A system and method for quantifying and simulating the movement behavior of quadrupeds based on image recognition

Abstract

The present invention discloses a system and method for quantifying and simulating the motion behavior of a quadruped based on image recognition, relates to the field of biological motion, and solves the problem of poor quantification effect of the motion behavior of a quadruped. The data acquisition module is used to acquire motion gait simulation test data, the image analysis module is used to perform image analysis on the motion gait simulation test data to obtain gait type data of the target animal image, and the motion quantification module is used to perform motion quantification on the target animal according to the motion gait simulation test data and the gait type data of the target animal image. The present invention acquires the motion distance of the quadruped in the image by acquiring the gait cycle moving distance corresponding to different types of motion gaits, thereby improving the pertinence of the corresponding quantification result of the quadruped.

Description

Technical Field

[0001] The invention belongs to the field of biological motion and relates to image recognition technology, in particular to a system and method for quantifying and simulating the motion behavior of quadrupeds based on image recognition. Background Art

[0002] The movement behavior of quadrupeds refers to the behavior of animals with four limbs when moving, running, jumping, etc. Quadrupeds usually use their limbs to support their bodies and achieve body stability and balance through the coordinated movement of their forelimbs and hind limbs. Specific movement behaviors include walking, running, jumping, crawling, etc. By studying the movement behavior of quadrupeds, we can gain a deep understanding of their physiological characteristics, movement mechanisms, and adaptability, which is of great significance to the research in the fields of animal behavior and exercise physiology.

[0003] Existing methods for quantifying the movement behavior of quadrupeds based on image recognition have the following defects:

[0004] 1. When evaluating the motion of quadrupeds, the scene in the image is usually used as a reference object to evaluate the motion distance, which is difficult to quantify;

[0005] 2. When evaluating the movement of quadrupeds, the changes in the gait of quadrupeds during movement are fully considered. When evaluating the movement distance based on the gait, there is a large error;

[0006] To this end, we propose a system and method for quantitative simulation of quadruped motion behavior based on image recognition. Summary of the invention

[0007] In view of the shortcomings of the prior art, the purpose of the present invention is to provide a system and method for quantitative simulation of quadruped movement behavior based on image recognition. The present invention obtains movement images of target animals, the moving distance of the first gait cycle and the moving distance of the second gait cycle, obtains movement gait simulation test data, analyzes the movement images of the target animals in the movement gait simulation test data, obtains the quantity values ​​of the first type of gait cycles and the quantity values ​​of the second type of gait cycles, obtains gait type data of the target animal images, and quantifies the movement of the target animal according to the movement gait simulation test data and the gait type data of the target animal images.

[0008] In order to achieve the above-mentioned purpose, the present invention adopts the following technical scheme to implement a system and method for quantifying the motion behavior of quadrupeds based on image recognition. The specific working process of each module is as follows:

[0009] Data acquisition module: used for dividing the motion gait of the quadruped into a first type of motion gait and a second type of motion gait, analyzing the first type of motion gait and the second type of motion gait respectively, obtaining the first gait cycle moving distance and the second gait cycle moving distance, obtaining the target animal motion image, and defining the target animal motion image, the first gait cycle moving distance and the second gait cycle moving distance as motion gait simulation test data;

[0010] Image analysis module: used to analyze the target animal motion image in the motion gait simulation test data, obtain the first type gait cycle quantity value and the second type gait cycle quantity value, and obtain the target animal image gait type data;

[0011] Motion quantification module: used to quantify the motion of the target animal based on the motion gait simulation test data and the gait type data of the target animal image.

[0012] Furthermore, the data acquisition module acquires the motion gait simulation test data, specifically as follows:

[0013] Select a quadrupeds of the same type as sample quadrupeds, and mark them as the first to the ath sample quadrupeds respectively;

[0014] Obtain the average length of the limbs of the sample animals;

[0015] The movement gaits of quadrupeds are divided into the first type of movement gaits and the second type of movement gaits;

[0016] Performing gait analysis on the first type of motion gait to obtain the first step cycle moving distance;

[0017] Repeat the analysis process of the first type of motion gait to obtain the gait cycle movement distance corresponding to the second type of motion gait, and obtain the second gait cycle movement distance;

[0018] Acquire a motion video image corresponding to a target quadruped animal to obtain a motion image of the target animal;

[0019] The target animal's motion image, the moving distance in the first gait cycle, and the moving distance in the second gait cycle are defined as motion gait simulation test data.

