actual application scenario, or can be determined by the user according to the personal operation habits, which is not limited in this application.

In addition, in practical applications, when the accuracy of the output results of the image feature extraction model pre-trained by the present application is sufficient, after the present application obtains a plurality of key feature points of the animal to be tested contained in the target image, it is also possible to directly obtain the corresponding The key feature point information is no longer required to perform position calibration on the extracted key feature points.

Step S85, using the position information of the first key feature point to determine the first distance between the first key feature point and the horizontal center line of the image collector;

Combining with the description of the corresponding part of the above embodiment, since the first key feature point refers to the key feature point whose distance from the ground is less than the first threshold value among the multiple key feature points of the animal to be tested, with reference to Figure 7 above, it is assumed that the first key feature point is A key feature point is G, the horizontal center line of the image collector is the X-axis in FIG. 7 , starting from the first key feature point G, a ray perpendicular to the X-axis is generated, and the first key feature point G is associated with the ray The distance between the intersection point with the X axis is determined as the first distance; with reference to Figure 9a and Figure 9b, the first distance can be the vertical distance from the pixel point g to the X axis, and the specific acquisition method of the first distance in the present application is not Do limit.

Step S86, using the first distance and the shooting roll angle to obtain the first included angle of the first key feature point relative to the horizontal center line of the image collector;

In conjunction with the above description of the shooting roll angle, with reference to the schematic diagram of the scene shown in FIG. 7 , after obtaining the first distance from the first key feature point to the horizontal centerline of the image collector, the present embodiment can be based on different coordinate systems ( Such as the coordinate conversion relationship between the image coordinate system, the world coordinate system, etc.), using the first distance and the shooting roll angle to obtain the first angle ∠ of the first key feature point G relative to the horizontal center line OX of the image collector XOG. According to actual needs, the present application can realize the calculation of the above-mentioned first included angle in combination with the configuration parameters currently possessed by the image collector, such as image resolution, zoom magnification, etc. limited.

Step S87, using the first included angle and the shooting pitch angle to obtain the second included angle formed by the first key feature point and the first vertical point relative to the center point of the image collector;

Among them, the first vertical point refers to the vertical projection point of the center point of the image collector relative to the ground plane where the animal to be tested stands, such as the point P shown in Figure 7, the horizontal center line OX and the vertical center line OY of the image collector The formed angle ∠XOY is a right angle. Therefore, as shown in FIG. 7 , the second angle ∠GOP formed by the first key feature point G and the first vertical point P relative to the center point O of the image collector is the same as that obtained above. The first included angle ∠XOG and the shooting pitch angle ∠YOP constitute ∠XOY, and the second included angle ∠GOP=90°-∠YOP-∠XOG. It should be understood that the first included angle, the shooting pitch angle, and the second included angle in step S87 are all determined by converting to the same coordinate system, so as to ensure that these angles can be reliably performed in step S87 according to the above-mentioned angular relationship.

Step S88, acquiring the shooting height of the image collector when shooting the target image;

In the process of image shooting, the animal to be tested and the user usually stand on a basically flat ground, and the distance between the two is 5 meters to 10 meters (but not limited to this, and can be adjusted according to actual needs). Specifically, it can be determined according to the habit, body shape and other parameters of the animal to be tested. At this time, it can be considered that the user who took the image and the animal to be tested are standing on the same ground plane, as shown in FIG. 7 . The shooting height of the image collector is determined according to the height of ordinary people. For example, if the user's height is between 1.6 meters and 1.8 meters (the user's height can be adjusted according to the actual situation), the height of the image collector when shooting images is usually the same as that of the user's eyes. The heights are similar. Therefore, this application can consider that the shooting height of the image collector is the height of the user minus 0.1 meters (but not limited to this, it can be adjusted according to the actual situation, and this application only uses this as an example to illustrate), That is, between 1.5 meters and 1.7 meters, but it is not limited to this. According to the actual situation, the shooting height when the current user uses the image collector to shoot the image of the animal to be tested can also be input in advance.

However, in some shooting scenarios, there may be a height difference between the ground on which the user taking the image stands and the ground on which the animal to be tested stands, which is recorded as the first height difference. As shown in FIG. 11 , the first height difference can be determined in advance. The height difference is input as a configuration parameter of the image collector, so that when the captured target image is processed in the above manner, the height difference can be combined with the processing. It should be noted that, in this shooting scene, the shooting height of the image collector obtained in this embodiment may be a relative shooting height, that is, the height of the center point O of the image collector relative to the ground plane on which the animal to be tested stands, as shown in the figure The shooting height OP in 11 includes the above-mentioned first height difference between different standing ground planes.

