CN112288793A - Livestock individual backfat detection method and device, electronic equipment and storage medium - Google Patents

Livestock individual backfat detection method and device, electronic equipment and storage medium Download PDF

Info

Publication number
CN112288793A
CN112288793A CN202011232362.5A CN202011232362A CN112288793A CN 112288793 A CN112288793 A CN 112288793A CN 202011232362 A CN202011232362 A CN 202011232362A CN 112288793 A CN112288793 A CN 112288793A
Authority
CN
China
Prior art keywords
individual
livestock
backfat
detected
body size
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202011232362.5A
Other languages
Chinese (zh)
Other versions
CN112288793B (en
Inventor
闫润强
杨梓钰
李旭强
邓柯珀
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Luoyang Voice Cloud Innovation Institute
Original Assignee
Luoyang Voice Cloud Innovation Institute
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Luoyang Voice Cloud Innovation Institute filed Critical Luoyang Voice Cloud Innovation Institute
Priority to CN202011232362.5A priority Critical patent/CN112288793B/en
Publication of CN112288793A publication Critical patent/CN112288793A/en
Application granted granted Critical
Publication of CN112288793B publication Critical patent/CN112288793B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
    • G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/50—Depth or shape recovery
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10004—Still image; Photographic image
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10028—Range image; Depth image; 3D point clouds
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
    • Y02A40/00—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production
    • Y02A40/70—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production in livestock or poultry

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Human Computer Interaction (AREA)
  • Multimedia (AREA)
  • Geometry (AREA)
  • Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)

Abstract

The embodiment of the invention provides a method and a device for detecting backfat of an individual livestock, electronic equipment and a storage medium, wherein the body ruler data of the individual livestock to be detected can be automatically determined through three-dimensional point cloud data; through the body ruler data, the backfat information of the individual livestock to be detected can be quickly and accurately determined; the identity information of the individual livestock to be detected is introduced and combined with the body size data of the individual livestock to be detected, so that the accuracy of the backfat information of the individual livestock to be detected can be further improved. The whole detection process can be automated, manual participation is not needed, the labor consumption brought by the complex backfat detection process can be reduced, the detection efficiency can be improved, and the phenomenon that the stress reaction of livestock and poultry is caused or the probability of the diseases of the livestock and the poultry is increased due to the contact of the livestock and the poultry can be avoided. In addition, the limitation conditions such as the posture of the livestock individual and the ambient light do not need to be considered in the detection process, so that the detection method can be suitable for detecting the backfat information of the livestock individual in various different scenes.