[0020] Furthermore, the data acquisition module acquires the average length of the limbs of the sample animals, as follows:

[0021] Acquire the limb lengths of the first sample quadruped animal by using a first data acquisition device to obtain a left forelimb length value, a right forelimb length value, a left hind limb length value, and a right hind limb length value, and calculate the average of the left forelimb length value, the right forelimb length value, the left hind limb length value, and the right hind limb length value to obtain the first sample limb length value;

[0022] The average number of limbs of each sample quadruped is obtained to obtain the length values ​​of the first to a limbs, and the average length of the limbs of the sample animal is obtained by calculating the length values ​​of the first to a limbs.

[0023] Furthermore, the data acquisition module acquires the first step periodic movement distance as follows:

[0024] Create a gait analysis runway, mark a first marking line and a second marking line at both ends of the gait analysis runway respectively, obtain a distance value between the first marking line and the second marking line, and obtain a gait test distance value;

[0025] Acquire a motion video of a first sample quadruped moving from a first marked line to a second marked line through a camera to obtain a first sample gait analysis video;

[0026] Obtain the cumulative number of gait cycles completed by the first sample quadruped in the first sample gait analysis video to obtain the first gait cycle number value;

[0027] The number of gait cycles of each sample quadruped is obtained respectively to obtain the number of gait cycles from the second to the ath;

[0028] The number of gait cycles from the first step to the ath step are averaged to obtain the average number of sample gait cycles;

[0029] The first gait cycle moving distance is obtained by calculating the gait test distance value, the average number of sample gait cycles and the average length of the limbs of the sample animals;

[0030] Calculate the moving distance of the first step cycle.

[0031] Furthermore, the image analysis module acquires gait type data of the target animal image, as follows:

[0032] Acquire motion gait simulation test data, and acquire a motion image of a target animal according to the motion gait simulation test data;

[0033] Respectively obtain a plurality of gait cycles of the target animal in the target animal motion image, and mark them as the first to bth target gait cycles respectively;

[0034] Performing image analysis on the first target gait cycle to obtain a gait type classification coefficient corresponding to the first target gait cycle;

[0035] Repeat the above steps for the gait type division coefficient corresponding to the first target gait cycle, and respectively obtain the gait type division coefficients corresponding to the second to b-th target gait cycles, to obtain a plurality of gait type division coefficients;

[0036] Obtaining a gait type division coefficient threshold, comparing the gait type division coefficient threshold with a plurality of gait type division coefficients respectively, dividing the gait cycle into a first type gait and a second type gait, and obtaining gait type division data of the target animal;

[0037] Counting the number of gait cycles corresponding to the first type of gait in the target animal motion image to obtain a value of the number of the first type of gait cycles;

[0038] Counting the number of gait cycles corresponding to the second type of gait in the target animal motion image to obtain a value of the number of the second type of gait cycles;

[0039] The first type gait cycle quantity value and the second type gait cycle quantity value are defined as the target animal image gait type data.

[0040] Furthermore, the image analysis module divides the gait type corresponding to the first target gait cycle into coefficients as follows:

[0041] In the first target gait cycle, obtain the time value corresponding to the landing of the left forelimb of the target animal, obtain the time value of the landing of the first limb, obtain the time value corresponding to the landing of the right forelimb of the target animal, obtain the time value of the landing of the second limb, obtain the time value corresponding to the landing of the left hind limb of the target animal, obtain the time value of the landing of the third limb, obtain the time value corresponding to the landing of the right hind limb of the target animal, and obtain the time value of the landing of the fourth limb;

[0042] The gait type division coefficient corresponding to the first target gait cycle is obtained by calculating the first limb landing time value, the second limb landing time value, the third limb landing time value and the fourth limb landing time value;

[0043] The gait type division coefficient corresponding to the first target gait cycle is calculated.

[0044] Furthermore, the image analysis module obtains the gait type classification coefficient threshold, as follows:

[0045] The forelimb landing benchmark time difference and the hindlimb landing benchmark time difference are obtained respectively, and the forelimb landing benchmark time difference and the hindlimb landing benchmark time difference are calculated to obtain the gait type classification coefficient threshold;

[0046] Calculate the gait type classification coefficient threshold.

[0047] Furthermore, the image analysis module compares the gait type classification coefficient threshold with the multiple gait type classification coefficients respectively, as follows:

[0048] When the gait type division coefficient is greater than or equal to the gait type division coefficient threshold, the target animal is judged to be of the first type of gait;

[0049] When the gait type classification coefficient is less than the gait type classification coefficient threshold, it is determined that the target animal has the second type of gait.

[0050] Furthermore, the motion quantification module performs motion quantification on the target animal, specifically as follows:

[0051] Acquire motion gait simulation test data, and acquire the first gait cycle moving distance and the second gait cycle moving distance according to the motion gait simulation test data;

[0052] Acquire gait type data of the target animal image, and acquire a first type gait cycle quantity value and a second type gait cycle quantity value according to the gait type data of the target animal image;

[0053] The moving distance of the first gait cycle, the moving distance of the second gait cycle, the number of first-type gait cycles and the number of second-type gait cycles are calculated to obtain the moving distance of the target animal in the target animal's moving image, thereby quantifying the movement behavior of the reptile;

[0054] The moving distance of the target animal in the target animal moving image is calculated.

[0055] A method for quantitatively simulating the movement behavior of a quadruped based on image recognition comprises the following specific steps:

[0056] Step S1: Acquire motion gait simulation test data;

[0057] Step S2: performing image analysis on the motion gait simulation test data to obtain gait type data of the target animal image;

[0058] Step S3: quantifying the movement of the target animal according to the movement gait simulation test data and the gait type data of the target animal image.

[0059] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0060] 1. The present invention divides the movement gait of the quadruped into a first type of movement gait and a second type of movement gait, and respectively obtains the gait cycle movement distance corresponding to different types of movement gaits to obtain the movement distance of the quadruped in the image, thereby improving the pertinence of the corresponding quantification result of the quadruped;

[0061] 2. The present invention collects multiple samples to evaluate the gait cycle movement distance, which can effectively avoid the randomness of the evaluation result caused by a single sample, thereby improving the accuracy of the motion quantification result. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0063] Figure 1 is the overall system block diagram of the present invention;

[0064] Figure 2 It is a diagram of the implementation steps of the present invention;

[0065] Figure 3 Schematic diagram of the radial gait of a quadruped. DETAILED DESCRIPTION

[0066] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0067] Embodiment 1

[0068] See also Figure 1 , the present invention provides a technical solution: a quadrupedal motion behavior quantification simulation system based on image recognition, comprising a data acquisition module, an image analysis module, a motion quantification module and a server, wherein the data acquisition module, the image analysis module and the motion quantification module are respectively connected to the server;

[0069] The data acquisition module acquires motion gait simulation test data;

[0070] Select a quadrupeds of the same type as sample quadrupeds, and mark them as the first to the ath sample quadrupeds respectively;

[0071] It should be noted here that:

[0072] In this application, a mentioned here specifically refers to the quantity value corresponding to the sample quadruped, and a is an integer greater than 0;

[0073] Acquire the limb lengths of the first sample quadruped animal by using a first data acquisition device to obtain a left forelimb length value, a right forelimb length value, a left hind limb length value, and a right hind limb length value, and calculate the average of the left forelimb length value, the right forelimb length value, the left hind limb length value, and the right hind limb length value to obtain the first sample limb length value;

[0074] The average number of limbs of each sample quadruped is obtained respectively to obtain the length values ​​of the first to a limbs, and the average length of the limbs of the sample animal is obtained by calculating the length values ​​of the first to a limbs;

[0075] The average length of the limbs of the sample animals is calculated, and the specific formula is configured as follows:

[0076]

[0077] Among them, Yzp is the average length of the limbs of the sample animals, Ys1 to Ysa are the length values ​​of the first to the ath limbs, and a is the corresponding number value of the sample quadrupeds;

[0078] The movement gaits of quadrupeds are divided into the first type of movement gaits and the second type of movement gaits;

[0079] It should be noted here that:

[0080] In the present application, the quadrupedal animal involved may be a pig;

[0081] See also Figure 3 The first type of gait corresponding to quadrupeds is the radial gait, in which the four limbs of the quadruped move in a front-to-back symmetrical manner, that is, the front limbs and hind limbs move alternately, forming a stable and rhythmic step, see Figure 3 ;

[0082] The second type of gait corresponding to quadrupeds is the gallop gait. In this gait, the four limbs of the quadruped move in a symmetrical manner, with the front and rear limbs landing on the ground and in the air at the same time, forming a fast and powerful running movement.

[0083] The gait analysis of the first type of motion gait is performed as follows:

[0084] Create a gait analysis runway, mark a first marking line and a second marking line at both ends of the gait analysis runway respectively, obtain a distance value between the first marking line and the second marking line, and obtain a gait test distance value;

[0085] Acquire a motion video of a first sample quadruped moving from a first marked line to a second marked line through a camera to obtain a first sample gait analysis video;

[0086] Obtain the cumulative number of gait cycles completed by the first sample quadruped in the first sample gait analysis video to obtain the first gait cycle number value;

[0087] It should be noted here that:

[0088] In this application, the gait cycle involved here specifically refers to the four limbs of the sample quadruped completing one gait movement;

[0089] The number of gait cycles of each sample quadruped is obtained respectively to obtain the number of gait cycles from the second to the ath;

[0090] The number of gait cycles from the first step to the ath step are averaged to obtain the average number of sample gait cycles;

[0091] The first gait cycle moving distance is obtained by calculating the gait test distance value, the average number of sample gait cycles and the average length of the limbs of the sample animals;

[0092] Calculate the moving distance of the first step cycle, and the specific formula configuration is as follows:

[0093]

[0094] Among them, Btq1 is the moving distance of the first gait cycle, Bcj is the gait test distance value, Bzs is the average number of sample gait cycles, Ybc is the average length of the limbs of the sample animals, a1 is the set proportional coefficient and a1 is greater than 0;

[0095] Repeat the analysis process of the first type of motion gait to obtain the gait cycle movement distance corresponding to the second type of motion gait, and obtain the second gait cycle movement distance;

[0096] Acquire a motion video image corresponding to a target quadruped animal to obtain a motion image of the target animal;

[0097] The target animal's motion image, the moving distance of the first gait cycle, and the moving distance of the second gait cycle are defined as motion gait simulation test data;

[0098] The data acquisition module acquires the motion gait simulation test data and transmits it to the image analysis module and the motion quantification module;

[0099] The image analysis module performs image analysis on the motion gait simulation test data to obtain the gait type data of the target animal image;

[0100] Acquire motion gait simulation test data, and acquire a motion image of a target animal according to the motion gait simulation test data;

[0101] Respectively obtain a plurality of gait cycles of the target animal in the target animal motion image, and mark them as the first to bth target gait cycles respectively;

[0102] It should be noted here that:

[0103] In the present application, b is the quantity value corresponding to the target gait cycle existing in the target animal motion image;

[0104] Performing image analysis on the first target gait cycle to obtain a gait type classification coefficient corresponding to the first target gait cycle;

[0105] The details are as follows:

[0106] In the first target gait cycle, obtain the time value corresponding to the landing of the left forelimb of the target animal, obtain the time value of the landing of the first limb, obtain the time value corresponding to the landing of the right forelimb of the target animal, obtain the time value of the landing of the second limb, obtain the time value corresponding to the landing of the left hind limb of the target animal, obtain the time value of the landing of the third limb, obtain the time value corresponding to the landing of the right hind limb of the target animal, and obtain the time value of the landing of the fourth limb;

[0107] The gait type division coefficient corresponding to the first target gait cycle is obtained by calculating the first limb landing time value, the second limb landing time value, the third limb landing time value and the fourth limb landing time value;

[0108] The gait type division coefficient corresponding to the first target gait cycle is calculated, and the specific formula configuration is as follows:

[0109] Btl1=|Tzl1-Tzl2|+|Tzl3-Tzl4|;

[0110] Wherein, Bt l 1 is the gait type division coefficient corresponding to the first target gait cycle, Tz l 1 is the value of the first limb landing time, Tz l 2 is the value of the second limb landing time, Tz l 3 is the value of the third limb landing time, and Tz l4 is the value of the fourth limb landing time;

[0111] Repeat the above steps for the gait type division coefficient corresponding to the first target gait cycle, and respectively obtain the gait type division coefficients corresponding to the second to b-th target gait cycles, to obtain a plurality of gait type division coefficients;

[0112] Obtaining a gait type division coefficient threshold, and numerically comparing the gait type division coefficient threshold with a plurality of gait type division coefficients to obtain gait type division data of the target animal;

[0113] The gait type classification coefficient threshold is obtained as follows:

[0114] The forelimb landing benchmark time difference and the hindlimb landing benchmark time difference are obtained respectively, and the forelimb landing benchmark time difference and the hindlimb landing benchmark time difference are calculated to obtain the gait type classification coefficient threshold;

[0115] The gait type classification coefficient threshold is calculated, and the formula is as follows:

[0116] Blty=Qsc+Hsc;

[0117] Among them, B l ty is the gait type classification coefficient threshold, Qsc is the reference time difference of forelimb landing, and Hsc is the reference time difference of hindlimb landing;

[0118] It should be noted here that:

[0119] In the present application, the reference time difference of forelimb landing is specifically the critical value between the time difference of the left forelimb and the right forelimb contacting the ground when the quadruped is walking and the time difference of the left forelimb and the right forelimb contacting the ground when the quadruped is running; the reference time difference of hind limb landing is specifically the critical value between the time difference of the left hind limb and the right hind limb contacting the ground when the quadruped is walking and the time difference of the left hind limb and the right hind limb contacting the ground when the quadruped is running;

[0120] The gait type division coefficient threshold is numerically compared with multiple gait type division coefficients, as follows:

[0121] When the gait type division coefficient is greater than or equal to the gait type division coefficient threshold, the target animal is judged to be of the first type of gait;

[0122] When the gait type division coefficient is less than the gait type division coefficient threshold, the target animal is judged to have the second type of gait;

[0123] Counting the number of gait cycles corresponding to the first type of gait in the target animal motion image to obtain a value of the number of the first type of gait cycles;

[0124] Counting the number of gait cycles corresponding to the second type of gait in the target animal motion image to obtain a value of the number of the second type of gait cycles;

[0125] The first type of gait cycle quantity value and the second type of gait cycle quantity value are defined as target animal image gait type data;

[0126] The image analysis module acquires the gait type data of the target animal image and transmits it to the motion quantification module;

[0127] The motion quantification module performs motion quantification on the target animal according to the motion gait simulation test data and the gait type data of the target animal image;

[0128] Acquire motion gait simulation test data, and acquire the first gait cycle moving distance and the second gait cycle moving distance according to the motion gait simulation test data;

[0129] Acquire gait type data of the target animal image, and acquire a first type gait cycle quantity value and a second type gait cycle quantity value according to the gait type data of the target animal image;

[0130] The moving distance of the first gait cycle, the moving distance of the second gait cycle, the number of first-type gait cycles and the number of second-type gait cycles are calculated to obtain the moving distance of the target animal in the target animal's moving image, thereby quantifying the movement behavior of the reptile;

[0131] The moving distance of the target animal in the target animal moving image is calculated, and the specific formula configuration is as follows:

[0132] Smd = Btq1*Blx1+Btq2*Blx2;

[0133] Wherein, Smd is the movement distance of the target animal in the target animal motion image, Btq1 is the movement distance of the first gait cycle, Btq2 is the movement distance of the second gait cycle, B lx1 is the number value of the first type of gait cycle, and B lx2 is the number value of the second type of gait cycle;

[0134] In this application, if corresponding calculation formulas appear, the above calculation formulas are all dimensionless and take their numerical calculations. The weight coefficients, proportional coefficients and other coefficients in the formulas are set to a result value obtained by quantifying each parameter. The size of the weight coefficient and the proportional coefficient can be determined as long as it does not affect the proportional relationship between the parameter and the result value.

[0135] Embodiment 2

[0136] See also Figure 2 Based on another concept of the same invention, a system and method for quantifying and simulating the motion behavior of quadrupeds based on image recognition is proposed, which comprises the following steps:

[0137] Step S1: Acquire motion gait simulation test data;

[0138] Step S11: selecting a quadrupeds of the same type as sample quadrupeds, and marking them as the first to ath sample quadrupeds, respectively;

[0139] Step S12: acquiring the limb lengths of the first sample quadruped animal through the first data acquisition device to obtain the left forelimb length value, the right forelimb length value, the left hind limb length value and the right hind limb length value, and calculating the average of the left forelimb length value, the right forelimb length value, the left hind limb length value and the right hind limb length value to obtain the first sample limb length value;

[0140] Step S13: Obtain the average number of limbs for each sample quadruped animal respectively to obtain the first to ath limb length values, and calculate the first to ath limb length values ​​to obtain the average length of the limbs of the sample animal;

[0141] Step S14: classifying the movement gait of the quadruped into a first type of movement gait and a second type of movement gait;

[0142] Step S15: performing gait analysis on the first type of motion gait to obtain a first gait cycle moving distance;

[0143] The step S15 further includes the following specific steps:

[0144] Step S151: creating a gait analysis track, marking a first marking line and a second marking line at both ends of the gait analysis track respectively, obtaining a distance value between the first marking line and the second marking line, and obtaining a gait test distance value;

[0145] Step S152: acquiring a motion video of a first sample quadruped moving from a first marked line to a second marked line, and obtaining a first sample gait analysis video;

[0146] Step S153: obtaining the cumulative number of gait cycles completed by the first sample quadruped in the first sample gait analysis video, and obtaining the first gait cycle number value;

[0147] Step S154: acquiring the number of gait cycles of each sample quadruped respectively, and obtaining the second to ath gait cycle number values;

[0148] Step S155: averaging the first gait cycle value to the ath gait cycle value to obtain the average gait cycle value of the sample;

[0149] Step S156: Calculate the gait test distance value, the average number of sample gait cycles, and the average length of the limbs of the sample animals to obtain the first gait cycle moving distance;

[0150] Calculate the moving distance of the first step cycle, and the specific formula configuration is as follows:

[0151]

[0152] Among them, Btq1 is the moving distance of the first gait cycle, Bcj is the gait test distance value, Bzs is the average number of sample gait cycles, Ybc is the average length of the limbs of the sample animals, a1 is the set proportional coefficient and a1 is greater than 0;

[0153] Step S16: Repeat the analysis process of the first type of motion gait to obtain the gait cycle movement distance corresponding to the second type of motion gait, and obtain the second gait cycle movement distance;

[0154] Step S17: acquiring a motion video image corresponding to the target quadruped animal to obtain a motion image of the target animal;

[0155] Step S18: defining the target animal's motion image, the first gait cycle moving distance, and the second gait cycle moving distance as motion gait simulation test data;

[0156] Step S2: performing image analysis on the motion gait simulation test data to obtain gait type data of the target animal image;

[0157] Step S21: acquiring motion gait simulation test data, and acquiring a motion image of a target animal according to the motion gait simulation test data;

[0158] Step S22: obtaining a plurality of gait cycles of the target animal in the target animal motion image respectively, and marking them as the first to bth target gait cycles respectively;

[0159] Step S23: performing image analysis on the first target gait cycle to obtain a gait type classification coefficient corresponding to the first target gait cycle;

[0160] The step S23 further includes the following specific steps:

[0161] Step S231: in the first target gait cycle, obtain the time value corresponding to the landing of the left forelimb of the target animal, obtain the time value corresponding to the landing of the first limb, obtain the time value corresponding to the landing of the right forelimb of the target animal, obtain the time value corresponding to the landing of the second limb, obtain the time value corresponding to the landing of the left hind limb of the target animal, obtain the time value corresponding to the landing of the third limb, obtain the time value corresponding to the landing of the right hind limb of the target animal, and obtain the time value corresponding to the landing of the fourth limb;

[0162] Step S232: obtaining a gait type division coefficient corresponding to the first target gait cycle by calculating the first limb landing time value, the second limb landing time value, the third limb landing time value, and the fourth limb landing time value;

[0163] The gait type division coefficient corresponding to the first target gait cycle is calculated, and the specific formula configuration is as follows:

[0164] Btl1=|Tzl1-Tzl2|+|Tzl3-Tzl4|;

[0165] Wherein, Bt l 1 is the gait type division coefficient corresponding to the first target gait cycle, Tz l 1 is the value of the first limb landing time, Tz l 2 is the value of the second limb landing time, Tz l 3 is the value of the third limb landing time, and Tz l4 is the value of the fourth limb landing time;

[0166] Step S24: Repeat the above steps for the gait type division coefficient corresponding to the first target gait cycle, and respectively obtain the gait type division coefficients corresponding to the second to b-th target gait cycles, to obtain a plurality of gait type division coefficients;

[0167] Step S25: obtaining a gait type division coefficient threshold, and numerically comparing the gait type division coefficient threshold with a plurality of gait type division coefficients to obtain gait type division data of the target animal;

[0168] The step S25 further includes the following specific steps:

[0169] Step S251: respectively obtaining the forelimb landing reference time difference and the hindlimb landing reference time difference, and calculating the forelimb landing reference time difference and the hindlimb landing reference time difference to obtain a gait type classification coefficient threshold;

[0170] The gait type classification coefficient threshold is calculated, and the formula is as follows:

[0171] Blty=Qsc+Hsc;

[0172] Among them, B l ty is the gait type classification coefficient threshold, Qsc is the reference time difference of forelimb landing, and Hsc is the reference time difference of hindlimb landing;

[0173] Step S252: when the gait type classification coefficient is greater than or equal to the gait type classification coefficient threshold, it is determined that the target animal has a first type of gait;

[0174] Step S253: when the gait type classification coefficient is less than the gait type classification coefficient threshold, it is determined that the target animal has a second type of gait;

[0175] Step S26: Counting the number of gait cycles corresponding to the first type of gait in the target animal's motion image to obtain a value of the number of first type of gait cycles;

[0176] Step S26: Counting the number of gait cycles corresponding to the second type of gait in the target animal's motion image to obtain a value of the number of second type of gait cycles;

[0177] Step S28: defining the first type gait cycle quantity value and the second type gait cycle quantity value as the target animal image gait type data;

[0178] Step S3: quantifying the movement of the target animal according to the movement gait simulation test data and the gait type data of the target animal image;

[0179] Step S31: acquiring motion gait simulation test data, and acquiring the first gait cycle moving distance and the second gait cycle moving distance according to the motion gait simulation test data;

[0180] Step S32: acquiring gait type data of the target animal image, and acquiring a first type gait cycle quantity value and a second type gait cycle quantity value according to the gait type data of the target animal image;

[0181] Step S33: calculating the first gait cycle moving distance, the second gait cycle moving distance, the first type gait cycle quantity value and the second type gait cycle quantity value to obtain the moving distance of the target animal in the target animal moving image, so as to quantify the movement behavior of the reptile;

[0182] The moving distance of the target animal in the target animal moving image is calculated, and the specific formula configuration is as follows:

[0183] Smd = Btq1*Blx1+Btq2*Blx2;

[0184] Wherein, Smd is the movement distance of the target animal in the target animal motion image, Btq1 is the movement distance of the first gait cycle, Btq2 is the movement distance of the second gait cycle, B lx1 is the number value of the first type of gait cycle, and B lx2 is the number value of the second type of gait cycle.

[0185] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A system for quantifying and simulating the movement behavior of quadrupeds based on image recognition, characterized in that: include: Data acquisition module: used for acquiring the motion gait of the quadruped and dividing it into a first type of motion gait and a second type of motion gait, respectively analyzing the first type of motion gait and the second type of motion gait, obtaining the first gait cycle moving distance and the second gait cycle moving distance, acquiring the motion image of the target animal, and defining the target animal motion image, the first gait cycle moving distance and the second gait cycle moving distance as motion gait simulation test data; Image analysis module: used to analyze the target animal motion image in the motion gait simulation test data, obtain the first type gait cycle quantity value and the second type gait cycle quantity value, and obtain the target animal image gait type data; Motion quantification module: used to quantify the motion of the target animal based on the motion gait simulation test data and the gait type data of the target animal image; The image analysis module acquires the gait type data of the target animal image, as follows: Acquire motion gait simulation test data, and acquire a motion image of a target animal according to the motion gait simulation test data; Respectively obtain a plurality of gait cycles of the target animal in the target animal motion image, and mark them as the first to bth target gait cycles respectively; Performing image analysis on the first target gait cycle to obtain a gait type classification coefficient corresponding to the first target gait cycle; Repeat the above steps for the gait type division coefficient corresponding to the first target gait cycle, and respectively obtain the gait type division coefficients corresponding to the second to b-th target gait cycles, to obtain a plurality of gait type division coefficients; Obtaining a gait type division coefficient threshold, comparing the gait type division coefficient threshold with a plurality of gait type division coefficients respectively, dividing the gait cycle into a first type gait and a second type gait, and obtaining gait type division data of the target animal; Counting the number of gait cycles corresponding to the first type of gait in the target animal motion image to obtain a value of the number of the first type of gait cycles; Counting the number of gait cycles corresponding to the second type of gait in the target animal motion image to obtain a value of the number of the second type of gait cycles; The first type of gait cycle quantity value and the second type of gait cycle quantity value are defined as target animal image gait type data; The image analysis module divides the gait type corresponding to the first target gait cycle into coefficients as follows: In the first target gait cycle, obtain the time value corresponding to the landing of the left forelimb of the target animal, obtain the time value of the landing of the first limb, obtain the time value corresponding to the landing of the right forelimb of the target animal, obtain the time value of the landing of the second limb, obtain the time value corresponding to the landing of the left hind limb of the target animal, obtain the time value of the landing of the third limb, obtain the time value corresponding to the landing of the right hind limb of the target animal, and obtain the time value of the landing of the fourth limb; The gait type division coefficient corresponding to the first target gait cycle is obtained by calculating the first limb landing time value, the second limb landing time value, the third limb landing time value and the fourth limb landing time value; Calculating a gait type division coefficient corresponding to a first target gait cycle; The image analysis module obtains the gait type classification coefficient threshold, which is as follows: The forelimb landing benchmark time difference and the hindlimb landing benchmark time difference are obtained respectively, and the forelimb landing benchmark time difference and the hindlimb landing benchmark time difference are calculated to obtain the gait type classification coefficient threshold; Calculating the gait type classification coefficient threshold; The image analysis module compares the gait type classification coefficient threshold with multiple gait type classification coefficients respectively, as follows: When the gait type division coefficient is greater than or equal to the gait type division coefficient threshold, the target animal is judged to be of the first type of gait; When the gait type division coefficient is less than the gait type division coefficient threshold, the target animal is judged to have the second type of gait; The motion quantification module performs motion quantification on the target animal, as follows: Acquire motion gait simulation test data, and acquire the first gait cycle moving distance and the second gait cycle moving distance according to the motion gait simulation test data; Acquire gait type data of the target animal image, and acquire a first type gait cycle quantity value and a second type gait cycle quantity value according to the gait type data of the target animal image; The moving distance of the first gait cycle, the moving distance of the second gait cycle, the number of first-type gait cycles and the number of second-type gait cycles are calculated to obtain the moving distance of the target animal in the target animal's moving image, thereby quantifying the movement behavior of the reptile; The moving distance of the target animal in the target animal moving image is calculated.

2. A quadrupedal motion behavior quantitative simulation system based on image recognition according to claim 1, characterized in that: The data acquisition module acquires the motion gait simulation test data, specifically as follows: Select a quadrupeds of the same type as sample quadrupeds, and mark them as the first to the ath sample quadrupeds respectively; Obtain the average length of the limbs of the sample animals; The movement gaits of quadrupeds are divided into the first type of movement gaits and the second type of movement gaits; Performing gait analysis on the first type of motion gait to obtain the first step cycle moving distance; Repeat the analysis process of the first type of motion gait to obtain the gait cycle movement distance corresponding to the second type of motion gait, and obtain the second gait cycle movement distance; Acquire a motion video image corresponding to a target quadruped animal to obtain a motion image of the target animal; The target animal's motion image, the moving distance in the first gait cycle, and the moving distance in the second gait cycle are defined as motion gait simulation test data.

3. A quadrupedal motion behavior quantification simulation system based on image recognition according to claim 2, characterized in that: The data acquisition module acquires the average length of the limbs of the sample animals, as follows: The limb lengths of the first sample quadrupeds are obtained to obtain a left forelimb length value, a right forelimb length value, a left hind limb length value, and a right hind limb length value, and the average of the left forelimb length value, the right forelimb length value, the left hind limb length value, and the right hind limb length value is calculated to obtain the limb length values ​​of the first sample; The average number of limbs of each sample quadruped is obtained to obtain the length values ​​of the first to a limbs, and the average length of the limbs of the sample animal is obtained by calculating the length values ​​of the first to a limbs.

4. A quadrupedal motion behavior quantification simulation system based on image recognition according to claim 2, characterized in that: The data acquisition module acquires the first phase period moving distance, specifically as follows: Mark a first marking line and a second marking line at both ends of the gait analysis track, respectively, obtain a distance value between the first marking line and the second marking line, and obtain a gait test distance value; Acquire a motion video of a first sample quadruped moving from a first marked line to a second marked line through a camera to obtain a first sample gait analysis video; Obtain the cumulative number of gait cycles completed by the first sample quadruped in the first sample gait analysis video to obtain the first gait cycle number value; The number of gait cycles of each sample quadruped is obtained respectively to obtain the number of gait cycles from the second to the ath; The number of gait cycles from the first step to the ath step are averaged to obtain the average number of sample gait cycles; The first gait cycle moving distance is obtained by calculating the gait test distance value, the average number of sample gait cycles and the average length of the limbs of the sample animals; Calculate the moving distance of the first step cycle.

5. A method for quantifying and simulating the motion behavior of a quadruped based on image recognition, applicable to a system for quantifying and simulating the motion behavior of a quadruped based on image recognition as claimed in any one of claims 1 to 4, characterized in that: The quantitative simulation method comprises the following specific steps: Step S1: Acquire motion gait simulation test data; Step S2: performing image analysis on the motion gait simulation test data to obtain gait type data of the target animal image; Step S3: quantifying the movement of the target animal according to the movement gait simulation test data and the gait type data of the target animal image.

Images