An image-based giant panda rapid identification system and identification method

Abstract

The application discloses a giant panda quick identification system and method based on images, first sets a first identification standard and a second identification standard, obtains an actual infrared temperature image and an environmental temperature parameter of an animal to be identified, and converts the actual infrared temperature image into an actual infrared pixel temperature image; then, first identification is performed according to the first identification standard and the actual infrared pixel temperature image, if the identification fails, the output result is other animals, and the identification program is terminated at the same time, otherwise, the barycentric coordinate parameter and the thermal area barycentric coordinate parameter of the animal to be identified are generated respectively, the thermal area barycentric actual offset is calculated according to the barycentric coordinate parameter and the thermal area barycentric coordinate parameter, and finally, whether the animal to be identified is a giant panda is judged according to the thermal area barycentric actual offset and the second identification standard; due to the special physiological characteristics of the giant panda, the thermal area barycenter of the giant panda will be offset downward relative to the barycenter; based on the above principle, the application performs twice identification on the animal to be identified, so that the identification accuracy is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of image recognition, in particular to a giant panda rapid identification system and method based on images. BACKGROUND

[0002] As a rare protected animal, the giant panda is known as the "national treasure" of China and is strictly protected. In order to improve the monitoring of wild giant pandas, a large number of cameras are currently set up in giant panda reserves to capture and capture the activities of wild giant pandas, further providing image basis materials for in-situ protection and scientific research of giant pandas. The above-mentioned equipment has good monitoring effect during the day, but at night, especially when taking pictures and identifying at a long distance, the color difference between different parts of the animal in the picture and between the animal and the background environment is small due to insufficient light, and the giant panda cannot be quickly identified by its typical black and white color, especially for bear animals with high similarity in body shape, the recognition accuracy is lower. SUMMARY

[0003] The main purpose of the present application is to provide a giant panda rapid identification system and method based on images, which aims to solve the low night giant panda identification accuracy defect in the prior art.

[0004] The present application achieves the above-mentioned purpose through the following technical solutions:

[0005] A giant panda rapid identification system based on images, comprising:

[0006] A parameter setting module for setting a first identification standard and a second identification standard for giant panda identification;

[0007] A data acquisition and processing module for acquiring an actual infrared temperature image of an animal to be identified and an environmental temperature parameter T1, and converting the actual infrared temperature image into an actual infrared pixel temperature image according to the environmental temperature parameter T1;

[0008] A first identification module for combining the first identification standard and the actual infrared pixel temperature image to identify the animal to be identified for the first time, and if the first identification fails, determining that the animal to be identified is another animal, terminating the identification program and outputting the identification result;

[0009] A first calculation module for generating a barycenter coordinate parameter and a hot area barycenter coordinate parameter of the animal to be identified according to the actual infrared pixel temperature image if the first identification is successful;

[0010] A second calculation module for calculating the actual offset of the hot area barycenter of the animal to be identified according to the barycenter coordinate parameter and the hot area barycenter coordinate parameter;

[0011] The second identification module is configured to determine whether the actual offset of the thermal center of gravity meets the second identification criterion, and if yes, output the to-be-identified animal as a panda, and if not, output the to-be-identified animal as another animal.

[0012] Correspondingly, the application further discloses an identification method based on the above rapid identification system, which comprises the following steps:

[0013] The first identification criterion and the second identification criterion for identifying a panda are set.

[0014] An actual infrared temperature image of a to-be-identified animal and an environmental temperature parameter T1 are obtained, and the actual infrared temperature image is converted into an actual infrared pixel temperature image according to the environmental temperature parameter T1.

[0015] The to-be-identified animal is identified for the first time in combination with the first identification criterion and the actual infrared pixel temperature image, and if the identification fails, the to-be-identified animal is determined as another animal, the identification procedure is terminated, and an identification result is output.

[0016] If the first-time identification succeeds, the center of gravity coordinate parameter and the thermal center of gravity coordinate parameter of the to-be-identified animal are respectively generated according to the actual infrared pixel temperature image.

[0017] The actual offset of the thermal center of gravity of the to-be-identified animal is calculated according to the center of gravity coordinate parameter and the thermal center of gravity coordinate parameter.

[0018] It is determined whether the actual offset of the thermal center of gravity meets the second identification criterion, and if yes, the to-be-identified animal is output as a panda, and if not, the to-be-identified animal is output as another animal.