Description

Livestock individual backfat detection method and device, electronic equipment and storage medium
Technical Field
The invention relates to the technical field of farming and animal husbandry breeding, in particular to a livestock individual backfat detection method, a device, electronic equipment and a storage medium.
Background
The measurement of the backfat of the livestock individual plays an important role in the whole breeding process of the livestock, particularly the breeding process of female livestock, and the backfat thickness of the female livestock individual at each stage reflects the health condition and the nutrition level of the female livestock individual.
In the prior art, the obtaining mode of the individual backfat of the livestock mainly comprises the steps of manually realizing through a feeder and automatically identifying the individual two-dimensional image of the livestock. However, the artificial implementation mode of the breeder requires manual participation, which not only has low efficiency and is easy to cause misjudgment, but also requires close contact of people and livestock, which can cause stress reaction of livestock and poultry on one hand and increase the disease probability of people and livestock on the other hand; and still need people and animals to contact closely; although the close contact between people and livestock can be avoided by the implementation mode of automatically identifying the two-dimensional image of the livestock individual, the deviation is easy to generate when the two-dimensional image of the livestock individual is identified, and the real backfat of the livestock individual cannot be truly reflected; moreover, the acquisition of the two-dimensional images of the livestock individuals has high requirements on the postures of the livestock individuals, ambient light and the like, and cannot be suitable for the detection of the backfat of the livestock individuals in various different scenes.
Disclosure of Invention
The embodiment of the invention provides a method and a device for detecting the backfat of an individual livestock, electronic equipment and a storage medium, which are used for solving the defects in the prior art.
The embodiment of the invention provides a method for detecting the backfat of an individual livestock, which comprises the following steps:
determining three-dimensional point cloud data of an individual of a livestock to be detected, and determining body size data of the individual of the livestock to be detected based on the three-dimensional point cloud data;
and determining backfat information of the individual of the livestock to be detected based on the body size data of the individual of the livestock to be detected or based on the body size data of the individual of the livestock to be detected and the identity information of the individual of the livestock to be detected.
According to the livestock individual backfat detection method provided by the embodiment of the invention, the determination of the body size data of the livestock individual to be detected based on the three-dimensional point cloud data specifically comprises the following steps:
projecting the three-dimensional point cloud data to a horizontal plane to obtain the back contour of the individual livestock to be detected, or projecting the three-dimensional point cloud data to the horizontal plane and a vertical plane perpendicular to the trunk of the individual livestock to be detected respectively to obtain the back contour and the hip contour of the individual livestock to be detected;
determining hip width in the body ruler data based on the back contour and/or the hip contour, and determining shoulder width, waist width and bust in the body ruler data based on the back contour.
According to the livestock individual backfat detection method provided by the embodiment of the invention, the determining of the shoulder width, the waist width and the chest circumference in the body size data based on the back contour specifically comprises the following steps:
determining two shoulder width key points and two waist width key points in the back contour, determining the shoulder width based on the two shoulder width key points, and determining the waist width based on the two waist width key points;
determining the chest contour of the individual livestock to be detected based on the three-dimensional point cloud data and a vertical plane containing the two shoulder width key points;
and determining the chest circumference of the individual livestock to be detected based on the chest contour.
According to the method for detecting the backfat of the livestock individual, disclosed by the embodiment of the invention, the determining of the backfat information of the livestock individual to be detected based on the body size data of the livestock individual to be detected or based on the body size data of the livestock individual to be detected and the identity information of the livestock individual to be detected specifically comprises the following steps:
inputting the body size data or the body size data and the identity information into a backfat detection model to obtain the backfat information output by the backfat detection model;
the backfat detection model is obtained based on body size data of the sample livestock individual carrying the backfat information label or based on body size data training of the sample livestock individual carrying the backfat information label and the identity information label.
According to the livestock individual backfat detection method provided by the embodiment of the invention, the backfat detection model comprises a plurality of backfat detection layers corresponding to different identity information;
correspondingly, will the body size data input is to backfat detection model, obtain by backfat detection model output backfat information specifically includes:
and inputting the body size data into the backfat detection layer corresponding to the identity information in the backfat detection model to obtain the backfat information output by the backfat detection layer corresponding to the identity information.
According to the method for detecting the backfat of the individual livestock, in an embodiment of the invention, the step of inputting the body size data to the backfat detection layer corresponding to the identity information in the backfat detection model to obtain the backfat information output by the backfat detection layer corresponding to the identity information further includes:
and inputting the body size data into an identity classification layer in the backfat detection model to obtain the identity information output by the identity classification layer.
According to the method for detecting the backfat of the livestock individual, disclosed by the embodiment of the invention, the determining of the backfat information of the livestock individual to be detected based on the body size data of the livestock individual to be detected or based on the body size data of the livestock individual to be detected and the identity information of the livestock individual to be detected specifically comprises the following steps:
and if the body size data is in a preset general body size data interval, determining the backfat information based on the body size data or the body size data and the identity information.
The embodiment of the invention also provides a livestock individual backfat detection device, which comprises: the back fat measurement system comprises a body size data determining module and a back fat information determining module. Wherein the content of the first and second substances,
the body size data determining module is used for determining three-dimensional point cloud data of the individual livestock to be detected and determining body size data of the individual livestock to be detected based on the three-dimensional point cloud data;
the backfat information determining module is used for determining backfat information of the individual to be detected based on the body size data of the individual to be detected, or based on the body size data of the individual to be detected and the identity information of the individual to be detected.
The embodiment of the invention also provides electronic equipment, which comprises a memory, a processor and a computer program which is stored on the memory and can run on the processor, wherein when the processor executes the program, the steps of the livestock individual backfat detection method are realized.
Embodiments of the present invention further provide a non-transitory computer readable storage medium, on which a computer program is stored, and the computer program, when executed by a processor, implements the steps of the method for detecting backfat of an individual livestock as described in any one of the above.
According to the livestock individual backfat detection method, the device, the electronic equipment and the storage medium provided by the embodiment of the invention, the body size data of the livestock individual to be detected can be automatically determined through the three-dimensional point cloud data; the backfat information of the individual livestock to be detected can be quickly and accurately determined through the body ruler data of the individual livestock to be detected; the identity information of the individual livestock to be detected is introduced and combined with the body size data of the individual livestock to be detected, so that the accuracy of the backfat information of the individual livestock to be detected can be further improved. The whole detection process can be automated, manual participation is not needed, the labor consumption brought by the complex backfat detection process can be reduced, the detection efficiency can be improved, and the phenomenon that the stress reaction of livestock and poultry is caused or the probability of the diseases of the livestock and the poultry is increased due to the contact of the livestock and the poultry can be avoided. In addition, the limitation conditions such as the posture of the livestock individual and the ambient light do not need to be considered in the detection process, so that the detection method can be suitable for detecting the backfat information of the livestock individual in various different scenes.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and those skilled in the art can also obtain other drawings according to the drawings without creative efforts.
FIG. 1 is a schematic flow chart of a method for detecting the backfat of an individual livestock according to an embodiment of the invention;
FIG. 2 is a schematic structural diagram of an apparatus for detecting backfat of an individual livestock according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
At present, the backfat information detection of the domestic animals such as pigs, cows, sheep, horses, donkeys and the like is crucial to the whole feeding process of the domestic animals, and by taking the breeding process of sows as an example, the backfat thickness of the individual sows at each stage reflects the nutrition level and the health condition of the individual sows. In intensive, automatic and intelligent modern breeding scenes, the backfat thickness of the individual sows is automatically and intelligently measured without contact, the nutrition level and the health condition of the individual sows can be known, the feeding amount of the feeding level of the individual sows is further adjusted, and the method has great significance for reducing the feed-meat ratio and improving the economic benefit of fattening pigs. Most of the existing sow body condition evaluation methods are that a grader visually observes the body condition of a sow according to experience and scores the sow, but the body condition score is not objective, has large errors and cannot accurately reflect the backfat level of the overall fat degree of a swinery. At present, the livestock individual backfat can be obtained mainly in three ways: 1) the backfat instrument is attached to the surface of the skin of the livestock individual by holding the backfat instrument by a feeder to measure the backfat; 2) estimating the hip of the livestock individual through human eye observation, and further roughly estimating the backfat of the livestock individual; 3) and obtaining the key point information of the individual livestock by using the two-dimensional image, and identifying the individual backfat by using a body type template obtained by similarity transformation so as to obtain the backfat value.
However, 1) through the scheme that a feeder holds a backfat instrument by hand, although the measuring precision is high, the efficiency is low because of the need of manual participation and the need of close contact of people and livestock, on one hand, the stress response of livestock can be caused, and on the other hand, the disease probability of people and livestock can be increased; 2) the proposal for estimating the buttocks of the livestock individual through human eye observation has higher requirements on the experience of workers, is easy to cause misjudgment, and still needs close contact of people and livestock; 3) although the mode of using the two-dimensional image can avoid close contact of people and livestock, the determination capacity of the key point information of the individual livestock is easy to generate deviation, and the real backfat of the individual livestock cannot be truly reflected; moreover, the acquisition of the two-dimensional image has high requirements on the posture of the livestock individual, ambient light and the like, and cannot be suitable for the measurement of the backfat of the livestock individual in various different scenes. Based on the above, the embodiment of the invention provides a method for detecting the backfat of an individual livestock, so as to solve the technical problems in the prior art.
Fig. 1 is a schematic flow chart of a method for detecting backfat of an individual livestock in an embodiment of the invention, as shown in fig. 1, the method includes:
s1, determining three-dimensional point cloud data of the individual livestock to be detected, and determining body size data of the individual livestock to be detected based on the three-dimensional point cloud data;
and S2, determining backfat information of the individual to be detected based on the body size data of the individual to be detected, or based on the body size data of the individual to be detected and the identity information of the individual to be detected.
Specifically, the method for detecting the backfat of the individual livestock provided in the embodiment of the invention aims to realize automatic detection of the backfat information of the individual livestock to be detected by acquiring point cloud data of the individual livestock to be detected. Wherein, the backfat information can be the backfat thickness.
Step S1 is performed first. The livestock to be detected refers to a target object needing backfat information detection, and the variety of the livestock to be detected can specifically comprise pigs, cows, sheep, horses, donkeys and the like. The individual of the livestock to be tested refers to an independent individual of the livestock to be tested, such as a pig, a cow, a horse, etc. In the embodiment of the present invention, the sex of the individual livestock to be tested is not particularly limited.
The three-dimensional point cloud data of the individual livestock to be detected is point cloud data representing a three-dimensional integral model of the individual livestock to be detected, in the embodiment of the invention, the point cloud data of the individual livestock to be detected can be obtained from different angles through a point cloud capturing device such as a depth camera, and the three-dimensional point cloud data of the individual livestock to be detected is obtained through further preprocessing and three-dimensional reconstruction of a registration algorithm. Compared with the traditional method for acquiring the two-dimensional image of the individual livestock to be detected, the method for acquiring the point cloud data of the individual livestock to be detected has lower requirements on the posture of the individual livestock, the ambient light and the like.
The body ruler data of the individual livestock to be detected can be determined by carrying out post-processing on the three-dimensional point cloud data in modes of projection and the like and combining the characteristics of each part of the individual livestock to be detected. The body size data can comprise data of shoulder width, waist width, hip width, waist circumference, chest depth, body length, body height and the like of the individual livestock to be detected, and can also comprise proportion data, such as back width-length ratio, chest depth-leg height ratio and the like, wherein the back width-length ratio is the ratio of the hip width to the body length, and the chest depth-leg height ratio is the ratio of the chest depth to the body height.
Then, step S2 is executed. In the embodiment of the invention, the backfat information of the individual to be detected of the livestock can be determined through the body scale data of the individual to be detected of the livestock, and the backfat information of the individual to be detected of the livestock can be determined by combining the identity information of the individual to be detected of the livestock on the basis of the body scale data of the individual to be detected of the livestock. When determining the backfat information of the individual of the livestock to be detected, the backfat detection method can be realized through a pre-trained backfat detection model, and can also be determined through a pre-determined corresponding relationship between the body size data and the backfat information, or through a pre-determined corresponding relationship between the body size data, the identity information and the backfat information, which is not particularly limited in the embodiment of the invention.
The identity information can comprise the variety, growth stage and other information of the individual livestock to be detected. Each individual of the domestic animals to be tested is raised in the corresponding column, each individual of the domestic animals to be tested has a unique ID which is used as an identification of the individual of the domestic animals to be tested, and a corresponding relation exists between the ID and the identity information. When the identity information of the individual livestock to be detected is obtained, the identity information of the individual livestock to be detected and the corresponding ID can be stored in the tag, and the identity information of the individual livestock to be detected can be obtained through equipment which can be in communication connection with the tag. The label can be worn on the body of the individual livestock to be detected or fixed in a column where the individual livestock to be detected is located, and the identity information stored in the label can be automatically updated along with the state of the individual livestock to be detected. The tag may specifically be a Radio Frequency Identification (RFID) tag, and correspondingly, the device that can be in communication connection with the tag may be an RFID reader.
According to the livestock individual backfat detection method provided by the embodiment of the invention, the body size data of the livestock individual to be detected can be automatically determined through the three-dimensional point cloud data; the backfat information of the individual livestock to be detected can be quickly and accurately determined through the body ruler data of the individual livestock to be detected; in addition, the identity information of the individual livestock to be detected is introduced and combined with the body size data of the individual livestock to be detected, so that the accuracy of the backfat information of the individual livestock to be detected can be further improved. The whole detection process can be automated, manual participation is not needed, the labor consumption brought by the complex backfat detection process can be reduced, the detection efficiency can be improved, and the phenomenon that the stress reaction of livestock and poultry is caused or the probability of the diseases of the livestock and the poultry is increased due to the contact of the livestock and the poultry can be avoided. In addition, the limitation conditions such as the posture of the livestock individual and the ambient light do not need to be considered in the detection process, so that the detection method can be suitable for detecting the backfat information of the livestock individual in various different scenes.
On the basis of the above embodiment, in the method for detecting backfat of an individual animal provided in the embodiment of the present invention, when point cloud data of an individual animal to be detected is obtained from different angles by a point cloud capturing device, a Region of Interest (ROI) can be established by using a straight-through filter, and the number of sampling points is reduced by using filtering algorithms such as voxel filter. And the point cloud data is acquired, a color image of the individual animal to be detected can be acquired, the point cloud data is combined with the color image to obtain four-dimensional data which relates to the individual animal to be detected and comprises position information and RGB color information in the directions of x, y and z under a three-dimensional coordinate system, and the determination of each part of the individual animal to be detected is convenient to realize.
The point cloud capturing device can be a movable and telescopic artificial image acquisition device carrying a depth camera, and can obtain point cloud data of the individual livestock to be detected in real time. Specifically, a patrol track can be arranged around a column where the individual stock to be detected is located, and the point cloud capturing device can move on the patrol track. In the motion process of the point cloud capturing device, the point cloud data of different angles can be collected for the same individual livestock to be detected for many times through the depth camera. It should be noted that point cloud data of all angles of the individual livestock to be tested need to be acquired synchronously, and meanwhile, the point cloud data of all angles need to be ensured to cover the whole range of the individual livestock to be tested, and partial point cloud data are overlapped between the point cloud data of two adjacent angles. The depth camera can specifically adopt kinect, axon and the like.
After point cloud data of the individual livestock to be detected are obtained from different angles, the point cloud data of the individual livestock to be detected under different angles can be preprocessed, specifically, a point cloud transformation matrix is determined firstly, namely, a rotation and translation matrix between the point cloud data of the individual livestock to be detected under different angles is solved, and the point cloud transformation matrix can be a rigid transformation matrix or an Euclidean transformation matrix. Then, point cloud data under different angles are converted to the same coordinate system through a registration algorithm; and finally, performing three-dimensional reconstruction by using an irregular triangulation algorithm to obtain three-dimensional point cloud data of the individual livestock to be detected, and obtaining a closed three-dimensional integral model of the individual livestock to be detected.
In order to ensure that the determined three-dimensional point cloud data has the same orientation in the livestock individuals to be detected represented in the space, the embodiment of the invention firstly carries out clustering operation on the three-dimensional point cloud data, determines a clustering center as the mass center of the livestock individuals to be detected, and determines the main shaft of the three-dimensional point cloud data by using a Principal Component Analysis (PCA) method by using the mass center as a coordinate origin, wherein the main shaft comprises axes in the directions of x, y and z. The x direction is the direction which passes through the origin of coordinates and faces the head of the individual livestock to be detected, the z direction is the direction which passes through the origin of coordinates and is vertically upward, and the z direction is the direction which is vertical to the x direction and the y direction. The xoz plane is a vertical plane parallel to the trunk of the individual livestock to be tested, the xoy plane is a horizontal plane, and the yoz plane is a vertical plane perpendicular to the trunk of the individual livestock to be tested. By the PCA method, the coordinate values of all the three-dimensional point cloud data can be linearly transformed, so that the correlation of the coordinate components of all the three-dimensional point cloud data in 3 main axis directions is minimized, and the three-dimensional point cloud data is obtained.
On the basis of the foregoing embodiment, the method for detecting backfat of an individual animal provided in the embodiment of the present invention, which determines the body size data of the individual animal to be detected based on the three-dimensional point cloud data, specifically includes:
projecting the three-dimensional point cloud data to a horizontal plane to obtain the back contour of the individual livestock to be detected, or projecting the three-dimensional point cloud data to the horizontal plane and a vertical plane perpendicular to the trunk of the individual livestock to be detected respectively to obtain the back contour and the hip contour of the individual livestock to be detected;
determining hip width in the body ruler data based on the back contour and/or the hip contour, and determining shoulder width, waist width and bust in the body ruler data based on the back contour.
Specifically, when the body size data of the individual livestock to be detected is determined based on the three-dimensional point cloud data, the three-dimensional point cloud data can be firstly projected downwards to a horizontal plane, namely to the xoy plane, and a two-dimensional back projection image can be obtained. After the two-dimensional back projection image is obtained, the plane shape can be reconstructed through the point set on the back projection image, so that the outline of the two-dimensional back projection shape is extracted, and the back outline of the individual livestock to be detected is obtained. And on the basis of determining the back contour, the three-dimensional point cloud data is projected to a vertical plane perpendicular to the trunk of the individual to be tested, namely to a yoz plane, towards the head direction of the individual to be tested, so that a two-dimensional hip projection image can be obtained. After the two-dimensional hip projection image is obtained, the plane shape can be reconstructed through the point set on the hip projection image, so that the contour of the two-dimensional hip projection shape is extracted, and the hip contour of the individual to be tested is obtained. Here, reconstructing the planar shape may be achieved by a pouch algorithm.
In the embodiment of the invention, the hip width in the body scale data of the livestock individual to be detected can be determined according to the back contour, the hip width in the body scale data of the livestock individual to be detected can be determined according to the hip contour, and the hip width in the body scale data of the livestock individual to be detected can be determined by combining the back contour and the hip contour.
For example, the shoulder and hip of the individual stock to be tested can be distinguished by the back contour, and then the shoulder width and hip width can be further determined. Specifically, the back contour can be divided into four quadrants according to an x axis and a y axis, and the maximum distance from contour points at two ends of the back contour to the xoz plane in each quadrant is respectively calculated, so that four body ruler key points respectively located in the four quadrants are obtained. Two of the four individual ruler key points are shoulder width key points, two of the four individual ruler key points are hip width key points, and the two shoulder width key points and the two hip width key points are respectively positioned at two ends of the back contour. And then, the curvature radius and the characteristic histogram of the point cloud data at the two ends of the back contour can be further determined, whether the curvature radius at the two ends of the back contour is smaller than a preset curvature radius or not is respectively compared, the distance between the characteristic histogram at the two ends of the back contour and the characteristic histogram of the template hip point cloud data in a characteristic space is respectively calculated, one end close to the characteristic histogram of the template hip point cloud data or one end with the curvature radius smaller than the preset curvature radius corresponds to the hip, the corresponding two body size key points are hip width key points, and the distance between the two hip width key points is the hip width.
The maximum and minimum y of the y-axis coordinate can also be found by measuring the width between the most outward protruding contour points in the hip contourmaxAnd yminThe hip width is dwidth＝ymax-ymin. The hip widths determined based on the back contour and the hip contour, respectively, can also be compared on the basis, if the difference between the two is within a preset range, both are considered to be correct, and either one is taken as the hip width or the average value of the two is taken as the hip width.
The shoulder width, waist width and chest circumference in the body ruler data of the livestock individual to be detected can be determined through the back contour of the livestock individual to be detected, for example, after the hip and hip width are determined through the back contour, the shoulder and shoulder width can be determined; the distance between two contour points which are near the y axis in the back contour and are farthest away from the coordinate origin can be used as the waist width; because the shoulder width can be equivalent to the chest width, under the condition of determining the chest width, the circumference of the tangent plane outline of the three-dimensional point cloud data formed by the chest width in the vertical plane is the bust.
In addition, after the back contour and/or the hip contour are determined, a first central point between the two shoulder width key points and a second central point between the two hip width key points can be obtained, the first central point and the second central point are two body length key points, and the distance between the two body length key points is the body length. The areas of the back contour and the hip contour can be calculated respectively, and the back area and the hip area in the body ruler data can be obtained respectively.
According to the embodiment of the invention, the back contour of the individual of the livestock to be detected is adopted or combined with the hip contour, the hip width, the shoulder width, the waist width and the chest circumference in the body ruler data are determined, and the accuracy of the body ruler data can be ensured by means of the three-dimensional point cloud data, so that the accuracy of the backfat detection result is ensured, and higher requirements on environment or light are not required.
On the basis of the above embodiments, in the method for detecting the backfat of the individual livestock provided in the embodiments of the present invention, the curvature radius of each end of the back contour can be specifically determined by the following method: and at the two body ruler key points of the end, reserving points with z coordinates larger than the two body ruler key points, omitting points with z coordinates smaller than the two body ruler key points, carrying out polynomial fitting by using a curve fitting algorithm, and calculating according to a curve obtained by fitting to obtain the curvature radius.
On the basis of the foregoing embodiment, the method for detecting backfat of an individual livestock provided in an embodiment of the present invention, which determines the shoulder width, waist width and chest circumference in the body size data based on the back contour, specifically includes:
determining two shoulder width key points and two waist width key points in the back contour, determining the shoulder width based on the two shoulder width key points, and determining the waist width based on the two waist width key points;
determining the chest contour of the individual livestock to be detected based on the three-dimensional point cloud data and a vertical plane containing the two shoulder width key points;
and determining the chest circumference of the individual livestock to be detected based on the chest contour.
Specifically, after the two hip width key points are determined through the back contour, the other end of the back contour, which is opposite to the two hip width key points, is the shoulder, the two corresponding body ruler key points are shoulder width key points, and the distance between the two shoulder width key points is the shoulder width.
The back contour is divided into two parts according to positive and negative values of the y axis, namely the x axis is taken as the axis to divide the back contour into two parts, the distances from contour points close to the central point in the two parts to the yoz plane are respectively calculated, and two contour points with the minimum distance in the two parts are found and are respectively the waist width key points. The distance between two waist width key points is the waist width.
Under the condition that the chest width is determined to be the shoulder width, a line segment between two shoulder width key points can be equivalent to a chest width line segment, a slicing method is used for processing the chest width line segment to form a plane perpendicular to a horizontal plane, namely a vertical plane containing the two shoulder width key points is obtained, and the outline formed by the vertical plane and the three-dimensional point cloud data is the chest outline of the individual livestock to be detected. The perimeter of the chest outline is the chest circumference of the individual livestock to be detected.
According to the embodiment of the invention, the back contour is adopted, the shoulder width and the waist width are determined by determining the shoulder width key point and the waist width key point, the chest circumference is determined by combining a slicing method, and the accuracy of the body ruler data can be ensured by means of three-dimensional point cloud data, so that the accuracy of a backfat detection result is ensured, and higher requirements on environment or light are not required.
