CN115345901B - Animal motion behavior prediction method and system and camera system - Google Patents
Animal motion behavior prediction method and system and camera system Download PDFInfo
- Publication number
- CN115345901B CN115345901B CN202211270218.XA CN202211270218A CN115345901B CN 115345901 B CN115345901 B CN 115345901B CN 202211270218 A CN202211270218 A CN 202211270218A CN 115345901 B CN115345901 B CN 115345901B
- Authority
- CN
- China
- Prior art keywords
- animal
- moving
- motion
- moving speed
- initial
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/207—Analysis of motion for motion estimation over a hierarchy of resolutions
-
- 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
- G06T2207/10016—Video; Image sequence
-
- 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)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a method and a system for predicting animal motion behaviors and a camera system, wherein the method comprises the following steps: s1: acquiring an exercise behavior prediction information base and an exercise behavior feature base; s2: acquiring a video image in real time, acquiring the animal motion gravity center in the video image according to a motion gravity center library, and calculating the moving speed and the moving direction of the animal motion gravity center according to the video data of the initial time period in the video image; s3: acquiring an initial moving speed and an initial moving direction which are matched with the moving speed and the moving direction of the moving gravity center of the animal in a moving behavior characteristic library; s4: and acquiring a corresponding motion track in the motion behavior prediction information base according to the initial moving speed and the initial moving direction which are matched with the moving speed and the moving direction of the center of gravity of the animal motion and acquired in the S3 as the animal motion behavior prediction track. The invention can quickly obtain the predicted motion track, quickly calculate the corrected track through the track and improve the quality of the camera shooting finished product.
Description
Technical Field
The invention relates to the technical field of animal motion behavior prediction, in particular to a method and a system for predicting animal motion behavior and a camera system.
Background
One of the key technologies in photography is to focus on a subject accurately. If the object is moving, the moving object needs to be kept in a following state, namely, in focus during shooting. The filming of moving objects generally places high demands on the cameraman's level of focus. Based on this need, auto focus following techniques are developed and applied in many scenarios.
For the focus-following photographing of a moving object, two approaches are generally adopted. One is to perform target identification and extraction on an image acquired by a camera, and calculate the position of an object to perform feedback type focus following adjustment; one is to perform active adjustment control of the focus following according to a preset track on a target with a specific track.
When the method is applied to living objects such as animals, pets and the like, the scheme for performing feedback type follow-up focus adjustment by target extraction and identification has the following defects: (1) the feedback type focus following adjustment requires that the camera system has stronger image recognition computing capability and focusing corresponding speed, and the requirements are difficult to meet when the moving speed of the object is higher or the requirement on focus following is higher; (2) the living body itself such as an animal and a pet is not fixed (the difference is large when different postures of the animal such as squatting, standing, running, jumping, curling and the like are taken as picture objects), and the living body has the characteristics of a plurality of change types and a high change speed, which brings great difficulty to the target identification process itself. The tracking is actively adjusted and controlled through the preset track, the target shooting effect on the specific track is good, but when the animal moves, the initial speed and the direction are not characteristic, so that the track of each movement is different, therefore, the tracking shooting on the non-specific track moving target can be completed by automatically following the moving target through common shooting equipment, the accurate animal moving track needs to be quickly predicted, and the tracking shooting is performed based on the track.
Disclosure of Invention
In order to solve the above-mentioned problems in the prior art, the present invention provides a method and a system for predicting animal motion behavior, and a camera system.
In order to achieve the purpose, the invention adopts the technical scheme that:
a method for predicting animal motion behavior is characterized by comprising the following specific steps:
s1: acquiring a motion behavior prediction information base and a motion behavior feature base, wherein the motion behavior prediction information base corresponds to the motion behavior feature base one by one, the motion behavior prediction information base is provided with a motion track of an animal, and the motion behavior feature base is provided with an initial moving direction and an initial moving speed of the animal;
s2: acquiring a video image in real time, acquiring the animal motion gravity center in the video image according to a motion gravity center library, and calculating the moving speed and the moving direction of the animal motion gravity center according to the video data of the initial time period in the video image;
s3: acquiring an initial moving speed matched with the moving speed of the moving gravity center of the animal in a moving behavior characteristic library, and acquiring an initial moving direction matched with the moving direction of the moving gravity center of the animal in the moving behavior characteristic library;
s4: and acquiring a corresponding motion track in the motion behavior prediction information base according to the initial moving speed and the initial moving direction which are matched with the moving speed and the moving direction of the center of gravity of the animal motion and acquired in the S3, and taking the motion track as the motion track predicted by the animal motion behavior.
Preferably, the method for establishing the motion center-of-gravity library in step S2 is as follows:
the building method of the motion gravity center library in the step S2 is as follows:
step S21, acquiring original data of an animal which leaves the ground and is only subjected to gravity, wherein the original data comprises video data and image data;
s22, acquiring a reference video image only containing animals in the original data;
step S23, acquiring the relative speed of the local body of the animal at the first moment from the reference video image;
step S24, fitting the relative speed of the local body of the first moment in the horizontal direction to obtain the average speed of the animal in the horizontal direction;
s25, acquiring a two-dimensional image block with the same average speed in a reference video image, wherein the coordinate of the two-dimensional image block is a motion gravity center image;
and S26, storing the two-dimensional image blocks into a motion gravity center library.