In a possible implementation manner, the above step S88 may include: calling the height and the first height difference of the operating body (that is, the above-mentioned user) of the shooting target image, and obtaining the image capture according to the height and the first height difference of the operating body The shooting height of the monitor relative to the ground plane where the animal to be tested stands, but is not limited to this implementation.

Step S89, using the shooting height to obtain the horizontal shooting distance between the center point of the image collector and the animal to be measured through the operation of the tangent function;

Referring to the shooting scenes shown in Figure 7 and Figure 11, combined with the operation principle of the trigonometric function, the center point of the image collector can be obtained according to the calculation formula of the horizontal shooting distance GP=(H-0.1+m)*tan(∠GOP) The horizontal shooting distance GP between O and the animal to be tested. In this calculation formula, H can represent the height of the user who shoots the image of the animal to be tested; 0.1 can represent the height difference between the user and the image collector; m can represent the first Height difference, it should be understood that in the case where there is no first height difference between the ground on which the user taking the image stands and the ground on which the animal to be tested stands, m=0; tan( ) represents a tangent function.

It should be noted that the calculation formula for calculating the horizontal shooting distance is not limited to the calculation formula listed above, and can be flexibly adjusted according to the requirements of actual application scenarios, which will not be described in detail in this application.

Step S810, according to the respective position information of any two key feature points of the animal to be tested, obtain a third angle formed by the corresponding two key feature points relative to the center point of the image collector;

In this embodiment, after the horizontal shooting distance GP from the center point O of the image collector to the animal to be measured is obtained in the above-mentioned manner, it can be used as a basic parameter to realize the calculation of various external characteristics of the animal to be measured. Specifically, in combination with the above Analysis, for a certain external feature to be tested, determine the two key feature points required to obtain the external feature to be tested, and then use the respective position information of the two key feature points to calculate the relative relationship between the two key feature points. The third included angle formed by the center point O of the image collector.

Exemplarily, referring to the schematic diagram of the shooting scene shown in FIG. 7 , for the key feature points A and B (which may correspond to the key feature points a and b shown in FIG. 9 a and FIG. 9 b ), the corresponding third angle is expressed as ∠AOB. In the same way, for any other two key feature points on the animal to be tested, the acute angle formed by connecting the center point O of the image collector can also be used as the third angle in this way. The position information of the key feature points, and the realization process of calculating the corresponding third included angle will not be described in detail.

Step S811, using the third included angle and the horizontal shooting distance to obtain the feature point distance between the corresponding two key feature points;

Following the above analysis, still take the third angle ∠AOB shown in Figure 7 as an example to illustrate, on the basis of knowing the shooting horizontal distance GP between the animal to be tested and the center point O of the image collector, the trigonometric function can be combined. According to the operation principle, the feature point distance AB=GP*sin(∠AOB) between the two key feature points is calculated, and the specific calculation process of the feature point distance between the two key feature points is not detailed in this embodiment. described.

Step S812, using the obtained feature point distances between at least two second key feature points corresponding to each type of body size to be measured, to obtain body size parameters to be measured corresponding to the animal to be measured;

Step S813, according to the obtained at least one body size parameter to be measured, obtain the corresponding attribute value to be measured of the animal to be measured.

In combination with the relevant description of the relationship between the body size parameters to be measured and the key feature points of the animal to be measured in the above embodiment, the present application can determine the corresponding second key feature points according to the body size parameters to be measured that actually need to be measured, and then , and using the determined feature point distances between the second key feature points, the body size parameter to be measured is obtained by calculation. For the key feature points A and B exemplified above, the bust size of the animal to be measured can be calculated according to the distance between the feature points; as shown in the key feature points C and D in Figure 10, the key feature points can be obtained in the above manner The distance between the feature points between C and D can be used to obtain the body length of the animal to be tested, etc. This application will not give examples one by one.

It should be understood that, for some body size parameters of the animal to be tested, it may be necessary to obtain the feature point distance between multiple second key feature points, and it is not limited to the above-mentioned two second key feature points. The distance between the feature points between the two can be directly calculated to obtain the parameters of the body size to be measured, which can be determined according to actual needs, and this application does not describe them in detail.