On the basis of the embodiment, the livestock individual backfat detection method provided by the embodiment of the invention can also determine the height, waist, chest depth and the like in the body size data of the livestock individual to be detected through the three-dimensional point cloud data.
Specifically, the height in the body size data refers to the vertical distance from the highest hip joint hip of the individual livestock to be tested to the ground. In the embodiment of the invention, the point coordinate with the maximum coordinate value on the z axis can be firstly determined as the hip highest point zmaxThe coordinate of the point with the minimum coordinate value on the z-axis is a point z on the groundminThen the height calculation formula is Dtall＝zmax-zmin。
Further, the posture in the body size data of the individual livestock to be detected can be determined according to the height. When the height D istallWhen the height is larger than the height threshold value, the individual of the livestock to be detected can be judged to be in a standing posture, and the height D is obtainedtallWhen the height is smaller than the height threshold value, the individual of the livestock to be detected can be judged to be in the lying posture.
The waist circumference in the body size data can be specifically determined by the following method: determining a waist width line segment between two waist width key points in the back contour, then using a slicing method to pass through the waist width line segment to form a vertical plane perpendicular to the horizontal plane, thereby obtaining the waist contour, and calculating the circumference of the waist contour, namely the waist circumference.
The chest depth in the body size data can be specifically determined by the following method: and taking the maximum value and the minimum value of the z coordinate in the chest contour, wherein the difference value is the chest depth.
In the embodiment of the invention, the body size data is supplemented, so that the body size data is more complete, the factors considered when backfat information is detected through the body size data are more comprehensive, and the detection result is more accurate.
On the basis of the above embodiment, because the back fat information finally detected is affected by the head twisting, body bending and other actions of the to-be-detected animal individual, the animal individual back fat detection method provided in the embodiment of the present invention may further perform symmetry detection on the to-be-detected animal individual on the basis of obtaining the shoulder contour, so as to determine whether the head twisting, body bending and other actions occur in the point cloud data acquisition of the to-be-detected animal individual. The symmetry of the individual of the livestock to be detected can also be used as a part of the body size data and used as a consideration factor for subsequent detection of backfat information, so that the symmetry of the individual of the livestock to be detected is considered when the backfat information is detected, and the accuracy of a detection result is ensured. Symmetry detection means that if one point exists on one side of the symmetry plane, one point should exist on the other side of the symmetry plane, the distances between the two points and the symmetry plane are equal, the distance difference is 0, and the connecting line of the two points is perpendicular to the symmetry plane. In the embodiment of the invention, the symmetry plane is an xoz plane, the back contour is divided into two parts according to a xoz plane, the sum of distances from all contour points of the two parts to a xoz plane is respectively calculated, and if the difference between the sum of the distances of the two parts is smaller than a distance threshold value, the back contour is symmetrical.
In the embodiment of the invention, the symmetry of the individual of the livestock to be detected is detected, the obtained symmetry is used as a part of the body size data, and the backfat information can be detected under the condition of considering the posture information of the individual of the livestock to be detected without acquiring the posture information of the individual of the livestock to be detected, so that the workload of a data preparation stage is reduced, and the accuracy of the detection result is ensured.
On the basis of the foregoing embodiment, the method for detecting backfat of an individual of a livestock provided in the embodiment of the present invention, which determines backfat information of the individual of the livestock to be detected based on the body size data of the individual of the livestock to be detected, or based on the body size data of the individual of the livestock to be detected and the identity information of the individual of the livestock to be detected, specifically includes:
inputting the body size data or the body size data and the identity information into a backfat detection model to obtain the backfat information output by the backfat detection model;
the backfat detection model is obtained based on body size data of the sample livestock individual carrying the backfat information label or based on body size data training of the sample livestock individual carrying the backfat information label and the identity information label.
Specifically, when determining the backfat information of the livestock individual to be detected, the backfat detection model can be adopted, two types of inputs can exist in the backfat detection model, the first type of input is body size data, and the second type of input is body size data and identity information. No matter which type of input, the backfat detection model can output backfat information. The backfat detection model may specifically be a machine learning model, such as: linear regression, random forest, xgboost, etc. For the first type of input, the backfat detection model can be obtained by training body size data of a sample livestock individual carrying a backfat information label. After the body size data are input into the backfat detection model, the backfat detection model can determine the identity information of the individual livestock to be detected through the body size data, and then the backfat information is detected by combining the identity information and the body size information. For the second type of input, the backfat detection model can be obtained through training the body size data of the sample livestock individual carrying the backfat information label and the identity information label.
In the embodiment of the invention, the backfat detection model is introduced to realize the detection of the backfat information, so that the detection process can be simplified, and meanwhile, the detection efficiency and accuracy can be improved.
On the basis of the embodiment, the livestock individual backfat detection method provided by the embodiment of the invention has the advantages that the backfat detection model comprises a plurality of backfat detection layers corresponding to different identity information;
correspondingly, will the body size data input is to backfat detection model, obtain by backfat detection model output backfat information specifically includes:
and inputting the body size data into the backfat detection layer corresponding to the identity information in the backfat detection model to obtain the backfat information output by the backfat detection layer corresponding to the identity information.
Specifically, because the correspondence between the size data and the backfat information of the to-be-detected livestock individuals with different identity information is different, backfat detection layers for detecting backfat information are respectively constructed for the livestock individuals with different identity information in the embodiment of the invention, each backfat detection layer corresponds to identity information, and the backfat detection model includes all backfat detection layers.
When the body size data of the to-be-detected livestock individual is input into the backfat detection model, the body size data of the to-be-detected livestock individual is specifically input into a backfat detection layer corresponding to the identity information of the to-be-detected livestock individual, and the backfat detection layer detects and outputs the backfat information of the to-be-detected livestock individual. Preferably, the identity information in the embodiment of the present invention may specifically be a variety, a growth stage, and the like, that is, the backfat detection layer may correspond to a variety of the individual livestock, a growth stage of the individual livestock, or different growth stages of different varieties. The growth stage may be in units of months or ten days, which is not specifically limited in the embodiment of the present invention.
When each backfat detection layer is trained, specifically, the body size data of a sample livestock individual of certain identity information is used as an independent variable, the backfat information of the sample livestock individual is used as a dependent variable, a plurality of models are used for training, and then the model structure and the parameters with the minimum detection error are selected as the layer structure and the layer parameters of the backfat detection layer corresponding to the identity information.
In the embodiment of the invention, the backfat detection model is divided into the backfat detection layers corresponding to the identity information, and the backfat information of the individual to be detected is detected through the backfat detection layer corresponding to the identity information of the individual to be detected, so that the backfat detection error can be effectively reduced, and the backfat detection accuracy is improved.
On the basis of the foregoing embodiment, in the method for detecting the backfat of the livestock individual provided in the embodiment of the present invention, the step of inputting the body size data to the backfat detection layer corresponding to the identity information in the backfat detection model to obtain the backfat information output by the backfat detection layer corresponding to the identity information further includes:
and inputting the body size data into an identity classification layer in the backfat detection model to obtain the identity information output by the identity classification layer.
Specifically, in the embodiment of the invention, the identity information of the individual livestock to be detected can be obtained by automatically classifying the back fat detection model through the input body size data. When the identity information of the to-be-detected livestock individuals is obtained through automatic classification of the backfat detection model, the backfat detection model can further comprise an identity classification layer for determining the identity information of the to-be-detected livestock individuals according to the input body size data of the to-be-detected livestock individuals.
The identity classification layer can be constructed based on classification models such as a generalized linear model and a classification tree. When the identity classification layer is trained, specifically, the body size data of the sample livestock individual is used as an independent variable, the identity information of the sample livestock individual is used as a dependent variable, a plurality of models are used for training, and then the model structure and the parameters with the minimum classification error are selected as the layer structure and the layer parameters of the identity classification layer.
According to the embodiment of the invention, the identity information of the taxi taking processing individuals can be automatically classified and acquired through the identity classification layer in the backfat detection model, so that the inconvenience of manually acquiring the identity information can be avoided, and the automation and convenience of acquiring the identity information are improved.
On the basis of the foregoing embodiment, the method for detecting backfat of an individual of a livestock provided in the embodiment of the present invention, which determines backfat information of the individual of the livestock to be detected based on the body size data of the individual of the livestock to be detected, or based on the body size data of the individual of the livestock to be detected and the identity information of the individual of the livestock to be detected, specifically includes:
and if the body size data is in a preset general body size data interval, determining the backfat information based on the body size data or the body size data and the identity information.
Specifically, in the embodiment of the invention, after the body size data of the individual livestock to be detected is determined, the usability of the body size data can be judged, that is, whether the calculated body size data is accurate or not can be judged by combining the body size data of the individual livestock under the normal condition.
And judging whether the body size data of the individual livestock to be detected is in a preset general body size data interval or not, if so, indicating that the body size data is accurate and available, and determining backfat information according to the body size data or the body size data and identity information. Otherwise, if the backfat information is not in the preset range, the problem that the body size data is possibly lack of data due to the reasons of acquisition or equipment conditions and the like is solved, and the usability is not available, so that the backfat information cannot be determined continuously according to the body size data or according to the body size data and the identity information. And subsequently, the body size data can be acquired again and judged. The general body size data section may be a section range determined by body size data of a large number of individual stock animals in a normal condition, for example, the shoulder width in the body size data may correspond to a general shoulder width section.
According to the embodiment of the invention, usability judgment is carried out on the obtained body size data, so that backfat information can be successfully detected, and the high accuracy of a detection result is ensured.
Fig. 2 is a schematic structural view of an animal individual backfat detection device provided in an embodiment of the present invention, and as shown in fig. 2, the animal individual backfat detection device includes: a body size data determination module 21 and a backfat information determination module 22. Wherein the content of the first and second substances,
the body size data determining module 21 is used for determining three-dimensional point cloud data of the individual livestock to be detected and determining body size data of the individual livestock to be detected based on the three-dimensional point cloud data;
the backfat information determining module 22 is configured to determine backfat information of the individual to be tested based on the body size data of the individual to be tested, or based on the body size data of the individual to be tested and the identity information of the individual to be tested.
Specifically, the functions of the modules in the livestock individual backfat detection device provided in the embodiment of the present invention correspond to the operation flows of the steps in the embodiments of the method one to one, and the achieved effects are also consistent, for specific reference, the embodiments of the present invention are not described again.
On the basis of the foregoing embodiment, in the livestock individual backfat detection device provided in the embodiment of the present invention, the body size data determining module is specifically configured to:
projecting the three-dimensional point cloud data to a horizontal plane to obtain the back contour of the individual livestock to be detected, or projecting the three-dimensional point cloud data to the horizontal plane and a vertical plane perpendicular to the trunk of the individual livestock to be detected respectively to obtain the back contour and the hip contour of the individual livestock to be detected;
determining hip width in the body ruler data based on the back contour and/or the hip contour, and determining shoulder width, waist width and bust in the body ruler data based on the back contour.
On the basis of the foregoing embodiment, in the livestock individual backfat detection device provided in the embodiment of the present invention, the body size data determining module is specifically configured to:
determining two shoulder width key points and two waist width key points in the back contour, determining the shoulder width based on the two shoulder width key points, and determining the waist width based on the two waist width key points;
determining the chest contour of the individual livestock to be detected based on the three-dimensional point cloud data and a vertical plane containing the two shoulder width key points;
and determining the chest circumference of the individual livestock to be detected based on the chest contour.
On the basis of the foregoing embodiment, in the livestock individual backfat detection device provided in the embodiment of the present invention, the backfat information determining module is specifically configured to:
inputting the body size data or the body size data and the identity information into a backfat detection model to obtain the backfat information output by the backfat detection model;
the backfat detection model is obtained based on body size data of the sample livestock individual carrying the backfat information label or based on body size data training of the sample livestock individual carrying the backfat information label and the identity information label.
On the basis of the embodiment, the livestock individual backfat detection device provided by the embodiment of the invention comprises a plurality of backfat detection layers corresponding to different identity information in a backfat detection model;
correspondingly, the backfat information determination module is specifically configured to:
and inputting the body size data into the backfat detection layer corresponding to the identity information in the backfat detection model to obtain the backfat information output by the backfat detection layer corresponding to the identity information.
On the basis of the above embodiments, the livestock individual backfat detection device provided in the embodiments of the present invention further includes: an identity classification layer;
correspondingly, the backfat information determination module is specifically configured to: and inputting the body size data into an identity classification layer in the backfat detection model to obtain the identity information output by the identity classification layer.
On the basis of the foregoing embodiment, in the livestock individual backfat detection device provided in the embodiment of the present invention, the backfat information determining module is specifically configured to:
and if the body size data is in a preset general body size data interval, determining the backfat information based on the body size data or the body size data and the identity information.
Fig. 3 illustrates a physical structure diagram of an electronic device, which may include, as shown in fig. 3: a processor (processor)310, a communication Interface (communication Interface)320, a memory (memory)330 and a communication bus 340, wherein the processor 310, the communication Interface 320 and the memory 330 communicate with each other via the communication bus 340. The processor 310 may invoke logic instructions in the memory 330 to perform a method for detecting backfat in an individual animal, comprising: determining three-dimensional point cloud data of an individual of a livestock to be detected, and determining body size data of the individual of the livestock to be detected based on the three-dimensional point cloud data; and determining backfat information of the individual of the livestock to be detected based on the body size data of the individual of the livestock to be detected or based on the body size data of the individual of the livestock to be detected and the identity information of the individual of the livestock to be detected.
In addition, the logic instructions in the memory 330 may be implemented in the form of software functional units and stored in a computer readable storage medium when the software functional units are sold or used as independent products. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
In another aspect, embodiments of the present invention further provide a computer program product, where the computer program product includes a computer program stored on a non-transitory computer-readable storage medium, the computer program includes program instructions, and when the program instructions are executed by a computer, the computer is capable of executing the method for detecting backfat of an individual livestock provided by the above-mentioned method embodiments, including: determining three-dimensional point cloud data of an individual of a livestock to be detected, and determining body size data of the individual of the livestock to be detected based on the three-dimensional point cloud data; and determining backfat information of the individual of the livestock to be detected based on the body size data of the individual of the livestock to be detected or based on the body size data of the individual of the livestock to be detected and the identity information of the individual of the livestock to be detected.
In still another aspect, an embodiment of the present invention further provides a non-transitory computer-readable storage medium, on which a computer program is stored, where the computer program is implemented by a processor to execute the method for detecting backfat of an individual livestock provided by the above embodiments, and the method includes: determining three-dimensional point cloud data of an individual of a livestock to be detected, and determining body size data of the individual of the livestock to be detected based on the three-dimensional point cloud data; and determining backfat information of the individual of the livestock to be detected based on the body size data of the individual of the livestock to be detected or based on the body size data of the individual of the livestock to be detected and the identity information of the individual of the livestock to be detected.
The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. With this understanding in mind, the above-described technical solutions may be embodied in the form of a software product, which can be stored in a computer-readable storage medium such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in the embodiments or some parts of the embodiments.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.

Claims (10)

1. A livestock individual backfat detection method is characterized by comprising the following steps:
determining three-dimensional point cloud data of an individual of a livestock to be detected, and determining body size data of the individual of the livestock to be detected based on the three-dimensional point cloud data;
and determining backfat information of the individual of the livestock to be detected based on the body size data of the individual of the livestock to be detected or based on the body size data of the individual of the livestock to be detected and the identity information of the individual of the livestock to be detected.
2. The livestock individual backfat detection method according to claim 1, wherein the determining of the body size data of the livestock individual to be detected based on the three-dimensional point cloud data specifically comprises:
projecting the three-dimensional point cloud data to a horizontal plane to obtain the back contour of the individual livestock to be detected, or projecting the three-dimensional point cloud data to the horizontal plane and a vertical plane perpendicular to the trunk of the individual livestock to be detected respectively to obtain the back contour and the hip contour of the individual livestock to be detected;
determining hip width in the body ruler data based on the back contour and/or the hip contour, and determining shoulder width, waist width and bust in the body ruler data based on the back contour.
3. The livestock individual backfat detection method according to claim 2, wherein the step of determining the shoulder width, waist width and chest circumference in the body size data based on the back contour specifically comprises the steps of:
determining two shoulder width key points and two waist width key points in the back contour, determining the shoulder width based on the two shoulder width key points, and determining the waist width based on the two waist width key points;
determining the chest contour of the individual livestock to be detected based on the three-dimensional point cloud data and a vertical plane containing the two shoulder width key points;
and determining the chest circumference of the individual livestock to be detected based on the chest contour.
4. The method for detecting the backfat of the livestock individual according to claim 1, wherein the step of determining the backfat information of the livestock individual to be detected based on the body ruler data of the livestock individual to be detected or based on the body ruler data of the livestock individual to be detected and the identity information of the livestock individual to be detected specifically comprises the steps of:
inputting the body size data or the body size data and the identity information into a backfat detection model to obtain the backfat information output by the backfat detection model;
the backfat detection model is obtained based on body size data of the sample livestock individual carrying the backfat information label or based on body size data training of the sample livestock individual carrying the backfat information label and the identity information label.
5. The livestock individual backfat detection method according to claim 4, wherein the backfat detection model comprises a plurality of backfat detection layers corresponding to different identity information;
correspondingly, will the body size data input is to backfat detection model, obtain by backfat detection model output backfat information specifically includes:
and inputting the body size data into the backfat detection layer corresponding to the identity information in the backfat detection model to obtain the backfat information output by the backfat detection layer corresponding to the identity information.
6. The livestock individual backfat detection method according to claim 5, wherein the step of inputting the body size data into the backfat detection layer corresponding to the identity information in the backfat detection model to obtain the backfat information output by the backfat detection layer corresponding to the identity information further comprises the steps of:
and inputting the body size data into an identity classification layer in the backfat detection model to obtain the identity information output by the identity classification layer.
7. The method for detecting the backfat of the livestock individual according to any one of claims 1 to 6, wherein the step of determining the backfat information of the livestock individual to be detected based on the body size data of the livestock individual to be detected or based on the body size data of the livestock individual to be detected and the identity information of the livestock individual to be detected specifically comprises the following steps:
and if the body size data is in a preset general body size data interval, determining the backfat information based on the body size data or the body size data and the identity information.
8. An individual backfat detection device of domestic animal, characterized by, includes:
the body size data determining module is used for determining three-dimensional point cloud data of the individual livestock to be detected and determining body size data of the individual livestock to be detected based on the three-dimensional point cloud data;
and the backfat information determining module is used for determining backfat information of the individual to be detected based on the body size data of the individual to be detected, or based on the body size data of the individual to be detected and the identity information of the individual to be detected.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor executes the program to implement the steps of the method for detecting backfat of an individual livestock as claimed in any one of claims 1 to 7.
10. A non-transitory computer readable storage medium having a computer program stored thereon, wherein the computer program when executed by a processor implements the steps of the method for detecting backfat in an individual livestock of claims 1 to 7.
CN202011232362.5A 2020-11-06 2020-11-06 Method and device for detecting backfat of livestock individuals, electronic equipment and storage medium Active CN112288793B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011232362.5A CN112288793B (en) 2020-11-06 2020-11-06 Method and device for detecting backfat of livestock individuals, electronic equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011232362.5A CN112288793B (en) 2020-11-06 2020-11-06 Method and device for detecting backfat of livestock individuals, electronic equipment and storage medium

Publications (2)

Publication Number Publication Date
CN112288793A true CN112288793A (en) 2021-01-29
CN112288793B CN112288793B (en) 2023-05-26

Family

ID=74351690

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011232362.5A Active CN112288793B (en) 2020-11-06 2020-11-06 Method and device for detecting backfat of livestock individuals, electronic equipment and storage medium

Country Status (1)

Country Link
CN (1) CN112288793B (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113124802A (en) * 2021-03-30 2021-07-16 章瀚文 Method and device for rapidly measuring shin circumference of poultry
CN114051973A (en) * 2022-01-17 2022-02-18 北京探感科技股份有限公司 Intelligent feeding method, device and system based on visual animal body ruler
CN116363141A (en) * 2023-06-02 2023-06-30 四川省畜牧科学研究院 Pregnant sow intelligent body type evaluation device and system

Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN203206909U (en) * 2013-05-09 2013-09-25 中国农业大学 Data acquisition device and pig growth form detection system using same
US20180189561A1 (en) * 2016-12-30 2018-07-05 Morphotrust Usa, Llc Identity document validation using biometric image data
CN108416003A (en) * 2018-02-27 2018-08-17 百度在线网络技术（北京）有限公司 A kind of picture classification method and device, terminal, storage medium
CN108733751A (en) * 2018-04-09 2018-11-02 广东温氏食品集团股份有限公司 Fat information monitoring system and method are surveyed in a kind of pig farm
CN110754426A (en) * 2019-11-28 2020-02-07 深圳市慧农科技有限公司 Method and system for automatically detecting backfat body and automatically feeding animals
CN110837768A (en) * 2018-08-16 2020-02-25 武汉大学 Rare animal protection oriented online detection and identification method
CN110986788A (en) * 2019-11-15 2020-04-10 华南农业大学 Automatic measurement method based on three-dimensional point cloud livestock phenotype body size data
CN111243005A (en) * 2020-01-07 2020-06-05 洛阳语音云创新研究院 Livestock weight estimation method, device, equipment and computer readable storage medium
CN111353416A (en) * 2020-02-26 2020-06-30 广东温氏种猪科技有限公司 Posture detection method, system and storage medium based on livestock three-dimensional measurement
US20200305801A1 (en) * 2019-03-27 2020-10-01 Panasonic Intellectual Property Management Co., Ltd. Determining health state of individuals

Patent Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN203206909U (en) * 2013-05-09 2013-09-25 中国农业大学 Data acquisition device and pig growth form detection system using same
US20180189561A1 (en) * 2016-12-30 2018-07-05 Morphotrust Usa, Llc Identity document validation using biometric image data
CN108416003A (en) * 2018-02-27 2018-08-17 百度在线网络技术（北京）有限公司 A kind of picture classification method and device, terminal, storage medium
CN108733751A (en) * 2018-04-09 2018-11-02 广东温氏食品集团股份有限公司 Fat information monitoring system and method are surveyed in a kind of pig farm
CN110837768A (en) * 2018-08-16 2020-02-25 武汉大学 Rare animal protection oriented online detection and identification method
US20200305801A1 (en) * 2019-03-27 2020-10-01 Panasonic Intellectual Property Management Co., Ltd. Determining health state of individuals
CN110986788A (en) * 2019-11-15 2020-04-10 华南农业大学 Automatic measurement method based on three-dimensional point cloud livestock phenotype body size data
CN110754426A (en) * 2019-11-28 2020-02-07 深圳市慧农科技有限公司 Method and system for automatically detecting backfat body and automatically feeding animals
CN111243005A (en) * 2020-01-07 2020-06-05 洛阳语音云创新研究院 Livestock weight estimation method, device, equipment and computer readable storage medium
CN111353416A (en) * 2020-02-26 2020-06-30 广东温氏种猪科技有限公司 Posture detection method, system and storage medium based on livestock three-dimensional measurement

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
MARTIN ENGELCKE ET.AL: "Vote3Deep: Fast Object Detection in 3D Point Clouds Using Efficient Convolutional Neural Networks" *
尹令等: "多视角深度相机的猪体三维点云重构及体尺测量", 《农业工程学报》 *
滕光辉 等: "基于Kinect传感器的无接触式母猪体况评分方法", 《农业工程学报》 *

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113124802A (en) * 2021-03-30 2021-07-16 章瀚文 Method and device for rapidly measuring shin circumference of poultry
CN114051973A (en) * 2022-01-17 2022-02-18 北京探感科技股份有限公司 Intelligent feeding method, device and system based on visual animal body ruler
CN116363141A (en) * 2023-06-02 2023-06-30 四川省畜牧科学研究院 Pregnant sow intelligent body type evaluation device and system
CN116363141B (en) * 2023-06-02 2023-08-18 四川省畜牧科学研究院 Pregnant sow intelligent body type evaluation device and system

Also Published As

Publication number Publication date
CN112288793B (en) 2023-05-26

Similar Documents

Publication Publication Date Title
Wang et al. ASAS-NANP SYMPOSIUM: Applications of machine learning for livestock body weight prediction from digital images
Qiao et al. Intelligent perception for cattle monitoring: A review for cattle identification, body condition score evaluation, and weight estimation
CN107635509B (en) Improved method for determining body condition score, body weight and fertility status and related device
CN107180438B (en) Method for estimating size and weight of yak and corresponding portable computer device
CN112288793B (en) Method and device for detecting backfat of livestock individuals, electronic equipment and storage medium
US11627726B2 (en) System and method of estimating livestock weight
CN111243005B (en) Livestock weight estimation method, apparatus, device and computer readable storage medium
US9104906B2 (en) Image analysis for determining characteristics of animals
CA2744146C (en) Arrangement and method for determining a body condition score of an animal
JP4128791B2 (en) Sexual differentiation technology for sperm eggs
Zhang et al. Algorithm of sheep body dimension measurement and its applications based on image analysis
CN111297367A (en) Animal state monitoring method and device, electronic equipment and storage medium
CN109068613A (en) Health status estimating device
Tscharke et al. Review of methods to determine weight and size of livestock from images
Lu et al. Extracting body surface dimensions from top-view images of pigs
Li et al. A posture-based measurement adjustment method for improving the accuracy of beef cattle body size measurement based on point cloud data
CN110414369B (en) Cow face training method and device
Burke et al. An analysis of the feasibility of using image processing to estimate the live weight of sheep
KR102372107B1 (en) Image-based sow farrowing notification system
CN113762745A (en) Live pig body shape assessment method and device based on machine vision
Yuan et al. Stress-free detection technologies for pig growth based on welfare farming: A review
Thomas et al. The use of image data in the assessment of equine conformation-“limitations and solutions”
EP3606349A1 (en) Method and system for classifying animal carcass
Siachos et al. Development and validation of a fully automated 2D imaging system generating body condition scores for dairy cows using machine learning
CN114266811A (en) Livestock body condition scoring method and device, electronic equipment and storage medium

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant