CN115721296A - Wild animal identification system based on PaddleDeprotection - Google Patents
Wild animal identification system based on PaddleDeprotection Download PDFInfo
- Publication number
- CN115721296A CN115721296A CN202211489040.8A CN202211489040A CN115721296A CN 115721296 A CN115721296 A CN 115721296A CN 202211489040 A CN202211489040 A CN 202211489040A CN 115721296 A CN115721296 A CN 115721296A
- Authority
- CN
- China
- Prior art keywords
- preset
- control unit
- central control
- joint
- contour
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 241001465754 Metazoa Species 0.000 title claims abstract description 129
- 241000894007 species Species 0.000 claims description 6
- 238000012544 monitoring process Methods 0.000 abstract description 4
- 230000007774 longterm Effects 0.000 abstract description 3
- 238000001514 detection method Methods 0.000 description 10
- 238000000034 method Methods 0.000 description 9
- 238000013135 deep learning Methods 0.000 description 3
- 230000033001 locomotion Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
Images
Landscapes
- Image Analysis (AREA)
Abstract
The invention relates to the technical field of long-term monitoring of wild animals, in particular to a wild animal identification system based on Paddledetection, which comprises the following components: the identification units are used for acquiring image information of wild animals to be identified; the cloud platform is used for storing preset contour characteristics matched with various wild animal species and corresponding joint node information; the database is used for storing preset contour features matched with various wild animals in the area where the identification unit is located; and the central control unit is used for preliminarily judging the type of the wild animal in the region according to the region where each identification unit is located so as to download the preset contour characteristics and the corresponding joint node information in the region to the database. According to the invention, whether the recognized contour features and the number of joint nodes meet the distribution standard is judged, so that the recognition efficiency of the system is effectively improved.
Description
Technical Field
The invention relates to the technical field of long-term monitoring of wild animals, in particular to a PaddleDeprotection-based wild animal identification system.
Background
The propeller (Paddle) is a hundred-degree independently-developed open source deep learning platform integrating a deep learning core framework, tool components and a service platform. The PaddleDetection is an excellent target detection development kit under a flight paddle deep learning platform, various mainstream target detection, instance segmentation and key point detection algorithms are provided, each network component is modularized, a data enhancement strategy, a loss function strategy and the like are provided, and compression of a model and cross-platform high-performance deployment can help an industrial project to land well.
Long-term monitoring and identification of wild animal diversity is a key link for wild animal management, protection, research and resource utilization. At present, methods for monitoring wild animals mainly comprise a sample line method, a trap method and an infrared camera shooting method.
Chinese patent publication no: CN114305389A discloses a wild animal detection and species identification system and method, the system includes: the first data acquisition unit comprises a pressure detection assembly, the pressure detection assembly is laid on the ground, and the first data acquisition unit outputs a trigger signal when the animal applies pressure to the pressure detection assembly; the shooting unit is used for shooting the ground where the pressure detection assembly is located and outputting a picture after receiving the trigger signal; a data processing unit that processes the photographs and outputs at least a corresponding animal category. When the wild animal crawls, jumps, walks or runs on the pressure detection assembly, the pressure detection assembly receives the pressure action of the wild animal so as to generate a trigger signal.
The prior art can not determine the animal species to be identified according to the profile characteristics in the identified image and the distribution condition of the joint nodes, so that the identification efficiency is low.
Disclosure of Invention
Therefore, the wildlife recognition system based on PaddleDetection is provided for overcoming the problem that the recognition efficiency is low due to the fact that the prior art cannot be based on contour similarity and the number of joint nodes meeting the standard.
In order to achieve the above object, the wildlife recognition system based on PaddleDetection of the present invention comprises:
the identification units are respectively arranged at corresponding positions of different areas and used for acquiring image information of wild animals to be identified;
the cloud platform is used for storing preset contour features matched with various wild animal species and corresponding joint node information;
the database is used for storing preset contour features matched with various wild animals in the area where the identification unit is located, corresponding joint node information and image information acquired by the identification unit;
the central control unit is respectively connected with each identification unit, the database and the cloud platform, and is used for preliminarily judging the type of the wild animals in the region according to the region where each identification unit is located so as to download preset contour features matched with the wild animals in the region and corresponding joint node information in the cloud platform to the database, matching the contour features acquired from the image information acquired by the identification units with the preset contour features in the database so as to preliminarily judge whether the contour features are the wild animals in the corresponding type, and further judging whether the contour features are the wild animals in the corresponding type according to the preset joint node information of the wild animals in the corresponding type when the contour features are preliminarily judged to be the wild animals in the corresponding type.
Further, the central control unit matches the contour features acquired from the image acquired by the identification unit with each preset contour feature in the database, counts and obtains the preset contour feature most similar to the contour features and obtains the contour similarity S between the contour features and the preset contour features, the central control unit compares the S with the preset standard similarity S0 set by the central control unit to preliminarily determine whether the contour features are wild animals of the corresponding type,
if S is larger than or equal to S0, the central control unit judges that the similarity of the contour feature and a preset contour feature most similar to the contour feature meets the standard, the central control unit preliminarily judges the contour feature as a wild animal, and preliminarily judges the type of the wild animal as the type of the wild animal to which the preset contour feature most similar to the contour feature belongs; the central control unit identifies and collects the joint node information in the contour feature and further judges whether the contour feature is a wild animal of the type according to preset joint node information corresponding to the wild animal type;
if S is less than S0, the central control unit judges that the similarity between the outline features and preset outline features most similar to the outline features does not meet the standard, and the central control unit matches the outline features with the preset outline features in the cloud platform to further judge whether the outline features are wild animals of corresponding types.