Preferably, in step S24, the average velocity of the living subject in the horizontal direction is,
wherein the local body has a head, eyes, front paws, rear paws;
wherein vm is the speed of the local body in the horizontal direction at the first moment, and m is an integer greater than 1;
where F is a fitted function of the average velocity.
Preferably, in step S2, the moving speed, the moving direction and the moving included angle of the center of gravity of the animal motion are calculated according to the video data of the initial time period in the video image, and the moving included angle is the included angle between the initial moving direction of the animal and the horizontal direction; and if the movement included angle is smaller than 10 degrees, repeating the step S2, and if the movement included angle is larger than or equal to 10 degrees, executing the subsequent steps.
Preferably, the initial moving speed in the motion behavior feature library in the step S1 is 0m/S-20m/S,
wherein the initial moving speed Vs = Fs (D, V1), D is a distance between the animal and the camera, V1 is a moving speed of the animal in the video data of the initial time period, and Fs is a calculation function of the initial moving speed Vs;
the initial moving speed has a multi-stage moving speed range comprising a first moving speed, a second moving speed, a third moving speed and a fourth moving speed;
the method for acquiring the initial moving speed matched with the moving speed of the moving gravity center of the animal in the moving behavior characteristic library comprises the following steps:
s311: respectively comparing the moving speed of the center of gravity of the animal movement with the first moving speed, the second moving speed, the third moving speed and the fourth moving speed, and taking the moving speed range with the minimum absolute value of the moving speed difference of the center of gravity of the animal movement as the moving speed to be compared;
s312: calculating absolute values K1, K2, K3, kn of differences between the moving speed of the center of gravity of the animal movement and a plurality of initial moving speeds in the moving speeds to be compared, and then calculating a minimum value Kmin;
s313: the initial moving speed of Kmin is calculated as the initial moving speed matched with the moving speed of the moving gravity center of the animal.
Preferably, the first moving speed is greater than 0m/s and less than 3m/s, the second moving speed is greater than or equal to 3m/s and less than 5m/s, the third moving speed is greater than or equal to 5m/s and less than 10m/s, and the fourth moving speed is greater than or equal to 10m/s and less than 20m/s.
Preferably, the initial moving direction in the motion behavior feature library in step S1 is a direction relative to a horizontal plane, and is represented as-90 degrees to 90 degrees, and the initial moving direction has a multi-stage moving direction range, including a first moving direction, a second moving direction, a third moving direction, and a fourth moving direction;
the step of obtaining the initial moving direction matched with the moving direction of the animal moving gravity center in the moving behavior feature library in the step S3 is as follows:
s321: comparing the moving direction of the center of gravity of the animal motion with a first moving direction, a second moving direction, a third moving direction and a fourth moving direction, and taking the moving direction range with the minimum absolute value of the moving direction difference of the center of gravity of the animal motion as the moving direction to be compared;
s322: calculating absolute values L1, L2, L3, ln of differences between the moving direction of the center of gravity of the animal motion and a plurality of initial moving directions in the moving direction to be compared, and then calculating a minimum value Lmin;
s323: and calculating the initial moving direction of the Lmin as the initial moving direction matched with the moving direction of the center of gravity of the animal motion.
Preferably, the motion behavior prediction information base corresponds to the motion behavior feature base one by one through the matching module, the actual animal motion track corresponding to the animal motion track in the motion behavior prediction information base is recorded at fixed time intervals, and the matching module and the animal motion track in the motion behavior prediction information base corresponding to the matching module are updated through the actual animal motion track.
Preferably, the fixed time period is 7 to 15 days.
An animal motion behavior prediction system is characterized in that the animal motion behavior prediction method is applied.
A camera system based on animal motion behavior prediction is characterized in that the camera system comprises a camera module,
an imaging module;
a processing module;
a storage module;
a communication module;
the imaging module can acquire a video image;
the storage module can store the video image acquired by the imaging module;
the communication module can receive and send signals;
the processing module can adjust the lens direction or/and the focal length of the imaging module according to the instruction signal;
the instruction signal comprises lens direction adjustment information or/and focal length adjustment information.
The invention has the beneficial effects that:
the animal action behaviors shot by the method and the system, particularly the running and jumping actions, have high definition of the shot video, no shaking of the picture and extremely high sharing value, and the method can be realized by general hardware equipment to shoot the animal motion video with high definition of the picture and high shooting quality.
Drawings
FIG. 1 is a schematic flow chart of a method for predicting animal movement behavior;
FIG. 2 is a schematic flow chart of obtaining an initial moving speed matched with the moving speed of the moving gravity center of an animal from a motion behavior feature library;
FIG. 3 is a schematic flow chart of obtaining an initial moving direction matching with the moving direction of the moving gravity center of an animal from a motion behavior feature library;
FIG. 4 is a schematic diagram of a two-dimensional image block representing an image of the center of gravity of an animal's motion;
fig. 5 is a block diagram of a camera system based on animal motion behavior prediction.
Detailed Description
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 only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without making any creative effort based on the embodiments in the present invention, belong to the protection scope of the present invention.
Referring to fig. 1-5, the present invention provides the following embodiments:
example 1
Referring to fig. 1 and 4, a method for predicting animal motion behavior is characterized by comprising the following specific steps:
s1: acquiring a motion behavior prediction information base and a motion behavior feature base, wherein the motion behavior prediction information base corresponds to the motion behavior feature base one by one, the motion behavior prediction information base is provided with a motion track of an animal, and the motion behavior feature base is provided with an initial moving direction and an initial moving speed of the animal;
s2: acquiring a video image in real time, acquiring the animal motion gravity center in the video image according to a motion gravity center library, and calculating the moving speed and the moving direction of the animal motion gravity center according to the video data of the initial time period in the video image;
s3: acquiring an initial moving speed matched with the moving speed of the moving gravity center of the animal in a moving behavior characteristic library, and acquiring an initial moving direction matched with the moving direction of the moving gravity center of the animal in the moving behavior characteristic library;
s4: and acquiring a corresponding motion track in the motion behavior prediction information base according to the initial moving speed and the initial moving direction which are matched with the moving speed and the moving direction of the center of gravity of the animal motion and acquired in the S3, and taking the motion track as the motion track predicted by the animal motion behavior.
With the rise of each large and short video platform, a pet owner who feeds pets can choose to shoot and record some interesting action behaviors of animals and share the interesting action behaviors to each large and short video platform, but wants to shoot a section of complete action behaviors (especially action behaviors with higher movement speed, such as running and jumping of the animals) of the animals clearly and with coordinated picture occupation, needs to skillfully operate photographic equipment, prejudges the movement track of the animals before the animals start to move, enables the lenses to synchronously move along with the animals in a time period from the beginning to the end of the actions of the animals, and simultaneously adjusts the focal length to a proper size according to the distance between the animals and the lenses, so that the requirement on the follow-up shooting capability is higher.
The existing shooting method for actively adjusting and controlling the follow focus through the preset track has a good shooting effect only on the target with the specific track, but when an animal moves, the track of each movement is different because the initial speed and the direction of the animal are not characteristic, and the shooting effect is poor due to the method for shooting the movement behavior of the animal.
Considering that most pet owners are not specialized in photography, in order to help most people to clearly shoot some interesting action behaviors of animals, referring to fig. 1, the embodiment provides an animal motion behavior prediction method, based on which focus following shooting can be well realized, and the specific steps are as follows: a method for predicting animal motion behavior is characterized by comprising the following specific steps: s1: acquiring a motion behavior prediction information base and a motion behavior characteristic base, wherein the motion behavior prediction information base corresponds to the motion behavior characteristic base one by one, the motion behavior prediction information base is provided with a motion track of an animal, and the motion behavior characteristic base is provided with an initial moving direction and an initial moving speed of the animal; s2: acquiring a video image in real time, acquiring the animal motion gravity center in the video image according to a motion gravity center library, and calculating the moving speed and the moving direction of the animal motion gravity center according to the video data of the initial time period in the video image; s3: acquiring an initial moving speed matched with the moving speed of the moving gravity center of the animal in a moving behavior characteristic library, and acquiring an initial moving direction matched with the moving direction of the moving gravity center of the animal in the moving behavior characteristic library; s4: and acquiring a corresponding motion track in the motion behavior prediction information base according to the initial moving speed and the initial moving direction which are matched with the moving speed and the moving direction of the center of gravity of the animal motion and acquired in the S3, and taking the motion track as the motion track predicted by the animal motion behavior.
The inventor finds that when an animal starts to move, the movement speed and the movement direction of the movement gravity center at the initial moment of the movement of the animal determine the subsequent movement type and movement track of the animal, for example, the movement direction of the movement gravity center of the animal is a horizontal direction, which indicates that the movement is walking or running, and at this time, if the movement speed of the movement center is greater than a certain threshold value, the movement is running; if the moving direction of the center of gravity of the animal is not horizontal, the action is jumping. When the animal movement action is jumping, the movement track of the animal is also influenced by gravity, and a certain time is consumed for predicting the track only by calculating the moving speed and the moving direction of the movement gravity center at the initial moment of the animal movement, so that the shooting opportunity is possibly missed. In the embodiment, a prediction information base and a motion behavior feature base of animal motion behaviors are constructed, when actual prediction is carried out, the moving speed and the moving direction of the motion gravity center of the animal at the initial motion moment are compared with the initial moving speed and the initial moving direction in the motion behavior feature base to obtain the matched initial moving speed and initial moving direction, then the motion track matched with the initial moving speed and the initial moving direction in the motion behavior prediction information base is obtained, and the prediction track can be quickly obtained for focus-following shooting.
Example 2
Referring to fig. 4, in this embodiment, as a further improvement of the technical solution of embodiment 1, it is characterized in that,
the building mode of the motion gravity center library in the step S2 is as follows:
the building method of the motion gravity center library in the step S2 is as follows:
step S21, acquiring original data of an animal which leaves the ground and is only subjected to gravity, wherein the original data comprises video data and image data;
step S22, acquiring a reference video image only containing animals in original data;
step S23, acquiring the relative speed of the local body of the animal at the first moment from the reference video image;
step S24, fitting the relative speed of the local body of the first moment in the horizontal direction to obtain the average speed of the animal in the horizontal direction;
step S25, acquiring a two-dimensional image block with the same average speed in a reference video image, wherein the coordinate of the two-dimensional image block is a motion gravity center image;
and S26, storing the two-dimensional image blocks into a motion gravity center library.
The inventor finds that a pet in a home is shot, a shot object is relatively fixed, so that the moving gravity center of the pet can be identified from the moving track of the pet by learning and storing the object and then tracking the object; further, the relation between the motion center of gravity and the two-dimensional image posture can be constructed, so that the purpose of quickly positioning the center of gravity of the object through the two-dimensional image posture is achieved; it is also possible to construct the relative relationship of the center of gravity of the motion to the pose, the limb object (e.g., head, forepaw, etc.). Referring to fig. 4, in the embodiment, a two-dimensional image block a with the same average speed in a reference video image is obtained, the coordinate of the two-dimensional image block a is set as a motion barycentric coordinate, and when the two-dimensional image block a is actually shot, the motion barycentric of an animal is quickly located through the two-dimensional image block a, so that the generation speed of the corrected motion track information is further increased.
Preferably, in step S24, the average velocity of the living subject in the horizontal direction is,
wherein the local body has a head, eyes, front paws, rear paws;
wherein vm is the speed of the local body in the horizontal direction at the first moment, and m is an integer greater than 1;
where F is a fitted function of the average velocity.
Example 3
In this example, as a further improvement of the technical means of example 1, the present invention is characterized in that,
step S2, calculating the moving speed, the moving direction and the moving included angle of the center of gravity of the animal movement according to the initial time period video data in the video image, wherein the moving included angle is the included angle between the initial moving direction and the horizontal direction of the animal; and if the movement included angle is smaller than 10 degrees, repeating the step S2, and if the movement included angle is larger than or equal to 10 degrees, executing the subsequent steps.
In this embodiment, a movement included angle of a center of gravity of an animal movement, that is, an included angle between an initial movement direction of the animal and a horizontal direction, is further calculated from a video image of an initial period of the animal movement, and by determining whether the movement included angle is greater than or equal to 10 degrees, if the movement included angle is greater than or equal to 10 degrees, subsequent steps are executed; if the movement included angle is smaller than 10 degrees, the step S2 is repeated, in the actual execution process of the method, after the movement direction of the movement gravity center of the animal is calculated, if the movement included angle is larger than or equal to 10 degrees, the movement direction of the animal faces upwards or downwards, the animal is very likely to jump, therefore, the subsequent steps can be continuously executed to obtain the predicted track for the follow focus shooting, if the movement included angle is smaller than 10 degrees, the movement direction of the animal is parallel to the horizontal direction, and the animal cannot jump, so that the step S2 is repeated, the situation that equipment for executing the method is always in a working state is avoided, and the equipment load is increased. In the walking process of the animal, along with the movement and bending of the limbs of the animal, the movement direction of the movement center of gravity of the animal is inclined relative to the horizontal direction in a small range, the judgment state of the movement included angle is set to be greater than or equal to 10 degrees and smaller than 10 degrees, and the situation that the movement direction of the movement center of gravity is inclined relative to the horizontal direction in a small range is judged that the animal performs jumping action can be effectively avoided.
Example 4
Referring to fig. 2, in this embodiment, as a further improvement of the technical solution of embodiment 3, it is characterized in that,
the initial moving speed in the motion behavior feature library in the step S1 is 0m/S-20m/S,
wherein the initial moving speed Vs = Fs (D, V1), D is a distance between the animal and the camera, V1 is a moving speed of the animal in the video data of the initial time period, and Fs is a calculation function of the initial moving speed Vs;
the initial moving speed has a multi-stage moving speed range comprising a first moving speed, a second moving speed, a third moving speed and a fourth moving speed;
the step of obtaining the initial moving speed matched with the moving speed of the center of gravity of the animal motion in the motion behavior feature library in the step S3 is as follows:
s311: respectively comparing the moving speed of the center of gravity of the animal movement with the first moving speed, the second moving speed, the third moving speed and the fourth moving speed, and taking the moving speed range with the minimum absolute value of the moving speed difference of the center of gravity of the animal movement as the moving speed to be compared;
s312: calculating absolute values K1, K2, K3, kn of differences between the moving speed of the center of gravity of the animal movement and a plurality of initial moving speeds in the moving speeds to be compared, and then calculating a minimum value Kmin;
s313: the initial moving speed of Kmin is calculated as the initial moving speed matched with the moving speed of the moving gravity center of the animal.
The inventors have found that the movement speed of domestic pets (usually cats, dogs) is at most 20m/s,
in this embodiment, the initial moving speed in the motion behavior feature library in step S1 is set to be 0m/S-20m/S, and the initial moving speed is set to have multiple stages, including a first moving speed, a second moving speed, a third moving speed, and a fourth moving speed, the initial moving speed to which the moving speed of the center of gravity of the animal motion belongs is determined first, then absolute values K1, K2, K3,. Kn of differences between the moving speed of the center of gravity of the animal motion and the multiple initial moving speeds in the initial moving speed are calculated, then a minimum value Kmin the initial moving speed is calculated, and finally the initial moving speed calculated by Kmin is used as the initial moving speed matched with the moving speed of the center of gravity of the animal motion.
Preferably, the first moving speed is greater than 0m/s and less than 3m/s, the second moving speed is greater than or equal to 3m/s and less than 5m/s, the third moving speed is greater than or equal to 5m/s and less than 10m/s, and the fourth moving speed is greater than or equal to 10m/s and less than 20m/s.
The inventor finds that the moving speed of pets such as cats and dogs in an indoor environment is less than 5m/s at most of the time, the first moving speed is set to be greater than 0m/s and less than 3m/s in the embodiment, and the second moving speed is greater than or equal to 3m/s and less than 5m/s, so that the moving speed of the pets at most of the time is included, only the initial moving speed within 5m/s is needed to be compared at most of the predicted time, and the calculation amount is further reduced.
Example 5
Referring to fig. 3, in this embodiment, as a further improvement of the technical solution of embodiment 3, it is characterized in that,
the initial moving direction in the motion behavior feature library in the step S1 is a direction relative to a horizontal plane, and is represented as-90 degrees to 90 degrees, and the initial moving direction has a multi-stage moving direction range, including a first moving direction, a second moving direction, a third moving direction, and a fourth moving direction;
the step of obtaining the initial moving direction matched with the moving direction of the animal moving gravity center in the moving behavior feature library in the step S3 is as follows:
s321: comparing the moving direction of the center of gravity of the animal motion with a first moving direction, a second moving direction, a third moving direction and a fourth moving direction, and taking the moving direction range with the minimum absolute value of the moving direction difference of the center of gravity of the animal motion as the moving direction to be compared;
s322: calculating absolute values L1, L2, L3, ln of differences between the moving direction of the center of gravity of the animal motion and a plurality of initial moving directions in the moving direction to be compared, and then calculating a minimum value Lmin;
s323: and calculating the initial moving direction of the Lmin as the initial moving direction matched with the moving direction of the moving gravity center of the animal.
The inventor finds that the moving direction (the initial moving direction is a direction relative to the horizontal plane) of the domestic pet is between-90 degrees and 90 degrees, the embodiment sets the initial moving direction range in the step S1 to have a plurality of stages, including a first moving direction, a second moving direction, a third moving direction and a fourth moving direction, firstly judges a stage of initial moving direction to which the moving direction of the moving gravity center of the animal belongs, then calculates absolute values L1, L2, L3, ln of the difference between the moving direction of the moving gravity center of the animal and a plurality of initial moving directions in the stage of initial moving direction, and then calculates the minimum value Lmin, and finally, the initial moving direction of the Lmin is calculated and used as the initial moving direction matched with the moving direction of the moving gravity center of the animal, the moving direction of the moving gravity center does not need to be compared with the initial moving directions in the moving behavior feature library one by one, and only the initial moving direction closest to the moving direction of the moving gravity center can be judged by comparing with a plurality of initial moving directions in a certain section of moving directions in the first moving direction, the second moving direction, the third moving direction and the fourth moving direction, so that the calculated calculation amount is effectively reduced, and the prediction speed is increased.
Example 6
In this embodiment, as a further improvement of the means of embodiment 3, the feature is that,
the motion behavior prediction information base corresponds to the motion behavior feature base one by one through the matching module, the actual animal motion track corresponding to the animal motion track in the motion behavior prediction information base is recorded at intervals of fixed time, and the matching module and the animal motion track in the motion behavior prediction information base corresponding to the matching module are updated through the actual animal motion track.
The inventor finds that, for a certain fixed pet, the movement track corresponding to the movement characteristics of the pet changes along with the change of the body type of the pet (becoming fat or thin) or the change of the age (young, strong or old), in order to accurately predict the movement track of the pet in different body types of the pet in different age groups, the movement behavior prediction information base is arranged in the embodiment to be in one-to-one correspondence with the movement behavior characteristic base through the matching module, wherein the matching module can contain the relation information of the combination of the single initial movement speed and the single initial movement direction in the movement behavior characteristic base and the movement track of the animal matched with the combination, the matched movement track of the animal can be rapidly found through the relation information, the actual movement track of the animal corresponding to the movement track in the movement behavior prediction information base is recorded after the interval of fixed time period, and the movement track of the animal in the matching module and the movement behavior prediction information base corresponding to the actual movement track of the animal is updated through the movement track of the animal, namely, the updating of the movement track of the animal in the prediction information base is realized, and the pet in different age groups, the pet in different body types can accurately predict the movement track.
Preferably, the fixed time period is 7 to 15 days.
The body type of the pet generally changes obviously after 7-15 days, the fixed time is set to be 7-15 days in the embodiment, namely the animal motion track in the prediction information base is updated after at least seven days, the load on the system and the equipment is small, the method can be realized through common data processing equipment, and the application range of the prediction method is further expanded.
Example 7
An animal athletic performance prediction system using the method of any one of embodiments 1 to 6.
Example 8
Referring to fig. 5, a camera system based on animal motion behavior prediction, which applies the animal motion behavior prediction method of embodiments 1-6, is characterized by comprising,
an imaging module;
a processing module;
a storage module;
a communication module;
the imaging module can acquire a video image;
the storage module can store the video image acquired by the imaging module;
the communication module can receive and send signals;
the processing module can adjust the lens direction or/and the focal length of the imaging module according to the instruction signal;
the instruction signal comprises lens direction adjustment information or/and focal length adjustment information.
The embodiment provides a camera system based on template materials, which can finish the follow-up shooting of animals based on the animal motion behavior prediction method described in the embodiments 1 to 6, and can obtain the product with high camera finished product quality.
In the description of the embodiments of the present invention, it is to be understood that the terms "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "center", "top", "bottom", "inner", "outer", and the like indicate an orientation or positional relationship.
In the description of the embodiments of the present invention, it should be noted that, unless otherwise explicitly specified or limited, the terms "mounted," "connected," and "assembled" are to be construed broadly and may, for example, be fixedly connected, detachably connected, or integrally connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood in a specific case to those of ordinary skill in the art.
In the description of the embodiments of the invention, the particular features, structures, materials, or characteristics may be combined in any suitable manner in any one or more embodiments or examples.
In the description of the embodiments of the present invention, it is to be understood that "-" and "-" denote ranges of two numerical values, and the ranges include endpoints. For example, "A-B" means a range greater than or equal to A and less than or equal to B. "A to B" represents a range of A or more and B or less.
In the description of the embodiments of the present invention, the term "and/or" herein is only one kind of association relationship describing an associated object, and means that there may be three relationships, for example, a and/or B, and may mean: a exists alone, A and B exist simultaneously, and B exists alone. In addition, the character "/" herein generally indicates that the former and latter related objects are in an "or" relationship.
Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (10)
1. A method for predicting animal motion behaviors is characterized by comprising the following specific steps:
s1: acquiring a motion behavior prediction information base and a motion behavior feature base, wherein the motion behavior prediction information base corresponds to the motion behavior feature base one by one, the motion behavior prediction information base is provided with a motion track of an animal, and the motion behavior feature base is provided with an initial moving direction and an initial moving speed of the animal;
s2: acquiring a video image in real time, acquiring the animal motion gravity center in the video image according to a motion gravity center library, and calculating the moving speed and the moving direction of the animal motion gravity center according to the video data of the initial time period in the video image;
s3: acquiring an initial moving speed matched with the moving speed of the moving gravity center of the animal in a moving behavior characteristic library, and acquiring an initial moving direction matched with the moving direction of the moving gravity center of the animal in the moving behavior characteristic library;
s4: acquiring a corresponding motion track in the motion behavior prediction information base according to the initial moving speed and the initial moving direction which are matched with the moving speed and the moving direction of the center of gravity of the animal motion and acquired in the S3, and taking the motion track as the motion track of animal motion behavior prediction;
the building mode of the motion gravity center library in the step S2 is as follows:
step S21, acquiring original data of an animal which leaves the ground and is only subjected to gravity, wherein the original data comprises video data and image data;
s22, acquiring a reference video image only containing animals in the original data;
step S23, acquiring the relative speed of the local body of the animal at the first moment from the reference video image;
step S24, fitting the relative speed of the local body of the first moment in the horizontal direction to obtain the average speed of the animal in the horizontal direction;
step S25, acquiring a two-dimensional image block with the same average speed in a reference video image, wherein the coordinate of the two-dimensional image block is a motion gravity center image;
and S26, storing the two-dimensional image blocks into a motion gravity center library.
2. The method of claim 1, wherein the step of predicting the motor behavior of the animal,
in step S24, the average speed of the animal in the horizontal direction is,
wherein the local body has a head, eyes, front paws, rear paws;
wherein v1, v2, … …, vm are the speed of each local body in the horizontal direction at the first moment respectively, and m is an integer greater than 1;
where F is a fitted function of the average velocity.
3. The method of claim 1, wherein the step of predicting the motor behavior of the animal,
step S2, calculating the moving speed, the moving direction and the moving included angle of the center of gravity of the animal movement according to the initial time period video data in the video image, wherein the moving included angle is the included angle between the initial moving direction and the horizontal direction of the animal; if the moving included angle is smaller than 10 degrees, repeating the step S2, and if the moving included angle is larger than or equal to 10 degrees, executing the subsequent steps.
4. The method for predicting animal's motor behavior according to claim 3,
the initial moving speed in the motion behavior feature library in the step S1 is 0m/S-20m/S,
wherein the initial moving speed Vs = Fs (D, V1), D is a distance between the animal and the camera, V1 is a moving speed of the animal in the video data of the initial time period, and Fs is a calculation function of the initial moving speed Vs;
the initial moving speed has a multi-stage moving speed range comprising a first moving speed, a second moving speed, a third moving speed and a fourth moving speed;
the step of obtaining the initial moving speed matched with the moving speed of the animal moving gravity center in the moving behavior feature library in the step S3 is as follows:
s311: respectively comparing the moving speed of the center of gravity of the animal movement with the first moving speed, the second moving speed, the third moving speed and the fourth moving speed, and taking the moving speed range with the minimum absolute value of the moving speed difference of the center of gravity of the animal movement as the moving speed to be compared;
s312: calculating absolute values K1, K2, K3, kn of differences between the moving speed of the center of gravity of the animal movement and a plurality of initial moving speeds in the moving speeds to be compared, and then calculating a minimum value Kmin;
s313: the initial moving speed of Kmin is calculated as the initial moving speed matched with the moving speed of the moving gravity center of the animal.
5. The method for predicting animal's motor behavior according to claim 4,
the first moving speed is more than 0m/s and less than 3m/s, the second moving speed is more than or equal to 3m/s and less than 5m/s, the third moving speed is more than or equal to 5m/s and less than 10m/s, and the fourth moving speed is more than or equal to 10m/s and less than 20m/s.
6. The method for predicting animal's motor behavior according to claim 3,
the initial moving direction in the motion behavior feature library in the step S1 is a direction relative to a horizontal plane, and is represented as-90 degrees to 90 degrees, and the initial moving direction has a multi-stage moving direction range, including a first moving direction, a second moving direction, a third moving direction, and a fourth moving direction;
the step of obtaining the initial moving direction matched with the moving direction of the animal movement gravity center in the movement behavior feature library in the step S3 is as follows:
s321: comparing the moving direction of the center of gravity of the animal motion with a first moving direction, a second moving direction, a third moving direction and a fourth moving direction, and taking the moving direction range with the minimum absolute value of the moving direction difference of the center of gravity of the animal motion as the moving direction to be compared;
s322: calculating absolute values L1, L2, L3, ln of differences between the moving direction of the center of gravity of the animal motion and a plurality of initial moving directions in the moving direction to be compared, and then calculating a minimum value Lmin;
s323: and calculating the initial moving direction of the Lmin as the initial moving direction matched with the moving direction of the center of gravity of the animal motion.
7. The method for predicting animal's motor behavior according to claim 3,
the motion behavior prediction information base corresponds to the motion behavior feature base one by one through the matching module, the actual animal motion track corresponding to the animal motion track in the motion behavior prediction information base is recorded at intervals of fixed time, and the matching module and the animal motion track in the motion behavior prediction information base corresponding to the matching module are updated through the actual animal motion track.
8. The method of claim 7, wherein the step of predicting the motor behavior of the animal,
the fixed time is 7-15 days.
9. An animal athletic performance prediction system, for implementing an animal athletic performance prediction method as claimed in any one of claims 1-8.
10. A camera system based on animal motion behavior prediction is used for realizing the animal motion behavior prediction method of any one of claims 1 to 8, and is characterized by comprising,
an imaging module;
a processing module;
a storage module;
a communication module;
the imaging module is used for acquiring a video image;
the storage module is used for storing the video image acquired by the imaging module;
the communication module is used for receiving and sending signals;
the processing module is used for adjusting the lens direction or/and the focal length of the imaging module according to the instruction signal;
the instruction signal comprises lens direction adjustment information or/and focal length adjustment information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211270218.XA CN115345901B (en) | 2022-10-18 | 2022-10-18 | Animal motion behavior prediction method and system and camera system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211270218.XA CN115345901B (en) | 2022-10-18 | 2022-10-18 | Animal motion behavior prediction method and system and camera system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115345901A CN115345901A (en) | 2022-11-15 |
| CN115345901B true CN115345901B (en) | 2023-01-10 |
Family
ID=83957320
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211270218.XA Active CN115345901B (en) | 2022-10-18 | 2022-10-18 | Animal motion behavior prediction method and system and camera system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115345901B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115334249B (en) * | 2022-10-17 | 2023-03-10 | 成都唐米科技有限公司 | Focus following method based on motion center, storage medium and camera system |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104872025A (en) * | 2015-06-08 | 2015-09-02 | 广东省农业科学院畜牧研究所 | Aquarium fish activity track record method |
| CN105264436A (en) * | 2013-04-05 | 2016-01-20 | 安德拉运动技术股份有限公司 | System and method for controlling equipment related to image capture |
| CN113923364A (en) * | 2021-11-12 | 2022-01-11 | 成都唐米科技有限公司 | Camera video recording method and camera |
| CN215729043U (en) * | 2021-04-30 | 2022-02-01 | 深圳市大疆创新科技有限公司 | Follow focus ware, shooting device and movable platform |
| CN115334249A (en) * | 2022-10-17 | 2022-11-11 | 成都唐米科技有限公司 | Focus following method and system based on motion gravity center and camera system |
| CN115348392A (en) * | 2022-10-19 | 2022-11-15 | 成都唐米科技有限公司 | Shooting method and system based on template material |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP5947601B2 (en) * | 2012-04-10 | 2016-07-06 | キヤノン株式会社 | FOCUS DETECTION DEVICE, ITS CONTROL METHOD, AND IMAGING DEVICE |
| CN114928721B (en) * | 2022-05-20 | 2023-03-24 | 沈阳体育学院 | Switching method and system of football shooting lens, computer equipment and storage medium |
-
2022
- 2022-10-18 CN CN202211270218.XA patent/CN115345901B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105264436A (en) * | 2013-04-05 | 2016-01-20 | 安德拉运动技术股份有限公司 | System and method for controlling equipment related to image capture |
| CN104872025A (en) * | 2015-06-08 | 2015-09-02 | 广东省农业科学院畜牧研究所 | Aquarium fish activity track record method |
| CN215729043U (en) * | 2021-04-30 | 2022-02-01 | 深圳市大疆创新科技有限公司 | Follow focus ware, shooting device and movable platform |
| CN113923364A (en) * | 2021-11-12 | 2022-01-11 | 成都唐米科技有限公司 | Camera video recording method and camera |
| CN115334249A (en) * | 2022-10-17 | 2022-11-11 | 成都唐米科技有限公司 | Focus following method and system based on motion gravity center and camera system |
| CN115348392A (en) * | 2022-10-19 | 2022-11-15 | 成都唐米科技有限公司 | Shooting method and system based on template material |
Non-Patent Citations (2)
| Title |
|---|
| Data-driven child behavior prediction system based on posture database for fall accident prevention in a daily living space;Tsubasa Nose等;《Journal of Ambient Intelligence and Humanized Computing》;20200518;5845-5855 * |
| 基于时空图注意力网络的跳高轨迹预测;钟亦强等;《数学的实践与认识》;20220426;第52卷(第05期);第64-73页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115345901A (en) | 2022-11-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115348392B (en) | Shooting method and system based on template material | |
| JP7028966B2 (en) | Modular Hierarchical Visual System for Autonomous Personal Companion | |
| CN110210417B (en) | Method, terminal and readable storage medium for predicting pedestrian motion trail | |
| KR101121763B1 (en) | Apparatus and method for recognizing environment | |
| CN115345901B (en) | Animal motion behavior prediction method and system and camera system | |
| EP1672456B1 (en) | Target object detection apparatus and robot provided with the same | |
| KR20210015211A (en) | Method of cloud slam in realtime and robot and cloud server implementing thereof | |
| US8583286B2 (en) | Hybrid control device | |
| US7369686B2 (en) | Robot apparatus, face recognition method, and face recognition apparatus | |
| CN115334249B (en) | Focus following method based on motion center, storage medium and camera system | |
| US20210258470A1 (en) | Mobile body and control method | |
| JP2003269917A (en) | Method of measuring distance, instrument therefor, program therefor, recording medium therefor, and distance measuring instrument mounting type robot apparatus | |
| CN110072064B (en) | Image shooting method, device, equipment and storage medium | |
| CN111474953A (en) | Multi-dynamic-view-angle-coordinated aerial target identification method and system | |
| KR20220002758A (en) | Method for estimating position by synchronizing multi-sensor and robot implementing the same | |
| CN112019757A (en) | Unmanned aerial vehicle collaborative shooting method and device, computer equipment and storage medium | |
| CN111862154A (en) | Robot vision tracking method and device, robot and storage medium | |
| CN111161334A (en) | Semantic map construction method based on deep learning | |
| CN115565130A (en) | Unattended system and monitoring method based on optical flow | |
| CN206726243U (en) | Indoor moving video tracking positioning auxiliary filming apparatus | |
| CN115457666A (en) | Method and system for identifying moving gravity center of living body object and computer readable storage medium | |
| Zheng et al. | A study of 3D feature tracking and localization using a stereo vision system | |
| WO2020026798A1 (en) | Control device, control method, and program | |
| US20200382695A1 (en) | Robot and electronic device for acquiring video, and method for acquiring video using robot | |
| CN115294508B (en) | Focus following method and system based on static space three-dimensional reconstruction and camera system |
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 |