After the body size parameters of the animal to be tested are obtained, the corresponding attribute values to be tested can be further calculated according to the empirical formula between the body size parameters of the animal to be tested and the attributes to be tested. The calculation relationship between the body size parameter and the body weight of the animal to be measured, and then after obtaining the corresponding body size parameter to be measured according to the above method, the body weight of the animal to be measured is obtained according to the calculation relationship, and the attribute value acquisition method for other categories of attributes Similarly, this application does not describe them in detail.

To sum up, in this embodiment, when it is necessary to obtain external characteristics such as weight and height of the animal to be measured, in order to avoid the threat of injury that may be caused by manual measurement of the animal to be measured or the user, this embodiment proposes to use The method of image analysis realizes the non-contact measurement of the external features of the animal to be tested, and also gets rid of the heavy manual labor caused by the manual measurement method.

Specifically, in order to reduce equipment costs and professional requirements for photographers, an ordinary optical camera (that is, the above-mentioned image collector) can be used to capture the image of the animal to be tested, and then use the relevant shooting parameters of the captured target image, and the The position information of the key feature points of the animal to be tested in the target image is calculated to obtain the shooting horizontal distance between the animal to be tested and the center point of the image collector, so as to obtain the feature point distance between the two key feature points, and then The body size parameters to be measured and the attribute values to be measured are obtained by calculation, which meets the measurement requirements of the external features of the animal to be measured in the application scenario, and improves the universality of the method for obtaining the external features of the animal proposed in the present application.

In combination with the above embodiments, the scheme of the method for obtaining external features of animals proposed in the present application is described. In order to understand the scheme more clearly, the following will take the computer equipment as an example of photographing the user's mobile phone and measuring the weight of a yak. It should be understood that for The computer equipment that implements the scheme of this application includes but is not limited to mobile phones, and the animals to be tested include but are not limited to yak, but also other large animals such as camels, horses, elephants, etc. The properties of the animals to be tested are tested. (That is, the external characteristic value of the animal's body) includes but is not limited to the body weight, which can be determined according to the actual animal measurement requirements, which will not be described in detail in this application.

Combined with the schematic diagram of the application scenario shown in Figure 4 above, when the weight of the yak needs to be measured, the user can use the camera of the mobile phone to aim at the yak to be tested, and check whether the currently collected yak image is through the preview frame of the display screen. Meet the measurement requirements, such as whether the yak to be tested is located in the center of the preview frame, and contains the complete outline of the yak to be tested, if the measurement requirements are met, press the shooting button, and the obtained image will be used as the target image. The extraction model performs feature extraction on it, and displays multiple key feature points of the extracted yak in the image, such as the 16 key feature points shown in Figure 10 above, but is not limited to this. During this period, the user can check whether the displayed yak's key feature points are accurate. If not, they can manually adjust the positions of the corresponding key feature points to obtain an ideal image of the yak to be tested. The adjustment process can be referred to but It is not limited to the description of the corresponding parts of the above embodiments. Of course, the mobile phone can also use the default parameters without performing the above adjustment steps, and directly perform subsequent processing on the obtained image, which can be determined according to the actual situation.

After that, the mobile phone can use the shooting posture information when shooting the above-mentioned ideal target image (the content of which can refer to the description of the corresponding part of the above-mentioned embodiment), such as the user's height H, the distance between the user and the ground plane where the animal to be tested is standing. Parameters such as the first height difference (which can be controlled within 10 centimeters or determined according to the actual situation), the height difference between the height of the camera and the height of the user (such as the above 0.1 meters, but not limited to this), and the parameters to be Measure the position of the contact point between the yak and the ground plane, etc., according to the processing method described in the above embodiment, estimate the shooting horizontal distance between the yak to be measured and the user, and then combine the extracted position information of each key feature point, Calculate the body size parameters of the yak to be tested, and then further estimate the weight of the yak to be tested.

It should be understood that the process of acquiring other attribute values of the yak to be tested and the process of acquiring the values of various external characteristics of other animals to be tested are similar, and will not be described in detail in this application. Moreover, in combination with the descriptions of the application scenarios of the method for obtaining external features of animals proposed in the present application, it is not limited to this application scenario of using a mobile phone to shoot the cattle to be tested. It can be adaptively adjusted according to actual needs. The essential methods of external features are similar, and this application does not list them all.

Referring to FIG. 12 , it is a schematic structural diagram of an optional example of the device for obtaining external features of animals proposed in the present application. The device can be applied to the above-mentioned computer equipment. As shown in FIG. 12 , the device can include:

The information acquisition module 121 is used to acquire the target image of the animal to be tested, and the shooting posture information when the image collector shoots the target image;

The feature extraction module 122 is configured to perform feature extraction on the target image to obtain a plurality of key feature point information of the animal to be tested contained in the target image, and the key feature point information includes the corresponding key feature points in the location information in the target image;

In a possible implementation, the feature extraction module 122 may include:

A model processing unit, configured to input the target image into an image feature extraction model to obtain a plurality of key feature points of the animal to be tested contained in the target image;

a key feature point output unit for outputting a plurality of key feature points of the animal to be tested contained in the target image;

The position adjustment unit is configured to obtain multiple key feature point information of the animal to be tested in response to the position adjustment operation for the output at least one key feature point.

Further, in order to improve processing reliability and accuracy, the device for acquiring external features of animals proposed in this application may also include:

The verification module is used to verify the integrity of the obtained multiple key feature points of the animal to be tested; if the verification is passed, the follow-up process is performed;

The prompt information output module is configured to output first prompt information when the result of the verification module is that the verification fails, and the first prompt information is used to remind the operator to re-shoot the target image of the animal to be tested.

The horizontal shooting distance obtaining module 123 is used to obtain the horizontal shooting distance between the center point of the image collector and the animal to be measured according to the shooting posture information and the position information of the first key feature point, the first The key feature points refer to the key feature points whose distance from the ground is less than the first threshold among the multiple key feature points of the animal to be tested;

The external feature value obtaining module 124 to be measured is configured to obtain the external feature value to be measured corresponding to the animal to be measured according to the horizontal shooting distance and the position information of the second key feature point corresponding to the body size to be measured.

In some embodiments, since the external feature to be tested may include at least one body size to be measured and/or at least one attribute to be tested of the animal to be tested, therefore, as shown in FIG. 13 , the above-mentioned external feature value to be tested Obtaining module 124 may include:

a body measure parameter obtaining unit, configured to obtain the body measure parameter corresponding to the animal to be measured according to the horizontal shooting distance and the position information of the second key feature point corresponding to the body measure to be measured;

The attribute value obtaining unit is configured to obtain the corresponding attribute value to be measured of the animal to be measured according to the obtained at least one body size parameter to be measured.

In some embodiments proposed in this application, the above-mentioned shooting attitude information may include shooting pitch angle, shooting roll angle, shooting zoom magnification and other information. For the meaning and acquisition method of each shooting attitude information, please refer to the corresponding parts of the above-mentioned embodiments. description will not be repeated. Based on this, as shown in FIG. 13 , the above-mentioned horizontal shooting distance obtaining module 123 may include:

a first distance determination unit 1231, configured to use the position information of the first key feature point to determine the first distance between the first key feature point and the horizontal center line of the image collector;

The first included angle obtaining unit 1232 is used to obtain the first included angle of the first key feature point relative to the horizontal center line by using the first distance and the shooting roll angle;

The second included angle obtaining unit 1233 is configured to use the first included angle and the shooting pitch angle to obtain the second angle formed by the first key feature point and the first vertical point relative to the center point of the image collector. Two included angles, the first vertical point refers to the vertical projection point of the center point of the image collector relative to the ground plane on which the animal to be tested stands;

a shooting height obtaining unit 1234, configured to obtain the shooting height of the image collector when shooting the target image;

In a possible implementation manner, if there is a first height difference between the operating body and the respective surfaces of the animal to be tested, as shown in FIG. 11 above, the shooting height obtaining unit 1234 may include:

a parameter calling unit, used for calling the height and the first height difference of the operating body that shoots the target image;

A shooting height obtaining unit, configured to obtain the shooting height of the image collector relative to the ground plane on which the animal to be tested stands according to the height of the operating body and the first height difference.

The horizontal shooting distance obtaining unit 1235 is used for obtaining the horizontal shooting distance between the center point of the image collector and the animal to be measured by using the above shooting height and through tangent function operation.

In combination with the description of the above embodiment, as shown in FIG. 14 , the above-mentioned body size parameter obtaining unit may include:

The third included angle obtaining unit 1241 is configured to obtain a third angle formed by the corresponding two key feature points relative to the center point of the image collector according to the respective position information of any two key feature points of the animal to be tested. included angle;

The feature point distance obtaining unit 1242 is used to obtain the feature point distance between the corresponding two key feature points by using the third included angle and the horizontal shooting distance;

The body measure parameter obtaining unit 1243 is used to obtain the corresponding body to be measured of the animal to be measured by using the obtained feature point distances between at least two second key feature points corresponding to each type of body measure to be measured. Ruler parameter.

It should be noted that various modules, units, etc. in the above-mentioned device embodiments can be stored in the memory as program modules, and the processor executes the above-mentioned program modules stored in the memory to realize corresponding functions. For the functions implemented by the program modules and their combinations, and the technical effects achieved, reference may be made to the descriptions in the corresponding parts of the foregoing method embodiments, and details are not repeated in this embodiment.

The present application also provides a storage medium on which a computer program can be stored, and the computer program can be called and loaded by a processor to implement each step of the method for acquiring external features of an animal described in the above embodiments.

The present application also provides an animal external feature acquisition system, which may include a mobile terminal and a server.

In some embodiments, combined with the above analysis, the mobile terminal may have an image collector and certain image processing capabilities, and it has the structure described in the above computer equipment, and can execute the various steps of the method for acquiring external features of animals proposed in this application. , the implementation process may refer to the description of the corresponding part of the above-mentioned embodiment, which is not repeated.

In practical applications, in order to facilitate the subsequent query of the state of each animal to be tested and the external feature information at different stages, the mobile terminal can also associate the obtained external feature value of the animal to be tested with its corresponding acquisition time, and upload it to the server for storage. According to needs, the captured target image of the animal to be tested and other shooting parameters can also be uploaded to the server for storage for subsequent calling and viewing.

In still other embodiments, the mobile terminal is only used as an image acquisition device and does not perform image processing. Then, in combination with the descriptions in the corresponding parts of the above embodiments, the user can use the mobile terminal to collect the target image of the animal to be tested, and upload the relevant shooting parameters to the Server: The server executes each step of the animal external feature acquisition method proposed in the present application, obtains the required external feature value, and then feeds it back to the mobile terminal for output, so as to reduce the processing burden of the mobile terminal and improve processing efficiency.

Moreover, in this application scenario, the user can use the mobile terminal to shoot images of multiple animals to be tested and upload them to the server, which can simultaneously acquire the external features of multiple animals to be tested. For the specific implementation process, refer to the corresponding embodiments in the above-mentioned embodiments. Part of the description is not repeated in this embodiment.

Finally, it should be noted that the various embodiments in this specification are described in a progressive or juxtaposed manner, and each embodiment focuses on the differences from other embodiments, and the same and similar parts between the various embodiments can be referred to each other. . For the apparatuses, computer equipment, and systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple, and for related parts, please refer to the descriptions of the methods.

Professionals may further realize that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two, in order to clearly illustrate the possibilities of hardware and software. Interchangeability, the above description has generally described the components and steps of each example in terms of functionality. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Skilled artisans may implement the described functionality using different methods for each particular application, but such implementations should not be considered beyond the scope of this application.

The steps of a method or algorithm described in conjunction with the embodiments disclosed herein may be directly implemented in hardware, a software module executed by a processor, or a combination of the two. A software module can be placed in random access memory (RAM), internal memory, read only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other in the technical field. in any other known form of storage medium.

The above description of the disclosed embodiments enables any person skilled in the art to make or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the core idea or scope of the present application. Therefore, this application is not intended to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

A method for acquiring external features of animals, characterized in that the method comprises:

Obtain the target image of the animal to be tested, and the shooting attitude information when the image collector shoots the target image;

Perform feature extraction on the target image to obtain a plurality of key feature point information of the animal to be tested contained in the target image, where the key feature point information includes position information of the corresponding key feature point in the target image;

According to the shooting attitude information and the position information of the first key feature point, the horizontal shooting distance between the center point of the image collector and the animal to be measured is obtained, and the first key feature point refers to the to-be-measured animal. Among the multiple key feature points of the animal, the key feature points whose distance from the ground is less than the first threshold;

According to the horizontal shooting distance and the position information of the second key feature point corresponding to the external feature to be tested, the external feature value to be tested corresponding to the animal to be tested is obtained.
The method according to claim 1, wherein the external feature to be measured comprises at least one body size to be measured and/or at least one attribute to be measured of the animal to be measured, and the shooting according to the horizontal distance, and the position information of the second key feature point corresponding to the external feature to be tested, to obtain the external feature value to be tested corresponding to the animal to be tested, including

According to the horizontal shooting distance and the position information of the second key feature point corresponding to the body measure to be measured, obtain the body measure parameter to be measured corresponding to the animal to be measured;

According to the obtained at least one body size parameter to be measured, the corresponding attribute value to be measured of the animal to be measured is obtained.
The method according to claim 2, wherein the shooting attitude information includes shooting pitch angle and shooting roll angle, and the image capture is obtained according to the shooting attitude information and the position information of the first key feature point. The horizontal shooting distance between the center point of the detector and the animal to be tested, including:

Using the position information of the first key feature point, determine the first distance between the first key feature point and the horizontal center line of the image collector;

Using the first distance and the shooting roll angle to obtain a first included angle of the first key feature point relative to the horizontal centerline;

Using the first included angle and the shooting pitch angle, the second included angle formed by the first key feature point and the first vertical point relative to the center point of the image collector is obtained, and the first vertical point Refers to the vertical projection point of the center point of the image collector relative to the ground plane on which the animal to be tested stands;

obtaining the shooting height of the image collector when shooting the target image;

Using the shooting height, the horizontal shooting distance between the center point of the image collector and the animal to be tested is obtained through the operation of the tangent function.
The method according to claim 2 or 3, characterized in that, according to the horizontal shooting distance and the position information of the second key feature point corresponding to the body size to be measured, the corresponding to-be-measured animal of the to-be-measured animal is obtained Body size parameters, including:

According to the respective position information of any two key feature points of the animal to be tested, the third angle formed by the corresponding two key feature points relative to the center point of the image collector is obtained;

Use the third included angle and the horizontal shooting distance to obtain the feature point distance between the corresponding two key feature points;

Using the obtained feature point distances between at least two second key feature points corresponding to each type of body size to be measured, the parameters of the body size to be measured corresponding to the animal to be measured are obtained.
The method according to claim 3, wherein, if there is a first height difference between the operating body and the ground where the animal to be tested is located, the acquisition of the target image when the target image is captured Shooting heights, including:

calling the height and the first height difference of the operating body that captures the target image;

According to the height of the operating body and the first height difference, the shooting height of the image collector relative to the ground plane where the animal to be tested stands is obtained.
The method according to claim 1, wherein the feature extraction is performed on the target image to obtain a plurality of key feature point information of the animal to be tested contained in the target image, including:

Inputting the target image into an image feature extraction model to obtain a plurality of key feature points of the animal to be tested contained in the target image;

outputting multiple key feature points of the animal to be tested contained in the target image;

In response to the position adjustment operation for the output at least one key feature point, multiple key feature point information of the animal to be tested is obtained.
The method according to claim 6, wherein the method further comprises:

Perform integrity verification on the obtained multiple key feature points of the animal to be tested;

If the verification fails, first prompt information is output, and the first prompt information is used to remind the operator to re-shoot the target image of the animal to be tested.
A device for acquiring external features of animals, characterized in that the device comprises:

an information acquisition module, used for acquiring a target image of the animal to be tested, and the shooting attitude information when the image collector shoots the target image;

A feature extraction module is used to perform feature extraction on the target image to obtain a plurality of key feature point information of the animal to be tested contained in the target image, and the key feature point information includes the corresponding key feature points in the target image. location information in the image;

The horizontal shooting distance obtaining module is used to obtain the horizontal shooting distance between the center point of the image collector and the animal to be measured according to the shooting posture information and the position information of the first key feature point. The first key Feature points refer to the key feature points whose distance from the ground is less than the first threshold among the multiple key feature points of the animal to be tested;

The external feature value obtaining module to be tested is used for obtaining the external feature value to be tested corresponding to the animal to be tested according to the horizontal shooting distance and the position information of the second key feature point corresponding to the body size to be tested.
A computer device, characterized in that the computer device comprises:

Communication Interface;

a memory for storing a program for realizing the method for obtaining an attribute value of an animal according to any one of claims 1-7;

The processor is configured to load and execute the program to implement each step of the method for obtaining an attribute value of an animal according to any one of claims 1-7.
The computer device according to claim 9, wherein the computer device is a mobile terminal or a server;

When the computer device is the mobile terminal, the computer device further includes:

An image collector for taking images of the animal to be tested;

a display screen for outputting the image of the animal to be tested captured by the image collector;

A variety of attitude sensors for sensing the shooting attitude information when the image collector shoots the target image of the animal to be tested;

When the computer device is a server, the communication interface is used to receive the target image of the animal to be tested shot and sent by the image collector, and the shooting posture information when the image collector shoots the target image; The corresponding external characteristic value of the tested animal is fed back to the preset mobile terminal for output.