Further, the central control unit establishes a coordinate system under a first preset condition, collects joint node information in the profile feature in the image information to generate corresponding coordinate values for each joint node, the central control unit matches the coordinate values of each joint node with each preset joint node information corresponding to the type of animal in the database, counts and obtains preset joint node information with the maximum number of joint points in the joint node information, and further determines whether the profile feature is the type of wild animal according to the number E of joint nodes which are not overlapped with the preset joint nodes in the preset joint node information in the joint node information, the central control unit is provided with the number E1 of the first preset joint nodes which do not accord with the distribution standard, the number E2 of the second preset joint nodes which do not accord with the distribution standard and the number E3 of the joint nodes which do not accord with the third preset distribution standard, wherein E1 is more than E2 and less than E3,
if E is less than E1, the central control unit judges the contour features as wild animals of corresponding types;
if E1 is more than or equal to E and less than E2, the central control unit cannot judge whether the contour feature is a wild animal of a corresponding type, the central control unit acquires the priority of each joint node in the joint node information by combining the contour feature, and further judges whether the contour feature is a wild animal of a specific type according to the priority of each joint node;
if E2 is more than or equal to E and less than E3, the central control unit cannot judge whether the contour feature is a wild animal of a corresponding type, the central control unit controls another identification unit to acquire image information of an object corresponding to the contour feature at another angle so as to acquire a corresponding second contour feature, and further judges whether the object corresponding to the contour feature is a wild animal of a corresponding type according to the second contour feature;
if E is larger than or equal to E3, the central control unit judges that the object corresponding to the contour feature is not a wild animal;
the first preset condition is that the central control unit judges that the contour similarity S of the contour feature and the preset contour feature meets the condition that S is larger than or equal to S0.
Further, the central control unit counts the number of the joint nodes which are not overlapped under a second preset condition so as to further judge whether the object corresponding to the joint node is a wild animal of the corresponding type or not according to the number R of the joint nodes which are not overlapped, the central control unit is provided with a preset number R0 of the joint nodes which are not overlapped,
if R is less than or equal to R0, the central control unit judges that the object corresponding to the joint node is a wild animal of a corresponding type;
if R is larger than R0, the central control unit judges that the object corresponding to the joint node is not a wild animal of the corresponding type;
the second preset condition is that the central control unit judges that the contour similarity S of the contour feature and the preset contour feature meets S which is not less than S0, and the number E of the joint nodes which are not overlapped with the preset joint nodes in the preset joint node information in the joint node information meets E1 which is not less than E and is less than E2.
Further, the central control unit counts coordinates of each joint node which is not overlapped and the level of each joint node, judges whether preset contour features corresponding to a new wild animal species and preset joint node information are acquired from the cloud platform according to the absolute distance H between the coordinates of the joint nodes which are not overlapped and the coordinates of the corresponding preset joint nodes so as to re-identify the contour features, is provided with a preset distance H0,
if H is less than or equal to H0, the central control unit judges that new preset contour features and preset joint node information do not need to be acquired from the cloud platform;
and if H is larger than H0, the central control unit judges that new preset contour features and preset joint node information are obtained from the cloud platform.
Further, aiming at a single preset joint node, the central control unit detects the maximum movable angle theta of a connecting line of the joint node and a superior joint node serving as a motion center of the joint node so as to grade the joint node, adjusts the preset distance H0 according to the grade, is provided with a first preset movable angle theta 1, a second preset movable angle theta 2 and a third preset movable angle theta 3, wherein theta 1 is more than theta 2 and more than theta 3,
if theta is less than or equal to theta 1, the central control unit judges that the joint node is a primary joint node and judges that the preset distance H0 does not need to be adjusted;
if theta is larger than or equal to theta 1 and smaller than or equal to theta 2, the central control unit judges that the joint node is a secondary joint node, and adjusts the preset distance H0 to a corresponding value by using e 3;
if theta 2 is larger than theta and smaller than or equal to theta 3, the central control unit judges that the joint node is a third-level joint node, and adjusts the preset distance H0 to a corresponding value by using e 2;
if theta is larger than theta 3, the central control unit judges that the joint node is a four-stage joint node, and adjusts the preset distance H0 to a corresponding value by using e 1;
the central control unit records the preset distance adjusted by ej as H0', and sets H0' = H0 × ej, wherein j =1,2,3.
Further, the central control unit counts joint node information corresponding to the second contour feature collected by other identification units in the area under a third preset condition, marks identification units with the number E of joint nodes in the joint node information which are not overlapped with the corresponding preset joint nodes and are larger than a preset value in the collected joint node information as qualified identification units, marks the ratio of the number of the qualified identification units to the number of the identification units with the collected contour as D, and compares the D with a preset standard ratio D0 set in the central control unit to judge whether the object corresponding to the contour feature is a wild animal of the corresponding type,
if D is larger than or equal to D0, the central control unit judges that the object corresponding to the second contour feature is a wild animal of a corresponding type;
if D is less than D0, the central control unit judges that the object corresponding to the second contour feature is not a wild animal of the corresponding type;
the third preset condition is that the central control unit judges that the contour similarity S of the contour feature and the preset contour feature meets S which is larger than or equal to S0, and the number E of the joint nodes which are not overlapped with the preset joint nodes in the preset joint node information in the joint node information meets E2 which is larger than or equal to E and is smaller than E3.
Further, the central control unit searches from the cloud platform according to each contour feature and corresponding joint node information acquired by each identification unit under a fourth preset condition to judge whether the object is a specific species;
the fourth preset condition is that the central control unit judges that the contour similarity S of the contour feature and the preset contour feature meets S which is not less than S0, the number E of the joint nodes which are not overlapped with the preset joint nodes in the preset joint node information in the joint node information meets E2 which is not less than E < E3, the absolute distance H between the coordinates of the joint nodes which are not overlapped and the coordinates of the corresponding preset joint nodes meets H which is more than H0, and the ratio D between the number of qualified identification units and the number of identification units for collecting the contour meets D < D0.
Further, the central control unit analyzes the contour feature information acquired by each identification unit in a preset period according to the identification result, and transmits the contour feature information to the cloud platform to update the identification standard of the cloud platform for the contour feature of the animal of the kind, the central control unit calculates the ratio B of the number of the identified images to the number of the acquired images, compares B with each preset ratio, and adjusts the standard similarity to a corresponding value according to the comparison result, the central control unit is provided with a first preset ratio B1, a second preset ratio B2, a first preset standard similarity adjustment coefficient alpha 1, a second preset standard similarity adjustment coefficient alpha 2 and a third preset standard similarity adjustment coefficient alpha 3, wherein B1 is less than B2,1 is less than alpha 1 and less than alpha 2 is less than alpha 3 and less than 1.4,
if B is less than or equal to B1, the central control unit adjusts the standard similarity to a corresponding value by using alpha 1, the adjusted standard similarity is recorded as S ', and S' = S multiplied by alpha 3 is set;
if B1 is larger than B and is not larger than B2, the central control unit adjusts the standard similarity to a corresponding value by using alpha 2, and the adjusted standard similarity is recorded as S ', and S' = S multiplied by alpha 2 is set;
if B > B2, the central control unit adjusts the standard similarity to a corresponding value by using α 3, and the adjusted standard similarity is recorded as S ', and S' = S × α 1 is set.