[0019] Optionally, the expression of the first identification criterion is wherein T0 is a standard temperature threshold, T1 represents the environmental temperature parameter, t1-t0 is a constant, and t1 represents the actual infrared temperature of the to-be-identified animal. 14 The expression of the second identification criterion is wherein a represents the actual offset of the thermal center of gravity, b represents the actual offset of the center of gravity, and c represents the actual offset of the thermal center of gravity in the vertical direction. y

[0020] Optionally, obtaining the actual infrared temperature image of the to-be-identified animal and the environmental temperature parameter T1, and converting the actual infrared temperature image into the actual infrared pixel temperature image according to the environmental temperature parameter T1 comprises the following steps:

[0021] The environmental temperature parameter T1 is obtained.

[0022] ​​A temperature detection deviation is set, and a screening temperature domain [(1-P)T1, (1+P)T1] is generated according to the temperature detection deviation and an ambient temperature parameter, wherein P represents the temperature detection deviation;

[0023] The actual infrared temperature image is converted into a pixel image, and an actual temperature value of each pixel unit is obtained respectively;

[0024] The pixel units with the actual temperature value in the screening temperature domain are deleted from the pixel image, and an actual infrared pixel temperature image is generated.

[0025] Optionally, a first identification is performed on the animal to be identified in combination with the first identification standard and the actual infrared pixel temperature image, if the identification fails, the animal to be identified is determined as another animal, the identification procedure is terminated, and an identification result is output, including the following steps:

[0026] An actual infrared pixel temperature image is obtained;

[0027] An actual temperature value of each pixel unit is obtained respectively;

[0028] A standard temperature threshold T0 is selected from the first identification standard according to an ambient temperature parameter T1;

[0029] Each actual temperature value is compared with the standard temperature threshold respectively, if the standard temperature threshold is met, it is determined as qualified, and the actual temperature value is collected into a first set, otherwise, the actual temperature value is collected into a second set;

[0030] A number Q1 of actual temperature values in the first set and a number Q2 of actual temperature values in the second set are counted;

[0031] If Q1 / Q2>1 is met, then the animal to be identified is determined as a giant panda, the first identification is successful, otherwise, the animal to be identified is determined as another animal, the first identification is failed, and the identification procedure is terminated.

[0032] Optionally, a gravity center coordinate parameter and a hot area gravity center coordinate parameter of the animal to be identified are generated respectively according to the actual infrared pixel temperature image, including the following steps:

[0033] An actual infrared pixel temperature image is obtained;

[0034] A minimum bounding box is generated for the actual infrared pixel temperature image, and a height value H of the minimum bounding box is obtained;

[0035] A pixel coordinate set and a pixel temperature coordinate set are obtained according to the actual infrared pixel temperature image with the minimum bounding box;

[0036] The gravity center coordinate parameter of the animal to be identified is calculated according to the pixel coordinate set;

[0037] calculating the heat area barycentric coordinate parameter of the animal to be identified according to the pixel temperature coordinate set.

[0038] Optionally, the pixel coordinate set and the pixel temperature coordinate set are obtained according to the actual infrared pixel temperature image with the minimum bounding box, and the obtaining includes the following steps:

[0039] constructing a standard two-dimensional coordinate system according to the minimum bounding box, wherein the horizontal coordinate of the standard two-dimensional coordinate system represents the length of the minimum bounding box, and the vertical coordinate represents the height of the minimum bounding box;

[0040] generating a unique coordinate (x i , y i ) for each pixel unit according to the standard two-dimensional coordinate system, and collecting all the coordinates to generate a pixel coordinate set { (x1, y1), (x2, y2),..., (x i , y i ),..., (x N , y N )}, wherein i represents the pixel unit number, and N represents the total number of pixel units;

[0041] obtaining the actual temperature value of each pixel unit, binding each actual temperature value with each pixel unit number, and generating a pixel temperature coordinate set {T1, T2,..., T i ,..., T N}, wherein i represents the pixel unit number, and N represents the total number of pixel units.

[0042] Optionally, the barycentric coordinate parameter includes a barycentric vertical coordinate, and the calculation formula of the barycentric vertical coordinate is: , wherein N represents the total number of pixel units, i represents the pixel unit number, and y i represents the vertical coordinate of the pixel unit numbered i.

[0043] Optionally, the heat area barycentric coordinate parameter includes a heat area barycentric vertical coordinate, and the calculation formula of the heat area barycentric vertical coordinate is: , wherein N represents the total number of pixel units, i represents the pixel unit number, and y i represents the vertical coordinate of the pixel unit numbered i.

[0044] Optionally, the calculation formula of the heat area barycentric actual offset is: , wherein H represents the height value of the minimum bounding box.

[0045] Compared with the prior art, the present application has the following beneficial effects:

[0046] The application first sets a first identification standard and a second identification standard for giant panda identification, then acquires an actual infrared temperature image of an animal to be identified and an environmental temperature parameter T1, and converts the actual infrared temperature image into an actual infrared pixel temperature image according to the environmental temperature parameter T1; the first identification standard and the actual infrared pixel temperature image are combined to perform a first identification on the animal to be identified, if the identification fails, the result is directly output as other animals, and the identification program is terminated, otherwise, the actual infrared pixel temperature image is used to generate a gravity center coordinate parameter and a thermal area gravity center coordinate parameter of the animal to be identified, the thermal area gravity center actual offset is calculated according to the gravity center coordinate parameter and the thermal area gravity center coordinate parameter, and the thermal area gravity center actual offset and the second identification standard are used to determine whether the animal to be identified is a giant panda.

[0047] Research shows that the body surface temperature of a giant panda changes regularly with the change of the environmental temperature, and there is a large difference between the body surface temperature of the giant panda and that of some homeothermic animals, so the infrared temperature identification can be used to preliminarily screen the animal to be identified, and the problem of being unable to identify due to small color difference can be avoided.

[0048] Secondly, compared with mammals such as bears, the thickness of the back hair of a giant panda is 5-10 cm, the dense down forms a heat insulation layer to reduce heat loss, while the abdominal hair is relatively sparse, which is beneficial to the core heat transfer to the body surface to maintain the internal organ temperature, thus resulting in that the abdominal temperature is significantly higher than the back temperature, and the thermal area (high temperature area) is mainly concentrated in the abdomen, so compared with animals such as black bears, the thermal area gravity center of the giant panda will be offset to the lower part, while the thermal area gravity center of the black bear will not be offset or will be offset to the upper part.

[0049] Based on the above physiological characteristics, the application calculates the gravity center coordinate and the thermal area gravity center coordinate of the animal to be detected through the infrared temperature image, and can quickly determine whether the animal to be detected is a giant panda by calculating whether the deviation between the two meets the requirement;

[0050] Compared with the prior art, the two identification modes of the application can not only effectively improve the identification accuracy, but also can quickly screen out wild animals that do not match the physiological characteristics of the giant panda in the first identification, thereby reducing the calculation amount and being beneficial to improving the identification efficiency;

[0051] Secondly, the infrared image can be used to complete the identification according to the application, compared with the traditional method of identifying the giant panda by the black back color, the technical solution of the application does not need to consider the influence of color and visible light and other external conditions on the identification, so it can effectively improve the identification accuracy of the giant panda at night. BRIEF DESCRIPTION OF DRAWINGS

[0052] Figure 1 A structure diagram of a giant panda rapid identification system based on images is provided for the embodiments of the application.

[0053] Figure 2 A flow chart of a big panda quick identification method based on images provided by the embodiment of the present application is shown in the figure;

[0054] Figure 3 A calculation principle diagram of the barycentric coordinate and the barycentric coordinate of the hot area is shown in the figure;

[0055] The object, function characteristics and advantages of the present application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0056] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all the other embodiments obtained by those skilled in the art without creative work under the premise of the present application, all belong to the scope of protection of the present application.

[0057] It should be noted that all the directionality indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present application are only used to explain the relative position relationship, movement condition, etc. between the components in a certain specific posture (as shown in the drawings). If the specific posture changes, the directionality indications will also change accordingly.

[0058] In the present application, unless otherwise explicitly specified and limited, the terms "connection", "fixation" and the like should be understood in a broad sense. For example, "fixation" can be fixed connection, or detachable connection, or integral; can be mechanical connection, or electrical connection; can be direct connection, or indirect connection through an intermediate medium; can be the internal communication of two elements or the interaction relationship between two elements, unless otherwise explicitly limited. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0059] In addition, if the description of "first", "second" and the like is involved in the embodiments of the present application, the description of "first", "second" and the like is only for the purpose of description, and cannot be understood as indicating or implying the relative importance of the indicated technical features or implicitly indicating the number of the indicated technical features. Therefore, the features limited by "first", "second" can be explicitly or implicitly included at least one of the features. In addition, the meaning of "and / or" appearing throughout the text includes three parallel schemes. For example, "robot coordinate system and / or m" includes robot coordinate system scheme, or m scheme, or robot coordinate system and m scheme. In addition, the technical solutions of each embodiment can be combined with each other, but it must be based on the realization of ordinary skilled in the art. When the combination of technical solutions appears contradictory or unachievable, it should be considered that the combination of technical solutions does not exist and is not within the protection scope required by the present application.

[0060] Embodiment 1

[0061] Referring to the description attached Figure 1 , as an optional embodiment of the present application, it discloses a kind of image-based giant panda fast identification system, including parameter setting module, data acquisition and processing module and first identification module, wherein the input end of the data acquisition and processing module is used to call actual infrared temperature image and environmental temperature parameter measured by temperature sensor and uploaded by infrared camera shooting;

[0062] The output end of the data acquisition and processing module is electrically connected with the first identification module, and the input end of the first identification module is also electrically connected with the parameter setting module to call the first identification standard.

[0063] The fast identification system further includes a first calculation module, a second calculation module and a second identification module, the input end of the first calculation module is electrically connected with the data processing and acquisition module to obtain the actual infrared pixel temperature image after processing;The output end of the first calculation module is connected with the second calculation module and the second identification module in turn, and the input end of the second identification module is electrically connected with the parameter setting module.

[0064] It should be noted that the input end of the first calculation module is also electrically connected with the output end of the first identification module;The above connection is mainly used to receive the result of first identification to determine whether the first calculation module is started.

[0065] Embodiment 2

[0066] Referring to the description attached Figure 2 And Figure 3 , as another optional embodiment of the present application, the present embodiment discloses an identification method of image-based giant panda fast identification system, including the following steps:

[0067] S1, set a first identification criterion and a second identification criterion for giant panda identification;

[0068] In combination with detection of the individual giant panda and collection of literature data, physiological characteristic parameters of the giant panda are obtained, and the first identification criterion and the second identification criterion are respectively set according to the physiological characteristic parameters;

[0069] The expression of the first identification criterion is wherein T0 is a standard temperature threshold, T1 represents an environmental temperature parameter, t1-t 14 is a constant set;

[0070] In the above expression, it should be noted that the subscript in t1-t 14 is only a number, and the above data are set by the staff according to actual experience or measured parameters, such as setting T1 to 7-14℃ and the standard temperature threshold T0 to 6-25℃ after consulting the data; so as to improve the fitness of the first identification criterion to the actual environment of the panda habitat and improve the identification accuracy;

[0071] Meanwhile, in the above expression, 4 gradients can be set, and 3 or 5 gradients can also be set, but considering the identification accuracy and actual working conditions, 2-4 gradients are preferred;

[0072] The expression of the second identification criterion is wherein a represents an actual offset of the heat zone center of gravity, represents a heat zone center of gravity longitudinal coordinate, C y represents a center of gravity longitudinal coordinate;

[0073] S2, obtaining an actual infrared temperature image of an animal to be identified and an environmental temperature parameter T1, and converting the actual infrared temperature image into an actual infrared pixel temperature image according to the environmental temperature parameter T1;

[0074] S21, obtaining an environmental temperature parameter T1;

[0075] S22, setting a temperature detection deviation, and generating a screening temperature domain [(1-P)T1, (1+P)T1] according to the temperature detection deviation and the environmental temperature parameter, wherein P represents the temperature detection deviation;

[0076] Setting the temperature detection deviation P, such as setting the deviation to 2%, and the environmental temperature parameter T1 to 30℃, then the lower limit value of the screening temperature domain is 98%*T1=29.4℃, and the upper limit value is 102%*T1=30.6℃; and the screening temperature domain is [29.4℃, 30.6℃];

[0077] S23, convert the actual infrared temperature image into a pixel image, and obtain actual temperature values of each pixel unit respectively;

[0078] The actual infrared temperature image is disassembled into a plurality of pixel units, and then a unique number is generated for each pixel unit, and the actual temperature values of each pixel unit are detected and read respectively;

[0079] S24, delete the pixel unit whose actual temperature value is in the screening temperature domain from the pixel image to generate an actual infrared pixel temperature image;

[0080] Obtain the screening temperature domain obtained in step S22, and compare each actual temperature value with the screening temperature domain respectively. If the actual temperature value is in the screening temperature domain, it means that the pixel unit corresponding to the actual temperature value is unqualified, and the pixel unit is deleted from the pixel image. Otherwise, the pixel unit corresponding to the actual temperature value is retained;

[0081] The above operation is repeatedly performed, and when the comparison of the actual temperature values of all pixel units is completed, an actual infrared pixel temperature image can be generated;

[0082] The screening temperature domain represents the environment temperature. For a constant temperature animal such as a giant panda, there is a certain temperature difference between the body surface temperature and the environment temperature. The above principle can quickly separate the background part from the actual infrared temperature image, thereby avoiding the adverse effects of the background part on subsequent calculation, improving the calculation accuracy, and reducing the calculation amount in the later stage, thereby improving the identification efficiency of the giant panda;

[0083] S3, combine the first identification standard and the actual infrared pixel temperature image to perform a first identification on the animal to be identified. If the identification fails, the animal to be identified is determined to be another animal, the identification program is terminated, and the identification result is output;

[0084] S31, obtain an actual infrared pixel temperature image;

[0085] S32, obtain actual temperature values of each pixel unit respectively;

[0086] Obtain the actual infrared pixel temperature image obtained in step S2, and obtain the actual temperature values of each pixel unit. It should be noted that in this step, each pixel unit can be numbered, and the pixel unit is corresponded to the number (i.e. a one-to-one mapping relationship is established). Alternatively, the actual temperature values can be collected into a same set without numbering;

[0087] S33, select a standard temperature threshold T0 from the first identification standard according to the environment temperature parameter T1;

[0088] The expression of the first identification criterion is Wherein, T0 is a standard temperature threshold, T1 represents an environmental temperature parameter, t1-t 14 is a constant set;

[0089] An environmental temperature parameter T1 is acquired, and a corresponding standard temperature threshold T0 is selected from the first identification criterion in combination with the environmental temperature parameter; for example, if the environmental temperature is 20℃, the selected standard temperature threshold is 6-25℃.

[0090] With the change of the environmental temperature, the body surface temperature of the large cat will change, and the temperature difference is large. The environmental temperature parameter is corresponded with the standard temperature threshold through the first identification criterion, so as to improve the accuracy of the determination;

[0091] S34, respectively, each of the actual temperature values is compared with the standard temperature threshold, if the standard temperature threshold is met, it is determined to be qualified, and the actual temperature value is collected into a first set, otherwise the actual temperature value is collected into a second set;

[0092] Each actual temperature value obtained in step S32 is compared with the standard temperature threshold, respectively, if the actual temperature value meets the standard temperature threshold, it is determined that the actual temperature value is qualified, and the temperature value is collected into a first set, otherwise the actual temperature value is collected into a second set;

[0093] S35, the number of actual temperature values in the first set Q1 and the number of actual temperature values in the second set Q2 are counted;

[0094] After all the actual temperature values are compared with the standard temperature threshold, the first set and the second set are obtained, and the above set contains a plurality of actual temperature values. The number of actual temperature values in the first set Q1 and the number of actual temperature values in the second set Q2 are counted;

[0095] S36, if , it is determined that the animal to be identified is a giant panda, and the first identification is successful, otherwise it is determined to be other animals, and the first identification fails.

[0096] If the number of actual temperature values in the first set Q1 and the number of actual temperature values in the second set Q2 meet , it is indicated that in the actual infrared pixel temperature image, the actual temperature values in the first set account for the majority;

[0097] Compared with other homeothermic animals, the body surface temperature of the giant panda is in a specific range, which is reflected in the actual infrared pixel temperature image, and it is shown that most pixel units meet the standard temperature threshold. Therefore, if the formula If the temperature deviation is small, it can be determined that the actual infrared pixel temperature image is within the standard temperature threshold calibration range, and the actual infrared pixel temperature image represents a giant panda, and the first identification is successful, and the identification procedure enters the next step.

[0098] If the temperature deviation is large, it can be determined that the actual infrared pixel temperature image is not within the standard temperature threshold calibration range, and the actual infrared pixel temperature image represents an animal other than a giant panda, and the entire identification procedure is automatically terminated.

[0099] Compared with the prior art, in the first identification process, the formula reduces the adverse effects of uneven temperature distribution on identification, improves the accuracy of the first identification, and realizes rapid statistics of the temperature partition area through pixel unit cutting and data statistics, which not only reduces the calculation amount, but also improves the identification efficiency.

[0100] Secondly, the body surface temperature can be used for the first rough identification operation, so as to quickly screen out wild animals that do not match the physiological characteristics of giant pandas, thereby reducing the calculation amount and improving the identification efficiency.

[0101] S4, if the first identification is successful, the actual infrared pixel temperature image is used to generate the center of gravity coordinate parameter and the center of gravity coordinate parameter of the heat area of the animal to be identified.

[0102] S41, an actual infrared pixel temperature image is obtained.

[0103] If the first identification in step S3 is successful, the actual infrared pixel temperature image generated in step S2 is obtained, and the step is entered, otherwise the entire identification process is terminated.

[0104] S42, a minimum bounding box is generated for the actual infrared pixel temperature image, and the height value H of the minimum bounding box is obtained.

[0105] The minimum bounding box is automatically generated for the actual infrared pixel temperature image by a computer, and then the height value H of the minimum bounding box is extracted.

[0106] It should be noted that the height value H is a vertical parameter value.

[0107] S43, a pixel coordinate set and a pixel temperature coordinate set are obtained from the actual infrared pixel temperature image with the minimum bounding box.

[0108] S431, a standard two-dimensional coordinate system is constructed according to the minimum bounding box, wherein the horizontal coordinate of the standard two-dimensional coordinate system represents the length of the minimum bounding box, and the vertical coordinate represents the height of the minimum bounding box.

[0109] A standard two-dimensional coordinate system is generated, wherein the horizontal coordinate of the standard two-dimensional coordinate system represents the length of the minimum bounding box, and the vertical coordinate of the standard two-dimensional coordinate system represents the height of the minimum bounding box;

[0110] Preferably, the length of the minimum bounding box is taken as the horizontal coordinate of the standard two-dimensional coordinate system, and the height of the minimum bounding box is taken as the vertical coordinate of the standard two-dimensional coordinate system;

[0111] S432, according to the standard two-dimensional coordinate system, a unique coordinate (x i , y i ) is generated for each pixel unit, and all the coordinates are collected to generate a pixel coordinate set { (x1, y1), (x2, y2),..., (x i , y i ),..., (x N , y N )}, wherein i represents the pixel unit number, and N represents the total number of pixel units;

[0112] Since the generation of the standard two-dimensional coordinate system depends on the minimum bounding box, after the standard two-dimensional coordinate system is generated, the actual infrared pixel temperature image will be automatically implanted into the standard two-dimensional coordinate system;

[0113] At this time, each pixel unit is numbered according to a specific rule, and a unique coordinate (x i , y i ) is generated, and after all the coordinates are generated, all the coordinates are collected to generate a pixel coordinate set { (x1, y1), (x2, y2),..., (x i , y i ),..., (x N , y N )}, wherein i represents the pixel unit number, and N represents the total number of pixel units;

[0114] It should be noted that the above specific rule refers to a rule specified by the staff for numbering each pixel unit, such as sorting according to an S-shaped curve and then numbering in order, or numbering according to other rules.

[0115] S433, the actual temperature value of each pixel unit is obtained, and each of the actual temperature values is bound to each pixel unit number to generate a pixel temperature coordinate set {T1, T2,..., T i ,..., T N}, wherein i represents the pixel unit number, and N represents the total number of pixel units.

[0116] When the coordinates of each pixel unit are generated, the actual temperature value of each pixel unit is obtained, and each of the actual temperature values is bound to each pixel unit number to generate a pixel temperature coordinate set {T1, T2,..., Ti ... T N}, where i represents the pixel unit number and N represents the total number of pixel units;

[0117] It should be noted that since both the coordinates and the actual temperature value are bound to the pixel unit number, a one-to-one mapping relationship can be established between the coordinates of the pixel unit and the actual temperature value through the number, thereby achieving rapid data pairing, improving calculation efficiency, and avoiding calculation errors caused by incorrect pairing, thus improving the accuracy of the calculation.

[0118] Meanwhile, since the pixel units are too small to be clearly expressed in the accompanying drawings, each pixel unit has been enlarged in the accompanying drawings to facilitate understanding of the technical principles of this application.

[0119] S44. Calculate the centroid coordinate parameters of the animal to be identified based on the pixel coordinate set;

[0120] Retrieve the pixel coordinate set {(x1, y1), (x2, y2), ..., (x...} obtained in step S432 i y i ), ..., (x N y N Then, based on the centroid coordinate parameters, including the centroid ordinate, the formula for calculating the centroid ordinate is: Where N represents the total number of pixel units, i represents the pixel unit number, and y i Represents the ordinate of the pixel unit numbered i;

[0121] To simplify the calculation model and improve recognition efficiency, this application sets the animal to be identified as an individual with uniform density, that is, each pixel unit represents the same mass, thus obtaining the above-mentioned centroid coordinate calculation formula.

[0122] S45. Calculate the centroid coordinate parameters of the thermal zone of the animal to be identified based on the pixel temperature coordinate set.

[0123] Retrieve the pixel coordinate set {(x1, y1), (x2, y2), ..., (x...} obtained in step S432 i y i ), ..., (x N y N The pixel temperature coordinate set {T1, T2, ..., T} in step S433 and step S433 i ... T N}, based on the geocentric coordinate parameters of the hot zone, including the ordinate of the geocentric coordinate, the formula for calculating the ordinate of the geocentric coordinate is: Where N represents the total number of pixel units, i represents the pixel unit number, and y iA longitudinal coordinate of a pixel unit numbered i.

[0124] It should be noted that the horizontal coordinate is not calculated in the calculation of the barycenter coordinate and the thermal zone barycenter coordinate, because the offset of the thermal zone barycenter coordinate of the giant panda mainly appears in the vertical direction, and the offset in the horizontal direction is not representative, so the offset in the horizontal direction is not considered in the actual calculation.

[0125] Referring to the drawings Figure 3 , in Figure 3 the C x indicates the barycenter horizontal coordinate and the thermal zone barycenter horizontal coordinate;

[0126] S5, calculating the actual offset of the thermal zone barycenter of the animal to be identified according to the barycenter coordinate parameter and the thermal zone barycenter coordinate parameter;

[0127] The barycenter longitudinal coordinate, the thermal zone barycenter longitudinal coordinate in step S4 and the height value H of the minimum bounding box in step S42 are called to be brought into the calculation formula of the actual offset of the thermal zone barycenter, and the actual offset of the thermal zone barycenter can be calculated, and the calculation formula of the actual offset of the thermal zone barycenter is: , wherein H represents the height value of the minimum bounding box.

[0128] In the above formula, the numerator clearly indicates the absolute offset value of the thermal zone barycenter relative to the barycenter, and the ratio of the height value can be seen as the proportion of the above offset value in the entire height direction, that is, the actual offset.

[0129] S6, determining whether the actual offset of the thermal zone barycenter meets the second identification standard, if it meets, outputting the animal to be identified as a giant panda, otherwise outputting the animal to be identified as another animal;

[0130] The second identification standard is obtained, and the expression of the second identification standard is , wherein a represents the actual offset of the thermal zone barycenter, the thermal zone barycenter longitudinal coordinate, C y indicates the barycenter longitudinal coordinate; the actual offset of the thermal zone barycenter, the thermal zone barycenter longitudinal coordinate and the barycenter longitudinal coordinate are obtained respectively, and the corresponding identification result can be quickly output according to the above parameters;

[0131] In the above formula, the <C y This condition can eliminate the case that the thermal zone barycenter moves upward, so as to improve the accuracy of the determination;

[0132] Studies have shown that the body surface temperature of the giant panda changes regularly with the change of the ambient temperature, and the body surface temperature of the giant panda is quite different from that of some homeothermic animals, so the identification of the infrared temperature can be used to preliminarily screen the animals to be identified;

[0133] Secondly, compared with the mammals such as bears, the thickness of the back hair of the giant panda is 5-10 cm, the dense wool forms a heat insulation layer to reduce the heat loss, while the abdominal hair is relatively sparse, which is beneficial to the core heat transfer to the body surface to maintain the internal temperature, thus resulting in that the abdominal temperature is significantly higher than the back temperature, and the heat zone (high temperature area) is mainly concentrated in the abdomen, so compared with the animals such as black bears, the gravity center of the heat zone of the giant panda will be downwardly deviated from the gravity center, while the gravity center of the heat zone of the animals such as black bears will not be deviated or will be upwardly deviated;

[0134] Based on the above physiological characteristics, the longitudinal coordinates of the gravity center and the gravity center of the heat zone of the giant panda are calculated through the infrared temperature image, and whether the deviation between the two reaches the requirement can be quickly determined whether the animal to be detected is the giant panda.

[0135] Compared with the prior art, the twice identification mode of the present application can not only effectively improve the accuracy of identification, but also the first identification can quickly screen out the wild animals which are obviously inconsistent with the physiological characteristics of the giant panda, thereby reducing the calculation amount and being beneficial to improve the identification efficiency.

[0136] Secondly, the present application can complete the identification through the infrared image, which does not need to consider the influence of the external conditions such as visible light on the identification, and can effectively improve the identification accuracy of the giant panda at night.

[0137] The above is only the preferred embodiment of the present application, and does not limit the patent scope of the present application, and any equivalent structure or equivalent process transformation using the content of the specification and the drawings, or direct or indirect application in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. An image-based rapid identification system for giant pandas, characterized in that, The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps:

2. The identification method of the image-based giant panda rapid identification system according to claim 1, characterized in that, The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps:

3. The identification method according to claim 2, characterized in that, The expression of the first identification criterion is where T0 is a standard temperature threshold, T1 represents an ambient temperature parameter, t1-t 14 is a constant set; the expression of the second identification criterion is where a represents the actual offset of the thermal zone barycenter, represents the longitudinal coordinate of the thermal zone barycenter, C y represents the longitudinal coordinate of the barycenter.

4. The identification method according to claim 2, characterized in that, The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps:

5. The identification method according to claim 2, characterized in that, The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the following steps: The method comprises the acquiring an actual infrared pixel temperature image; acquiring actual temperature values of each pixel unit respectively; selecting a standard temperature threshold T0 from the first identification standard according to an environmental temperature parameter T1; comparing each actual temperature value with the standard temperature threshold respectively, and if the actual temperature value meets the standard temperature threshold, determining that it is qualified and collecting the actual temperature value into a first set, otherwise collecting the actual temperature value into a second set; counting the number of actual temperature values in the first set Q1 and the number of actual temperature values in the second set Q2; If the following conditions are met then the animal to be identified is determined to be a giant panda, the first identification is successful, otherwise the animal is determined to be another animal, the first identification fails, and the identification program terminates.

6. The identification method according to claim 2, characterized in that, the steps of generating the barycentric coordinate parameter and the thermal area barycentric coordinate parameter of the animal to be identified according to the actual infrared pixel temperature image respectively, comprising the following steps: acquiring an actual infrared pixel temperature image; generating a minimum bounding box for the actual infrared pixel temperature image and acquiring a height value H of the minimum bounding box; acquiring a pixel coordinate set and a pixel temperature coordinate set according to the actual infrared pixel temperature image with the minimum bounding box; calculating the barycentric coordinate parameter of the animal to be identified according to the pixel coordinate set; calculating the thermal area barycentric coordinate parameter of the animal to be identified according to the pixel temperature coordinate set.

7. The identification method according to claim 6, characterized in that, the steps of acquiring a pixel coordinate set and a pixel temperature coordinate set according to the actual infrared pixel temperature image with the minimum bounding box, comprising the following steps: constructing a standard two-dimensional coordinate system according to the minimum bounding box, wherein the horizontal coordinate of the standard two-dimensional coordinate system represents the length of the minimum bounding box and the vertical coordinate represents the height of the minimum bounding box; According to the standard two-dimensional coordinate system, a unique coordinate (x i , y i ) is generated for each pixel unit, and all the coordinates are collected to generate a pixel coordinate set { (x1, y1), (x2, y2),..., (x i , y i ),..., (x N , y N )}, wherein i represents the pixel unit number, and N represents the total number of pixel units. The actual temperature value of each pixel unit is obtained, and each actual temperature value is bound to the pixel unit number to generate a pixel temperature coordinate set {T1, T2, ..., T...}. i ... T N }, where i represents the pixel unit number and N represents the total number of pixel units.

8. The identification method according to claim 2 or 7, characterized in that, The barycentric coordinate parameter comprises a barycentric longitudinal coordinate, and a calculation formula of the barycentric longitudinal coordinate is: Wherein, N represents a total number of pixel units, i represents a pixel unit number, y i represents a longitudinal coordinate of the pixel unit numbered i.

9. The identification method according to claim 2 or 7, characterized in that, The heat zone barycenter coordinate parameter comprises a heat zone barycenter longitudinal coordinate, and a calculation formula of the heat zone barycenter longitudinal coordinate is: Wherein, N represents a total number of pixel units, i represents a pixel unit number, y i represents a longitudinal coordinate of the pixel unit numbered i.

10. The identification method according to claim 3, characterized in that, The formula for calculating the actual offset of the hot area gravity center is: where H represents the height value of the minimum bounding box.

Images