Further, the central control unit determines the usage amount of the identification unit according to the vegetation density G of the area to be detected, the central control unit is provided with a first preset density G1, a second preset density G2, a first preset usage amount adjustment coefficient beta 1, a second preset usage amount adjustment coefficient beta 2 and a third preset usage amount adjustment coefficient beta 3, wherein G1 is more than G2,1 is more than beta 1 and more than beta 2 is more than beta 3 and less than 1.6,
if G is less than or equal to G1, the central control unit adjusts the usage amount of the identification unit to a corresponding value by using beta 1, the adjusted usage amount of the identification unit is marked as W ', and W' = Wx (beta 1-1) is set, wherein W is the initial usage amount of the identification unit;
if G1 is larger than G and smaller than or equal to G2, the central control unit adjusts the use amount of the identification unit to a corresponding value by using beta 2, and the adjusted use amount of the identification unit is marked as W ', and W' = W multiplied by beta 2 is set;
if G > G2, the central control unit adjusts the use amount of the identification unit to a corresponding value by using beta 3, and the adjusted use amount of the identification unit is recorded as W ', and W' = W multiplied by beta 3 is set.
Compared with the prior art, the method has the advantages that the obtained similarity of the outline characteristics is compared with the preset standard similarity, whether the outline characteristics are wild animals of corresponding types or not is judged according to the comparison result, and the identification efficiency of the system is effectively improved; meanwhile, the invention further determines the types of the animals according to the distribution condition of the joint nodes and the number of the joint nodes meeting the distribution standard, effectively further confirms the attributes of the animals to be recognized, avoids the phenomenon of recognition error caused by contour similarity, and further improves the recognition efficiency of the system.
Furthermore, the central control unit is provided with a preset standard similarity S0, the contour similarity is determined by comparing the collected contour features with the preset contour features, the contour similarity is compared with the preset standard similarity, whether the contour features meet the standard or not is judged, and the animal species is preliminarily judged from the contour, so that the identification efficiency of the system is effectively improved.
Furthermore, the central control unit is provided with the number of a plurality of preset joint nodes which do not accord with the distribution standard, and the number of the joint nodes which do not accord with the distribution standard is compared with the number of each preset joint node which does not accord with the distribution standard to judge whether the contour feature is the wild animal of the type, so that the type of the animal to be identified is further determined, and the identification efficiency of the system is further improved.
Furthermore, the central control unit is provided with a preset standard ratio D0, the identification units with the number E of the joint nodes in the joint node information, which are not overlapped with the joint nodes, being larger than the preset value are recorded as qualified identification units, the number of the qualified identification units and the number of the identification units with the contours being acquired are recorded as D, the D and the D0 are compared, whether the object corresponding to the contour features is a wild animal of the corresponding type or not is further judged according to the comparison result, the object corresponding to the contour features is judged in multiple angles, the identification efficiency of the system is effectively improved, and the phenomenon that misjudgment is caused by the fact that only images of a single angle are referred to, and therefore the identification efficiency is influenced is avoided.
Further, when the similarity of the outline is judged to be lower than the similarity of the preset standard, the ratio of the number of the collected identification units with the joint nodes meeting the distribution standard being larger than the preset value is judged to be lower than the preset standard ratio, the absolute distance between the coordinates of the joint nodes which are not overlapped and the coordinates of the corresponding preset joint nodes is larger than the preset distance, and the ratio of the number of the qualified identification units to the number of the collected identification units of the outline is smaller than the preset ratio, the types of the animals to be identified are further judged by searching the collected characteristics of the outline and the joint nodes through the cloud platform, the types of the animals to be identified are further determined by means of network big data, and the identification efficiency of the system is effectively improved.
Furthermore, the central control unit is provided with a plurality of preset ratios and a plurality of preset standard similarity adjusting coefficients, the ratio B of the number of the recognized images to the number of the collected images is compared with the preset ratios, the preset standard similarity is adjusted to a corresponding value according to the comparison result, the recognition accuracy is effectively improved by adjusting the preset standard similarity, and the recognition efficiency of the system is further improved.
Drawings
FIG. 1 is a block diagram of a wildlife recognition system based on PaddleDetection according to an embodiment of the present invention;
FIG. 2 is a flow chart of the method for determining whether the contour feature is a wild animal of the species according to the number E of misaligned joint nodes according to the embodiment of the present invention;
FIG. 3 is a flowchart illustrating an embodiment of determining whether an object corresponding to the contour feature is a wild animal of a corresponding type according to a ratio of the number of qualified identification units to the number of identification units with which the contour is acquired;
fig. 4 is a flowchart illustrating an embodiment of adjusting the similarity of the preset criteria according to a ratio of the number of completed identifications to the number of captured images.
Detailed Description
In order that the objects and advantages of the invention will be more clearly understood, the invention is further described below with reference to examples; it should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only for explaining the technical principle of the present invention, and do not limit the scope of the present invention.
It should be noted that in the description of the present invention, the terms of direction or positional relationship indicated by the terms "upper", "lower", "left", "right", "inner", "outer", etc. are based on the directions or positional relationships shown in the drawings, which are only for convenience of description, and do not indicate or imply that the device or element must have a specific orientation, be constructed in a specific orientation, and be operated, and thus, should not be construed as limiting the present invention.
Furthermore, it should be noted that, in the description of the present invention, unless otherwise explicitly specified or limited, the terms "mounted," "connected," and "connected" are to be construed broadly, and may be, for example, fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood by those skilled in the art according to specific situations.
Fig. 1 is a block diagram showing a structure of a wildlife recognition system based on PaddleDetection according to an embodiment of the present invention, and the wildlife recognition system based on PaddleDetection according to the present invention includes:
the identification units are respectively arranged at corresponding positions of different areas and are used for acquiring image information of wild animals to be identified;
the cloud platform is used for storing preset contour features matched with various wild animal species and corresponding joint node information;
the database is used for storing preset contour features matched with various wild animals in the area where the identification unit is located, corresponding joint node information and image information acquired by the identification unit;
the central control unit is respectively connected with each identification unit, the database and the cloud platform, and is used for preliminarily judging the type of the wild animals in the region according to the region where each identification unit is located so as to download preset contour features matched with the wild animals in the region and corresponding joint node information into the database, matching the contour features acquired from the image information acquired by the identification units with the preset contour features in the database so as to preliminarily judge whether the contour features are the wild animals in the corresponding type, and further judging whether the contour features are the wild animals in the corresponding type according to the preset joint node information of the wild animals when the contour features are preliminarily judged to be the wild animals in the corresponding type.
Specifically, the central control unit matches the profile features acquired from the image acquired by the identification unit with each preset profile feature in the database, counts and obtains the preset profile feature most similar to the profile features and obtains the profile similarity S between the profile features and the preset profile features, the central control unit compares the S with the preset standard similarity S0 set by the central control unit to preliminarily judge whether the profile features are wild animals of the corresponding type or not,
if S is larger than or equal to S0, the central control unit judges that the similarity of the contour feature and a preset contour feature most similar to the contour feature meets the standard, the central control unit preliminarily judges the contour feature as a wild animal, and preliminarily judges the type of the wild animal as the type of the wild animal to which the preset contour feature most similar to the contour feature belongs; the central control unit identifies and collects the joint node information in the contour feature and further judges whether the contour feature is a wild animal of the type according to preset joint node information corresponding to the wild animal type;
if S is less than S0, the central control unit judges that the similarity between the outline features and preset outline features most similar to the outline features does not meet the standard, and the central control unit matches the outline features with the preset outline features in the cloud platform to further judge whether the outline features are wild animals of corresponding types.
It can be understood that the method for calculating the similarity between the contour feature acquired from the image information and the corresponding preset standard contour feature is as follows: and counting the proportion of the length of the completely coincident contour line in the two contours to the total length of the corresponding preset standard contour.
The central control unit is provided with a preset standard similarity S0, the contour similarity is determined by comparing the collected contour features with the preset contour features, whether the contour features meet the standard or not is judged by comparing the contour similarity with the preset standard similarity, and the identification efficiency of the system is effectively improved by preliminarily judging the animal types from the contours.
Please refer to fig. 2, which is a flowchart illustrating an embodiment of the present invention for determining a type of an animal to be identified according to a number of joint nodes that do not meet a distribution standard, wherein a central control unit establishes a coordinate system under a first preset condition, collects joint node information in a profile feature in image information to generate corresponding coordinate values for the joint nodes, the central control unit matches the coordinate values of the joint nodes with preset joint node information corresponding to the type of animal in a database, counts and obtains preset joint node information with the maximum number of joint points overlapping in the joint node information, and further determines whether the profile feature is a wild animal of the type according to a number E of joint nodes that do not overlap with the preset joint nodes in the preset joint node information, the central control unit is provided with a number E1 of joint nodes that do not meet the distribution standard, a number E2 of joint nodes that do not meet the distribution standard, and a number E3 of joint nodes that do not meet the distribution standard, where E1 < E2 < E3,
if E is less than E1, the central control unit judges the contour features as wild animals of corresponding types;
if E1 is more than or equal to E and less than E2, the central control unit cannot judge whether the contour feature is a wild animal of a corresponding type, the central control unit acquires the priority of each joint node in the joint node information by combining the contour feature, and further judges whether the contour feature is a wild animal of a specific type according to the priority of each joint node;
if E2 is more than or equal to E and less than E3, the central control unit cannot judge whether the contour feature is a wild animal of a corresponding type, the central control unit controls another identification unit to acquire image information of an object corresponding to the contour feature at another angle so as to acquire a corresponding second contour feature from the image information, and further judges whether the object corresponding to the contour feature is a wild animal of a corresponding type according to the second contour feature;
if E is larger than or equal to E3, the central control unit judges that the object corresponding to the contour feature is not a wild animal;
the first preset condition is that the central control unit judges that the contour similarity S of the contour feature and the preset contour feature meets S which is larger than or equal to S0.
It can be understood that, regarding the preset joint node information, the program sequentially simulates all the movements of the animal to form a plurality of continuous joint node distribution point positions, and projects the joint nodes into the coordinate system to obtain a series of coordinate values.
The central control unit is provided with the number of a plurality of preset joint nodes which do not accord with the distribution standard, and the number of the joint nodes which do not accord with the distribution standard is compared with the number of each preset joint node which does not accord with the distribution standard to judge whether the contour feature is the wild animal of the type, so that the type of the animal to be identified is further determined, and the identification efficiency of the system is further improved.
Specifically, the central control unit counts the number of the joint nodes which are not overlapped under the second preset condition so as to further judge whether the object corresponding to the joint node is a wild animal of the corresponding type or not according to the number R of the joint nodes which are not overlapped, the central control unit is provided with a preset number R0 of the joint nodes which are not overlapped,
if R is less than or equal to R0, the central control unit judges that the object corresponding to the joint node is a wild animal of a corresponding type;
if R is larger than R0, the central control unit judges that the object corresponding to the joint node is not a wild animal of the corresponding type;
the second preset condition is that the central control unit judges that the contour similarity S of the contour feature and the preset contour feature meets the condition that S is larger than or equal to S0, and the number E of the joint nodes which are not overlapped with the preset joint nodes in the preset joint node information meets the condition that E1 is larger than or equal to E and smaller than E2.
Specifically, the central control unit counts coordinates of each joint node which is not overlapped and the level to which each joint node belongs, judges whether preset contour features corresponding to a new wild animal species and preset joint node information are acquired from the cloud platform according to the absolute distance H between the coordinates of the joint nodes which are not overlapped and the coordinates of the corresponding preset joint nodes so as to re-identify the contour features, is provided with a preset distance H0,
if H is less than or equal to H0, the central control unit judges that new preset contour features and preset joint node information do not need to be acquired from the cloud platform;
and if H is larger than H0, the central control unit judges that new preset contour features and preset joint node information are obtained from the cloud platform.
Specifically, for a single preset joint node, the central control unit detects the maximum movable angle theta of a connecting line between the joint node and a superior joint node serving as a motion center of the joint node to grade the joint node, adjusts the preset distance H0 according to the grade, is provided with a first preset movable angle theta 1, a second preset movable angle theta 2 and a third preset movable angle theta 3, wherein theta 1 is more than theta 2 and less than theta 3,
if theta is less than or equal to theta 1, the central control unit judges that the joint node is a primary joint node and judges that the preset distance H0 does not need to be adjusted;
if theta 1 is larger than theta and is smaller than or equal to theta 2, the central control unit judges the joint node to be a secondary joint node, and the preset distance H0 is adjusted to a corresponding value by using e 3;
if theta 2 is larger than theta and is smaller than or equal to theta 3, the central control unit judges that the joint node is a three-level joint node, and the preset distance H0 is adjusted to a corresponding value by using e 2;
if theta is larger than theta 3, the central control unit judges that the joint node is a four-stage joint node, and adjusts the preset distance H0 to a corresponding value by using e 1;
the central control unit records the preset distance adjusted by ej as H0', and sets H0' = H0 × ej, wherein j =1,2,3.
Please refer to fig. 3, which is a flowchart illustrating whether the ratio of the number of qualified identification units to the number of identification units acquiring a contour meets a standard according to an embodiment of the present invention, where the central control unit counts joint node information corresponding to the second contour feature acquired by other identification units in the area under a third preset condition, marks an identification unit, in which the number E of joint nodes not coinciding with a corresponding preset joint node in the acquired joint node information is greater than a preset value, as a qualified identification unit, marks the ratio of the number of qualified identification units to the number of identification units acquiring a contour as D, and compares the D with a preset standard ratio D0 set in the central control unit to determine whether an object corresponding to the contour feature is a wild animal of a corresponding type,
if D is larger than or equal to D0, the central control unit judges that the object corresponding to the second contour feature is a wild animal of a corresponding type;
if D is less than D0, the central control unit judges that the object corresponding to the second contour feature is not a wild animal of the corresponding type;
the third preset condition is that the central control unit judges that the contour similarity S of the contour feature and the preset contour feature meets S which is larger than or equal to S0, and the number E of the joint nodes which are not overlapped with the preset joint nodes in the preset joint node information in the joint node information meets E2 which is larger than or equal to E and is smaller than E3.
The central control unit is provided with a preset standard proportion D0, the identification units with the number E of the joint nodes in the joint node information which is collected and not overlapped with the joint nodes and larger than a preset value are marked as qualified identification units, the number of the qualified identification units and the number of the identification units with the collected outlines are marked as D, whether the object corresponding to the outline characteristics is a wild animal of the corresponding type or not is further judged according to the comparison result by comparing the D with the D0, and the object corresponding to the outline characteristics is judged in multiple angles, so that the identification efficiency of the system is effectively improved, and the misjudgment caused by only referring to an image of a single angle is avoided, and the identification efficiency is further influenced.
Specifically, the central control unit searches from the cloud platform according to each contour feature and corresponding joint node information acquired by each identification unit under a fourth preset condition to determine whether the object is a specific species;
the fourth preset condition is that the central control unit judges that the contour similarity S of the contour feature and the preset contour feature meets S & gt S0, the number E of joint nodes which are not overlapped with the preset joint nodes in the preset joint node information in the joint node information meets E2 & lt E3, the absolute distance H between the coordinates of the joint nodes which are not overlapped and the coordinates of the corresponding preset joint nodes meets H & gt H0, and the ratio of the number of qualified identification units to the number of identification units for acquiring the contour D & lt D0.
When the similarity of the outline is judged to be lower than the similarity of the preset standard, the proportion of the identification units, the number of which is more than the preset value and accords with the distribution standard, is judged to be lower than the proportion of the preset standard, the absolute distance between the coordinates of the joint nodes which are not overlapped and the coordinates of the corresponding preset joint nodes is more than the preset distance, and the proportion of the number of the qualified identification units and the number of the identification units, which are collected to be the outline, is less than the preset proportion, the types of the animals to be identified are further judged by searching the collected outline characteristics and the joint nodes through the cloud platform, and the types of the animals to be identified are further determined by means of network big data, so that the identification efficiency of the system is effectively improved.
Please refer to fig. 4, which is a flowchart illustrating an embodiment of the present invention adjusting a preset standard similarity according to a ratio of a number of recognized images to a number of collected images, wherein a central control unit analyzes profile feature information collected by each recognition unit in a preset period according to a recognition result, and transmits the profile feature information to a cloud platform to update a recognition standard of the cloud platform for a profile feature of an animal of a type, the central control unit calculates a ratio B of the number of recognized images to the number of collected images, compares B with each preset ratio, and adjusts the standard similarity to a corresponding value according to a comparison result, the central control unit is provided with a first preset ratio B1, a second preset ratio B2, a first preset standard similarity adjustment coefficient α 1, a second preset standard similarity adjustment coefficient α 2, and a third preset standard similarity adjustment coefficient α 3, wherein B1 < B2,1 < α 2 < α 3 < 1.4,
if B is less than or equal to B1, the central control unit adjusts the standard similarity to a corresponding value by using alpha 1, the adjusted standard similarity is recorded as S ', and S' = S multiplied by alpha 3 is set;
if B1 is larger than B and is not larger than B2, the central control unit adjusts the standard similarity to a corresponding value by using alpha 2, and the adjusted standard similarity is recorded as S ', and S' = S multiplied by alpha 2 is set;
if B > B2, the central control unit adjusts the standard similarity to a corresponding value by using α 3, and the adjusted standard similarity is recorded as S ', and S' = sxα 1 is set.
The central control unit is provided with a plurality of preset ratios and a plurality of preset standard similarity adjusting coefficients, the ratio B of the number of the recognized images to the number of the collected images is compared with the preset ratios, the preset standard similarity is adjusted to a corresponding value according to the comparison result, the recognition accuracy is effectively improved by adjusting the preset standard similarity, and the recognition efficiency of the system is further improved.
Specifically, the central control unit determines the usage amount of the identification unit according to the vegetation density G of the area to be detected, the central control unit is provided with a first preset density G1, a second preset density G2, a first preset usage amount adjusting coefficient beta 1, a second preset usage amount adjusting coefficient beta 2 and a third preset usage amount adjusting coefficient beta 3, wherein G1 is more than G2,1 is more than beta 1 and more than beta 2 and more than beta 3 and less than 1.6,
if G is less than or equal to G1, the central control unit adjusts the use amount of the identification unit to a corresponding value by using beta 1, the adjusted use amount of the identification unit is marked as W ', and W' = W x (beta 1-1) is set, wherein W is the initial use amount of the identification unit;
if G1 is less than or equal to G2, the central control unit adjusts the use amount of the identification unit to a corresponding value by using beta 2, and the adjusted use amount of the identification unit is recorded as W ', and W' = W multiplied by beta 2 is set;
if G > G2, the central control unit adjusts the use amount of the identification unit to a corresponding value by using beta 3, and the adjusted use amount of the identification unit is recorded as W ', and W' = W multiplied by beta 3 is set.
So far, the technical solutions of the present invention have been described in connection with the preferred embodiments shown in the drawings, but it is easily understood by those skilled in the art that the scope of the present invention is obviously not limited to these specific embodiments. Equivalent changes or substitutions of related technical features can be made by those skilled in the art without departing from the principle of the invention, and the technical scheme after the changes or substitutions can fall into the protection scope of the invention.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention; various modifications and alterations to this invention will become apparent to those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. A wildlife identification system based on PaddleDetection, comprising:
the identification units are respectively arranged at corresponding positions of different areas and are used for acquiring image information of wild animals to be identified;
the cloud platform is used for storing preset contour features matched with various wild animal species and corresponding joint node information;
the database is used for storing preset contour features matched with various wild animals in the area where the identification unit is located, corresponding joint node information and image information acquired by the identification unit;
the central control unit is respectively connected with each identification unit, the database and the cloud platform, and is used for preliminarily judging the type of the wild animals in the region according to the region where each identification unit is located so as to download preset contour features matched with the wild animals in the region and corresponding joint node information in the cloud platform to the database, matching the contour features acquired from the image information acquired by the identification units with the preset contour features in the database so as to preliminarily judge whether the contour features are the wild animals in the corresponding type, and further judging whether the contour features are the wild animals in the corresponding type according to the preset joint node information of the wild animals in the corresponding type when the contour features are preliminarily judged to be the wild animals in the corresponding type.
2. The wildlife identification system according to claim 1, wherein the central control unit matches the contour features obtained from the images collected by the identification unit with the preset contour features in the database, counts and obtains the preset contour feature most similar to the contour features and finds out the contour similarity S between the contour features and the preset contour features, the central control unit compares S with the preset standard similarity S0 set by the central control unit to preliminarily determine whether the contour features are wildlife of the corresponding type,
if S is larger than or equal to S0, the central control unit judges that the similarity of the outline feature and a preset outline feature most similar to the outline feature meets the standard, the central control unit preliminarily judges the outline feature as a wild animal, and preliminarily judges the type of the wild animal as the type of the wild animal to which the preset outline feature most similar to the outline feature belongs; the central control unit identifies and collects the joint node information in the contour feature and further judges whether the contour feature is a wild animal of the type according to preset joint node information corresponding to the wild animal type;
if S is less than S0, the central control unit judges that the similarity between the outline features and preset outline features most similar to the outline features does not meet the standard, and the central control unit matches the outline features with the preset outline features in the cloud platform to further judge whether the outline features are wild animals of corresponding types.
3. The wildlife identification system based on PaddleDetection as claimed in claim 2, wherein said central control unit establishes a coordinate system under a first preset condition and collects the joint node information in the profile feature in said image information to generate corresponding coordinate values for each joint node, the central control unit matches the coordinate values of each joint node with each preset joint node information corresponding to the type of animal in said database, counts and obtains the preset joint node information with the most number of joint points in the joint node information, and further determines whether the profile feature is the wildlife of the type according to the number E of joint nodes in said joint node information that do not coincide with the preset joint nodes in the preset joint node information, the central control unit has the number E1 of joint nodes that do not accord with the distribution standard in the first preset, the number E2 of joint nodes that do not accord with the distribution standard in the second preset, and the number E3 of joint nodes that do not accord with the distribution standard in the third preset, wherein E1 < E2 < E3,
if E is less than E1, the central control unit judges the contour features as wild animals of corresponding types;
if E1 is less than or equal to E < E2, the central control unit cannot judge whether the contour feature is a wild animal of a corresponding type, the central control unit acquires the priority of each joint node in the joint node information by combining the contour feature, and further judges whether the contour feature is a wild animal of a specific type according to the priority of each joint node;
if E2 is more than or equal to E and less than E3, the central control unit cannot judge whether the contour feature is a wild animal of a corresponding type, the central control unit controls another identification unit to acquire image information of an object corresponding to the contour feature at another angle so as to acquire a corresponding second contour feature, and further judges whether the object corresponding to the contour feature is a wild animal of a corresponding type according to the second contour feature;
if E is larger than or equal to E3, the central control unit judges that the object corresponding to the outline feature is not a wild animal;
the first preset condition is that the central control unit judges that the contour similarity S of the contour feature and the preset contour feature meets the condition that S is larger than or equal to S0.
4. The wildlife recognition system based on PaddleDetection as claimed in claim 3, wherein said central control unit counts the number of non-overlapped joint nodes under a second preset condition to further determine whether the object corresponding to the joint node is wildlife of the corresponding kind according to the number R of non-overlapped joint nodes, the central control unit is provided with a preset number R0 of non-overlapped joint nodes,
if R is less than or equal to R0, the central control unit judges that the object corresponding to the joint node is a wild animal of a corresponding type;
if R is larger than R0, the central control unit judges that the object corresponding to the joint node is not a wild animal of the corresponding type;
the second preset condition is that the central control unit judges that the contour similarity S of the contour feature and the preset contour feature meets the condition that S is larger than or equal to S0, and the number E of the joint nodes which are not overlapped with the preset joint nodes in the preset joint node information meets the condition that E1 is larger than or equal to E and smaller than E2.
5. The wildlife identification system based on PaddleDetection according to claim 4, wherein the central control unit counts the coordinates of each joint node which is not overlapped and the level of each joint node, determines whether to obtain the preset contour feature corresponding to the new wildlife species and the preset joint node information from the cloud platform according to the absolute distance H between the coordinates of the joint nodes which are not overlapped and the coordinates of the corresponding preset joint nodes to re-identify the contour feature, and is provided with a preset distance H0,
if H is less than or equal to H0, the central control unit judges that new preset contour features and preset joint node information do not need to be acquired from the cloud platform;
and if H is larger than H0, the central control unit judges that new preset contour features and preset joint node information are obtained from the cloud platform.
6. The wildlife recognition system based on PaddleDetection according to claim 5, wherein said central control unit detects a maximum moving angle θ of a connecting line of the joint node and a superior joint node as a moving center of the joint node for ranking the joint node, wherein the central control unit adjusts said preset distance H0 according to the level, wherein the central control unit is provided with a first preset moving angle θ 1, a second preset moving angle θ 2, a third preset moving angle θ 3, a first preset distance adjusting coefficient e1, a second preset distance adjusting coefficient e2, and a third preset distance adjusting coefficient e3, wherein θ 1 < θ 2 < θ 3,0 < e1 < e2 < e3 < 1,
if theta is less than or equal to theta 1, the central control unit judges that the joint node is a primary joint node and judges that the preset distance H0 does not need to be adjusted;
if theta 1 is larger than theta and is smaller than or equal to theta 2, the central control unit judges the joint node to be a secondary joint node, and the preset distance H0 is adjusted to a corresponding value by using e 3;
if theta 2 is larger than theta and smaller than or equal to theta 3, the central control unit judges that the joint node is a third-level joint node, and adjusts the preset distance H0 to a corresponding value by using e 2;
if theta is larger than theta 3, the central control unit judges that the joint node is a four-stage joint node, and adjusts the preset distance H0 to a corresponding value by using e 1;
the central control unit records the preset distance adjusted by ej as H0', and sets H0' = H0 × ej, wherein j =1,2,3.
7. The wildlife identification system based on PaddleDetection according to claim 6, wherein the central control unit counts the joint node information corresponding to the second profile feature collected by other identification units in the area under a third preset condition, records the identification unit with the number E of joint nodes not coinciding with the corresponding preset joint node in the collected joint node information larger than a preset value as a qualified identification unit, records the ratio of the number of qualified identification units to the number of identification units with outlines collected as D, and compares D with a preset standard ratio D0 set in the central control unit to determine whether the object corresponding to the profile feature is the wildlife of the corresponding type,
if D is larger than or equal to D0, the central control unit judges that the object corresponding to the second contour feature is a wild animal of a corresponding type;
if D is less than D0, the central control unit judges that the object corresponding to the second contour feature is not a wild animal of the corresponding type;
the third preset condition is that the central control unit judges that the contour similarity S of the contour feature and the preset contour feature meets S which is larger than or equal to S0, and the number E of the joint nodes which are not overlapped with the preset joint nodes in the preset joint node information in the joint node information meets E2 which is larger than or equal to E and is smaller than E3.
8. The wildlife recognition system based on PaddleDetection as claimed in claim 7, wherein said central control unit searches from said cloud platform based on each contour feature and corresponding joint node information collected by each recognition unit under a fourth preset condition to determine whether the object is a specific species;
the fourth preset condition is that the central control unit judges that the contour similarity S of the contour feature and the preset contour feature meets S & gt S0, the number E of joint nodes which are not overlapped with the preset joint nodes in the preset joint node information in the joint node information meets E2 & lt E3, the absolute distance H between the coordinates of the joint nodes which are not overlapped and the coordinates of the corresponding preset joint nodes meets H & gt H0, and the ratio of the number of qualified identification units to the number of identification units for acquiring the contour D & lt D0.
9. The wildlife recognition system based on PaddleDetection as claimed in claim 8, wherein said central control unit analyzes the profile feature information collected by each recognition unit in a preset period according to the recognition result and transmits the profile feature information to said cloud platform to update the recognition standard of the cloud platform for the profile feature of the kind of animal, the central control unit calculates the ratio B of the number of recognized images to the number of collected images, compares B with each preset ratio, adjusts the standard similarity to a corresponding value according to the comparison result, the central control unit is provided with a first preset ratio B1, a second preset ratio B2, a first preset standard similarity adjustment coefficient α 1, a second preset standard similarity adjustment coefficient α 2 and a third preset standard similarity adjustment coefficient α 3, wherein B1 < B2,1 < α 2 < α 3 < 1.4,
if B is less than or equal to B1, the central control unit adjusts the standard similarity to a corresponding value by using alpha 1, the adjusted standard similarity is recorded as S ', and S' = S multiplied by alpha 3 is set;
if B1 is greater than B and less than or equal to B2, the central control unit adjusts the standard similarity to a corresponding value by using alpha 2, and the adjusted standard similarity is recorded as S ', and S' = S multiplied by alpha 2 is set;
if B > B2, the central control unit adjusts the standard similarity to a corresponding value by using α 3, and the adjusted standard similarity is recorded as S ', and S' = S × α 1 is set.
10. The wildlife recognition system based on PaddleDetection according to claim 9, wherein the central control unit determines the usage amount of the recognition unit according to the vegetation density G of the area to be detected, the central control unit is provided with a first preset density G1, a second preset density G2, a first preset usage amount adjustment coefficient β 1, a second preset usage amount adjustment coefficient β 2 and a third preset usage amount adjustment coefficient β 3, wherein G1 < G2,1 < β 2 < β 3 < 1.6,
if G is less than or equal to G1, the central control unit adjusts the use amount of the identification unit to a corresponding value by using beta 1, the adjusted use amount of the identification unit is marked as W ', and W' = W x (beta 1-1) is set, wherein W is the initial use amount of the identification unit;
if G1 is larger than G and smaller than or equal to G2, the central control unit adjusts the use amount of the identification unit to a corresponding value by using beta 2, and the adjusted use amount of the identification unit is marked as W ', and W' = W multiplied by beta 2 is set;
if G > G2, the central control unit adjusts the use amount of the identification unit to a corresponding value by using beta 3, and the adjusted use amount of the identification unit is marked as W ', and W' = W multiplied by beta 3 is set.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211489040.8A CN115721296B (en) | 2022-11-25 | 2022-11-25 | PaddleDetection-based wild animal identification system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211489040.8A CN115721296B (en) | 2022-11-25 | 2022-11-25 | PaddleDetection-based wild animal identification system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115721296A true CN115721296A (en) | 2023-03-03 |
| CN115721296B CN115721296B (en) | 2024-09-13 |
Family
ID=85298332
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211489040.8A Active CN115721296B (en) | 2022-11-25 | 2022-11-25 | PaddleDetection-based wild animal identification system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115721296B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116933138A (en) * | 2023-07-26 | 2023-10-24 | 广州淦源智能科技有限公司 | Intelligent racing product-based athletic control system |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107820616A (en) * | 2015-07-01 | 2018-03-20 | 维京遗传学Fmba | System and method for identifying individual animals based on back image |
| KR20190033892A (en) * | 2017-09-22 | 2019-04-01 | 국립생태원 | System and method for monitoring wild animals |
| CN109829888A (en) * | 2018-12-27 | 2019-05-31 | 北京林业大学 | Wild animal monitoring analysis system and method based on depth convolutional neural networks |
| CN111160361A (en) * | 2019-11-18 | 2020-05-15 | 中国科学院自动化研究所 | Image classification method and device and computer readable medium |
| CN112016380A (en) * | 2019-05-30 | 2020-12-01 | 中国科学院动物研究所 | Wild animal monitoring method and system |
| CN113963298A (en) * | 2021-10-25 | 2022-01-21 | 东北林业大学 | Wild animal identification tracking and behavior detection system, method, equipment and storage medium based on computer vision |
| CN114373113A (en) * | 2021-12-03 | 2022-04-19 | 浙江臻善科技股份有限公司 | Wild animal species image identification system based on AI technology |
| WO2022138584A1 (en) * | 2020-12-21 | 2022-06-30 | 株式会社リヴァンプ | Information processing device, information processing method, and program |
-
2022
- 2022-11-25 CN CN202211489040.8A patent/CN115721296B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107820616A (en) * | 2015-07-01 | 2018-03-20 | 维京遗传学Fmba | System and method for identifying individual animals based on back image |
| KR20190033892A (en) * | 2017-09-22 | 2019-04-01 | 국립생태원 | System and method for monitoring wild animals |
| CN109829888A (en) * | 2018-12-27 | 2019-05-31 | 北京林业大学 | Wild animal monitoring analysis system and method based on depth convolutional neural networks |
| CN112016380A (en) * | 2019-05-30 | 2020-12-01 | 中国科学院动物研究所 | Wild animal monitoring method and system |
| CN111160361A (en) * | 2019-11-18 | 2020-05-15 | 中国科学院自动化研究所 | Image classification method and device and computer readable medium |
| WO2022138584A1 (en) * | 2020-12-21 | 2022-06-30 | 株式会社リヴァンプ | Information processing device, information processing method, and program |
| CN113963298A (en) * | 2021-10-25 | 2022-01-21 | 东北林业大学 | Wild animal identification tracking and behavior detection system, method, equipment and storage medium based on computer vision |
| CN114373113A (en) * | 2021-12-03 | 2022-04-19 | 浙江臻善科技股份有限公司 | Wild animal species image identification system based on AI technology |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116933138A (en) * | 2023-07-26 | 2023-10-24 | 广州淦源智能科技有限公司 | Intelligent racing product-based athletic control system |
| CN116933138B (en) * | 2023-07-26 | 2024-03-19 | 广州淦源智能科技有限公司 | Intelligent racing product-based athletic control system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115721296B (en) | 2024-09-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111353413B (en) | Low-missing-report-rate defect identification method for power transmission equipment | |
| CN107667903B (en) | Livestock breeding living body weight monitoring method based on Internet of things | |
| CN104185275B (en) | A kind of indoor orientation method based on WLAN | |
| CN110456363A (en) | The target detection and localization method of three-dimensional laser radar point cloud and infrared image fusion | |
| CN107256017B (en) | Route planning method and system | |
| CN111666855B (en) | Animal three-dimensional parameter extraction method and system based on unmanned aerial vehicle and electronic equipment | |
| CN108844541A (en) | A kind of machine based on inspection scene patrols paths planning method | |
| CN114266891A (en) | Railway operation environment abnormity identification method based on image and laser data fusion | |
| CN115721296A (en) | Wild animal identification system based on PaddleDeprotection | |
| US20230104047A1 (en) | Track fusion method and device for unmanned surface vehicle | |
| CN112285733A (en) | Urban and rural planning verification and mapping data processing method | |
| CN113850312A (en) | Forest ecological condition monitoring method and device, electronic equipment and storage medium | |
| CN116863452A (en) | Plant disease identification method, storage medium, and identification system | |
| CN117630907B (en) | Sea surface target tracking method integrating infrared imaging and millimeter wave radar | |
| CN118297094A (en) | Chicken house assembly line egg screening and counting method and system based on low-calculation-force edge equipment | |
| CN114037895A (en) | Unmanned aerial vehicle pole tower inspection image identification method | |
| CN117765420A (en) | terrain surveying method and system based on remote sensing data | |
| CN116740378B (en) | Garden plant diseases and insect pests evaluation system based on image processing | |
| CN115297303B (en) | Image data acquisition and processing method and device suitable for power grid power transmission and transformation equipment | |
| CN115661363B (en) | Three-dimensional building model construction method utilizing unmanned aerial vehicle oblique photography | |
| EP4064120A1 (en) | Method for determining images plausible to have a false negative object detection | |
| WO2022181130A1 (en) | Information processing method, information processing device, information processing system, and program | |
| CN115830342A (en) | Method and device for determining detection frame, storage medium and electronic device | |
| CN113836251B (en) | Cognitive map construction method, device, equipment and medium | |
| CN115240091A (en) | Method and device for quickly detecting rice panicle grains |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |