CN117809099A - Method and system for predicting bird category by means of key part prediction network - Google Patents
Method and system for predicting bird category by means of key part prediction network Download PDFInfo
- Publication number
- CN117809099A CN117809099A CN202311846498.9A CN202311846498A CN117809099A CN 117809099 A CN117809099 A CN 117809099A CN 202311846498 A CN202311846498 A CN 202311846498A CN 117809099 A CN117809099 A CN 117809099A
- Authority
- CN
- China
- Prior art keywords
- bird
- target
- key part
- key
- gesture
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 35
- 230000002159 abnormal effect Effects 0.000 claims abstract description 23
- 238000001514 detection method Methods 0.000 claims abstract description 22
- 210000003323 beak Anatomy 0.000 claims abstract description 21
- 241000894007 species Species 0.000 claims description 24
- 241000271566 Aves Species 0.000 claims description 17
- 230000037237 body shape Effects 0.000 claims description 16
- 241000272814 Anser sp. Species 0.000 claims description 5
- 230000015572 biosynthetic process Effects 0.000 claims description 3
- 238000003786 synthesis reaction Methods 0.000 claims description 3
- 241001529251 Gallinago gallinago Species 0.000 claims description 2
- 230000015654 memory Effects 0.000 description 14
- 238000004590 computer program Methods 0.000 description 10
- 230000006870 function Effects 0.000 description 8
- 241000208152 Geranium Species 0.000 description 4
- 208000028752 abnormal posture Diseases 0.000 description 2
- 210000003128 head Anatomy 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 241000607479 Yersinia pestis Species 0.000 description 1
- 230000001133 acceleration Effects 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 235000013372 meat Nutrition 0.000 description 1
- 235000015097 nutrients Nutrition 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
Landscapes
- Image Analysis (AREA)
Abstract
The invention relates to a method and a system for predicting bird species by means of a critical-position prediction network, wherein the method comprises the following steps: detecting a bird target frame according to data to be detected by a target detection network, wherein the bird target frame comprises the position of a bird target and the bird category; predicting bird key part positions and scores in the bird target frame according to a key part prediction model obtained through training, wherein the bird key parts comprise beaks, crowns, backs, eyes, wings, legs and tails; judging the current posture of the target according to the positions and scores of the key parts of the birds and by combining the geometric relations among different parts, wherein the current posture of the target comprises a normal posture and an abnormal shielding posture; judging the final bird type according to the current posture of the target; the invention distinguishes the current gesture of the detection target, and carries out category judgment by combining auxiliary information such as the length of the key part, the geometric relationship and the like, thereby improving the accuracy of category prediction.
Description
Technical Field
The invention relates to the technical field of bird identification, in particular to a method and a system for predicting bird types by means of a key part prediction network.
Background
Birds are excellent ecosystem service providers that pollinate flowers, remove rotten meat, spread seeds, phagocytize pests, circulate nutrients and benefit the environment to benefit other species and humans, while birds play a critical ecological role in the health and consistency of many ecosystems as an important element of the ecosystem. In addition, birds are also very sensitive to changes in habitat structure and composition, a good indicator of habitat quality and biodiversity. With the acceleration of global climate and land utilization changes of vermilion, bird species and numbers are decreasing, and the importance of wild bird population monitoring is increasing. The classification and counting of birds is essential in almost all bird monitoring studies.
Because vegetation, topographic features and the like of the bird living environment are complex and changeable, the bird living environment is easy to be shielded by surrounding environments, and the bird living environment is various in motion gestures, so that interference is caused to the identification and class judgment of the target detection network. The existing target detection network cannot accurately identify bird gestures and categories under the condition of being blocked, the existing key point detection network is a detection frame and predicts key point positions, the problem is that the detection effect is poor for small targets, a large training sample is needed, and the data size of the single key point prediction network is small. The invention provides a method for improving bird detection precision under the conditions of abnormal posture, shielding and the like by means of a key part prediction network.
Disclosure of Invention
Aiming at the problems, the invention provides a method and a system for predicting bird types by means of a key part prediction network.
In order to achieve the above purpose, the present invention provides the following technical solutions:
a method of predicting bird species by means of a critical site prediction network, the method comprising the steps of:
detecting a bird target frame according to target detection network, wherein the bird target frame comprises a bird target position and a bird category, the data to be detected comprises picture and video data, and the detection network comprises yolo and a master-rcnn;
training a critical part prediction model, and predicting bird critical part positions and scores in the bird target frame according to the trained critical part prediction model, wherein the bird critical parts comprise beaks, crowns, backs, eyes, wings, legs and tails;
judging the current posture of the target according to the positions and scores of the key parts of the birds and by combining the geometric relations among different parts, wherein the current posture of the target comprises a normal posture and an abnormal shielding posture;
and judging the final bird type according to the current posture of the target.
As a further technical scheme of the invention, the normal posture means that an asymmetric key part in the bird key parts is visible, and at least one side of the key point of the bird key part with symmetry is visible, wherein the asymmetric key part comprises a beak and a guan, and the bird key part with symmetry comprises eyes and wings; and when the normal gesture is not satisfied, the abnormal shielding gesture is obtained.
As a further technical scheme of the invention, when the bird key position is output by the key position prediction model, the bird key position scores are also output, when the bird key position score is larger than a preset threshold value, the current bird key position is judged to be a visible key position, and when the bird key position score is smaller than the preset threshold value, the current bird key position is judged to be an invisible key position.
As a further technical solution of the present invention, the step of determining the current pose of the target according to the position and score of the key part of the bird and by combining the geometric relations between different parts includes:
setting a key part score threshold value, and judging that the gesture is normal when the score of the asymmetric key part is larger than the threshold value and the score of at least one side of the bird key part with symmetry is larger than the threshold value;
and judging the abnormal shielding gesture when the normal gesture is not satisfied.
As a further technical solution of the present invention, the step of determining the bird category according to the current posture of the target includes:
when the current gesture of the target is a normal gesture, directly outputting the bird type detected by the target detection network as a final type;
when the current gesture of the target is an abnormal shielding gesture, determining bird types together by combining priori knowledge;
wherein the prior knowledge comprises: according to the bird species distribution condition of the protection area, the common bird types of the protection area are determined, the bird types are subdivided according to the body shape size and the key part length, and the key part length is used for estimating the beak length, neck length and foot length of the bird according to the key point position.
As a further technical scheme of the invention, when the current gesture of the target is an abnormal shielding gesture, the step of jointly determining the bird category by combining prior knowledge comprises the following steps:
when the key part scores output by the key part prediction model are smaller than the preset threshold value and smaller than the preset quantity, the prior knowledge is combined to assist in judging the final category, if the legs are short-time judged to be wild goose type, and if the body shape is small, the final category is judged to be snipe type;
when the key point scores output by the key position prediction network are smaller than the set threshold value number and larger than the preset number, the visible key positions are considered to be too few, the specific categories are not judged any more, and birds are output as final labels.
It is another object of an embodiment of the present invention to provide a system for predicting bird species by means of a critical-site prediction network, the system comprising:
the bird target frame comprises a bird target position and a bird category, wherein the data to be detected comprises pictures and video data, and the detection network comprises yolo and faster-rcnn;
the key part prediction module is used for training a key part prediction model and predicting bird key part positions and scores in the bird target frame according to the trained key part prediction model, wherein the bird key parts comprise beaks, crowns, backs, eyes, wings, legs and tails;
the key part judging module is used for judging the current gesture of the target according to the position and the score of the bird key part and combining the geometric relations among different parts, wherein the current gesture of the target comprises a normal gesture and an abnormal shielding gesture;
and the result synthesis module is used for judging the final bird category according to the current posture of the target.
As a further technical scheme of the invention, the invention further comprises a priori knowledge module: the method is used for determining common bird types in the protection area according to the bird species distribution situation in the protection area, and subdividing the bird types according to the body shape size and the key part length refers to estimating the beak length, neck length and foot length of the birds according to the key point positions.
Compared with the prior art, the invention has the beneficial effects that: the invention provides a method and a system for predicting bird types by means of a key part prediction network. The invention provides a method for improving bird detection precision under the conditions of abnormal posture, shielding and the like by means of a key part prediction network.
Drawings
FIG. 1 is a block diagram of a system for predicting bird species with a critical-site prediction network.
Detailed Description
The technical scheme of the patent is further described in detail below with reference to the specific embodiments.
As an embodiment of the present invention, a method of predicting bird species by means of a critical-site prediction network, the method comprising the steps of:
detecting a bird target frame according to target detection network, wherein the bird target frame comprises a bird target position and a bird category, the data to be detected comprises picture and video data, and the detection network comprises yolo and a master-rcnn;
training a critical part prediction model, and predicting bird critical part positions and scores in the bird target frame according to the trained critical part prediction model, wherein the bird critical parts comprise beaks, crowns, backs, eyes, wings, legs and tails;
judging the current posture of the target according to the positions and scores of the key parts of the birds and by combining the geometric relations among different parts, wherein the current posture of the target comprises a normal posture and an abnormal shielding posture;
and judging the final bird type according to the current posture of the target.
In this embodiment, when the key part prediction model is trained, a dataset is first manufactured by labeling positions of different parts of birds in a bird picture only, and then the key part prediction model is trained to obtain the key part prediction model. According to the method, firstly, a bird target frame is initially detected through a target detection network, the bird target frame comprises the position of a bird target and the bird category, then a key part prediction model predicts the position and the score of the key part of the bird in the bird target frame, the current gesture of a target is judged according to the position and the score of the key part of the bird and by combining the geometric relations among different parts, the current gesture of the target comprises a normal gesture and an abnormal shielding gesture, and finally, whether the final bird category is judged by means of priori knowledge or not is judged according to the current gesture of the target.
In the embodiment of the invention, the normal posture means that an asymmetric key part in the bird key parts is visible, and at least one side of the key point of the symmetric bird key part is visible, wherein the asymmetric key part comprises a beak and a bird crown, and the symmetric bird key part comprises eyes and wings; the abnormal shielding posture is the abnormal shielding posture when the normal posture is not satisfied (the key parts of the bird body are invisible, such as predation or the head is invisible when the head is pricked under the wings, etc.).
In the embodiment of the invention, when the bird key position is output by the key position prediction model, the bird key position scores are also output, when the bird key position score is larger than a preset threshold value, the current bird key position is judged to be a visible key position, and when the bird key position score is smaller than the preset threshold value, the current bird key position is judged to be an invisible key position. The key part prediction network has high prediction score for the bird visible key part and low prediction score for the invisible key part, and can distinguish whether the key point is visible or not by setting a threshold value, if the threshold value is 0.8, the key part is visible when the threshold value is more than 0.8, and the key part is invisible when the threshold value is less than 0.8.
In the embodiment of the invention, the step of judging the current posture of the target according to the positions and the scores of the key parts of the birds and by combining the geometric relations among different parts comprises the following steps:
setting a key part score threshold value, and judging that the gesture is normal when the score of the asymmetric key part is larger than the threshold value and the score of at least one side of the bird key part with symmetry is larger than the threshold value;
and judging the abnormal shielding gesture when the normal gesture is not satisfied.
The invention can also judge the current gesture of the target through the position relation between the key parts of the birds, such as the beak and the back, the normal condition is the misregistration of the beak and the back, the abnormal condition is the near coincidence of the beak and the back (the abnormal condition is judged when the threshold value is set, such as less than 2 pixels), and the like.
In an embodiment of the present invention, the step of determining the bird category according to the current pose of the target includes:
when the current gesture of the target is a normal gesture, directly outputting the bird type detected by the target detection network as a final type;
when the current gesture of the target is an abnormal shielding gesture, determining bird types together by combining priori knowledge;
wherein the prior knowledge comprises: according to the distribution condition of bird species in the protection area, the common bird species in the protection area are determined, and the bird species are subdivided according to the body shape size and the length of key parts, such as a crane with a larger body shape, a geranium with a smaller body shape, a sensor with a smaller body shape and the like, the length of the key parts refers to the estimation of the beak length, the neck length and the foot length of the bird according to the position of the key points, such as the general short legs of the goose, the general high body shape of the crane and the geranium, the neck, the beak length, the leg length and the like.
In the embodiment of the present invention, when the current pose of the target is an abnormal occlusion pose, the step of determining the bird category together with a priori knowledge includes:
when the key part scores output by the key part prediction model are smaller than the preset threshold value number and smaller than the preset number (the preset number is an adjustable parameter), the prior knowledge is combined to assist in judging the final category, such as the short-time judgment of the legs as the wild goose category, the small-sized judgment as the saddle category, and the like;
when the key point score output by the key position prediction network is smaller than the set threshold value number and larger than the preset number (the preset number is an adjustable parameter), the visible key position is considered to be too few, the specific category is not judged any more, and birds are output as final labels.
Another object of an embodiment of the present invention, as shown in fig. 1, is to provide a system for predicting bird species by means of a critical location prediction network, the system comprising:
the bird target frame comprises a bird target position and a bird category, wherein the data to be detected comprises pictures and video data, and the detection network comprises yolo and faster-rcnn;
the key part prediction module is used for training a key part prediction model and predicting bird key part positions and scores in the bird target frame according to the trained key part prediction model, wherein the bird key parts comprise beaks, crowns, backs, eyes, wings, legs and tails;
the key part judging module is used for judging the current gesture of the target according to the position and the score of the bird key part and combining the geometric relations among different parts, wherein the current gesture of the target comprises a normal gesture and an abnormal shielding gesture;
and the result synthesis module is used for judging the final bird category according to the current posture of the target.
In the embodiment of the invention, the method further comprises a priori knowledge module: the method is used for determining common bird types in the protection area according to the bird species distribution situation in the protection area, and subdividing the bird types according to the body shape size and the key part length, such as a crane with a large body shape, a geranium with a small body shape, a sensor with a small body shape and the like, wherein the key part length is used for estimating the beak length, the neck length and the foot length of the bird according to the key point position, such as the general short leg of the goose, the general high body shape of the crane and the geranium, the neck, the beak length, the leg length and the like.
The functions that can be achieved by the above-described method for predicting bird species with the aid of a critical-position prediction network are accomplished by a computer device comprising one or more processors and one or more memories having stored therein at least one program code that is loaded and executed by the one or more processors to perform the functions of the method for predicting bird species with the aid of a critical-position prediction network.
The processor takes out instructions from the memory one by one, analyzes the instructions, then completes corresponding operation according to the instruction requirement, generates a series of control commands, enables all parts of the computer to automatically, continuously and cooperatively act to form an organic whole, realizes the input of programs, the input of data, the operation and the output of results, and the arithmetic operation or the logic operation generated in the process is completed by the arithmetic unit; the Memory comprises a Read-Only Memory (ROM) for storing a computer program, and a protection device is arranged outside the Memory.
For example, a computer program may be split into one or more modules, one or more modules stored in memory and executed by a processor to perform the present invention. One or more of the modules may be a series of computer program instruction segments capable of performing specific functions for describing the execution of the computer program in the terminal device.
It will be appreciated by those skilled in the art that the foregoing description of the service device is merely an example and is not meant to be limiting, and may include more or fewer components than the foregoing description, or may combine certain components, or different components, such as may include input-output devices, network access devices, buses, etc.
The processor may be a central processing unit (Central Processing Unit, CPU), other general purpose processors, digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), off-the-shelf programmable gate arrays (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or the like. The general purpose processor may be a microprocessor or the processor may be any conventional processor or the like, which is the control center of the terminal device described above, and which connects the various parts of the entire user terminal using various interfaces and lines.
The memory may be used for storing computer programs and/or modules, and the processor may implement various functions of the terminal device by running or executing the computer programs and/or modules stored in the memory and invoking data stored in the memory. The memory may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function (such as an information acquisition template display function, a product information release function, etc.), and the like; the storage data area may store data created according to the use of the berth status display system (e.g., product information acquisition templates corresponding to different product types, product information required to be released by different product providers, etc.), and so on. In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as a hard disk, memory, plug-in hard disk, smart Media Card (SMC), secure Digital (SD) Card, flash Card (Flash Card), at least one disk storage device, flash memory device, or other volatile solid-state storage device.
The modules/units integrated in the terminal device may be stored in a computer readable storage medium if implemented in the form of software functional units and sold or used as separate products. Based on this understanding, the present invention may implement all or part of the modules/units in the system of the above-described embodiments, or may be implemented by instructing the relevant hardware by a computer program, which may be stored in a computer-readable storage medium, and which, when executed by a processor, may implement the functions of the respective system embodiments described above. Wherein the computer program comprises computer program code, which may be in the form of source code, object code, executable files or in some intermediate form, etc. The computer readable medium may include: any entity or device capable of carrying computer program code, a recording medium, a U disk, a removable hard disk, a magnetic disk, an optical disk, a computer Memory, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), an electrical carrier signal, a telecommunications signal, a software distribution medium, and so forth.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
The foregoing description is only of the preferred embodiments of the present invention, and is not intended to limit the scope of the invention, but rather is intended to cover any equivalents of the structures or equivalent processes disclosed herein or in the alternative, which may be employed directly or indirectly in other related arts.
Claims (8)
1. A method of predicting bird species by means of a critical site prediction network, the method comprising the steps of:
detecting a bird target frame according to target detection network, wherein the bird target frame comprises a bird target position and a bird category, the data to be detected comprises picture and video data, and the detection network comprises yolo and a master-rcnn;
training a critical part prediction model, and predicting bird critical part positions and scores in the bird target frame according to the trained critical part prediction model, wherein the bird critical parts comprise beaks, crowns, backs, eyes, wings, legs and tails;
judging the current posture of the target according to the positions and scores of the key parts of the birds and by combining the geometric relations among different parts, wherein the current posture of the target comprises a normal posture and an abnormal shielding posture;
and judging the final bird type according to the current posture of the target.
2. A method of predicting avian species by means of a critical site prediction network according to claim 1 wherein the normal pose is that an asymmetric critical site of avian critical sites, including beak and bird crown, is visible and at least one side of a critical point of avian critical sites with symmetry, including eyes and wings; and when the normal gesture is not satisfied, the abnormal shielding gesture is obtained.
3. The method according to claim 2, wherein the key part prediction model further outputs a score of each bird key part when outputting the bird key part position, determines that the current bird key part is a visible key part when the bird key part score is greater than a preset threshold, and determines that the current bird key part is an invisible key part when the bird key part score is less than the preset threshold.
4. A method of predicting bird species by means of a critical-part prediction network as claimed in claim 3 wherein said step of determining the current pose of the target based on said bird critical-part positions and scores in combination with geometric relationships between different parts comprises:
setting a key part score threshold value, and judging that the gesture is normal when the score of the asymmetric key part is larger than the threshold value and the score of at least one side of the bird key part with symmetry is larger than the threshold value;
and judging the abnormal shielding gesture when the normal gesture is not satisfied.
5. The method of predicting bird species with the aid of a critical-part prediction network of claim 4 wherein said step of determining bird species from said target current pose comprises:
when the current gesture of the target is a normal gesture, directly outputting the bird type detected by the target detection network as a final type;
when the current gesture of the target is an abnormal shielding gesture, determining bird types together by combining priori knowledge;
wherein the prior knowledge comprises: according to the bird species distribution condition of the protection area, the common bird types of the protection area are determined, the bird types are subdivided according to the body shape size and the key part length, and the key part length is used for estimating the beak length, neck length and foot length of the bird according to the key point position.
6. The method of claim 5, wherein when the current pose of the target is an abnormal occlusion pose, the step of jointly determining the bird species in combination with a priori knowledge comprises:
when the key part scores output by the key part prediction model are smaller than the preset threshold value and smaller than the preset quantity, the prior knowledge is combined to assist in judging the final category, if the legs are short-time judged to be wild goose type, and if the body shape is small, the final category is judged to be snipe type;
when the key point scores output by the key position prediction network are smaller than the set threshold value number and larger than the preset number, the visible key positions are considered to be too few, the specific categories are not judged any more, and birds are output as final labels.
7. A system for predicting bird species by means of a strategic location prediction network, said system comprising:
the bird target frame comprises a bird target position and a bird category, wherein the data to be detected comprises pictures and video data, and the detection network comprises yolo and faster-rcnn;
the key part prediction module is used for training a key part prediction model and predicting bird key part positions and scores in the bird target frame according to the trained key part prediction model, wherein the bird key parts comprise beaks, crowns, backs, eyes, wings, legs and tails;
the key part judging module is used for judging the current gesture of the target according to the position and the score of the bird key part and combining the geometric relations among different parts, wherein the current gesture of the target comprises a normal gesture and an abnormal shielding gesture;
and the result synthesis module is used for judging the final bird category according to the current posture of the target.
8. The system for predicting bird species by means of a critical-site prediction network of claim 7 further comprising a priori knowledge module: the method is used for determining common bird types in the protection area according to the bird species distribution situation in the protection area, subdividing the bird types according to the body shape and the length of the key part, wherein the length of the key part is used for estimating the beak length, the neck length and the foot length of the bird according to the position of the key point.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311846498.9A CN117809099A (en) | 2023-12-29 | 2023-12-29 | Method and system for predicting bird category by means of key part prediction network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311846498.9A CN117809099A (en) | 2023-12-29 | 2023-12-29 | Method and system for predicting bird category by means of key part prediction network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117809099A true CN117809099A (en) | 2024-04-02 |
Family
ID=90426420
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311846498.9A Pending CN117809099A (en) | 2023-12-29 | 2023-12-29 | Method and system for predicting bird category by means of key part prediction network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117809099A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112381830A (en) * | 2020-10-23 | 2021-02-19 | 山东黄河三角洲国家级自然保护区管理委员会 | Method and device for extracting bird key parts based on YCbCr superpixels and graph cut |
| CN114387499A (en) * | 2022-01-19 | 2022-04-22 | 国家海洋环境监测中心 | Island coastal wetland waterfowl identification method, distribution query system and medium |
| CN115481680A (en) * | 2022-09-02 | 2022-12-16 | 武汉大学苏州研究院 | Exogenous radar-based flying bird and unmanned aerial vehicle track target classification method and equipment |
| US20230099113A1 (en) * | 2021-09-29 | 2023-03-30 | Beijing Baidu Netcom Science Technology Co., Ltd. | Training method and apparatus for a target detection model, target detection method and apparatus, and medium |
| CN116229518A (en) * | 2023-03-17 | 2023-06-06 | 百鸟数据科技(北京)有限责任公司 | Bird species observation method and system based on machine learning |
| CN116630726A (en) * | 2023-07-26 | 2023-08-22 | 成都大熊猫繁育研究基地 | Multi-mode-based bird classification method and system |
| CN116912880A (en) * | 2023-07-13 | 2023-10-20 | 百鸟数据科技(北京)有限责任公司 | Bird recognition quality assessment method and system based on bird key point detection |
-
2023
- 2023-12-29 CN CN202311846498.9A patent/CN117809099A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112381830A (en) * | 2020-10-23 | 2021-02-19 | 山东黄河三角洲国家级自然保护区管理委员会 | Method and device for extracting bird key parts based on YCbCr superpixels and graph cut |
| US20230099113A1 (en) * | 2021-09-29 | 2023-03-30 | Beijing Baidu Netcom Science Technology Co., Ltd. | Training method and apparatus for a target detection model, target detection method and apparatus, and medium |
| CN114387499A (en) * | 2022-01-19 | 2022-04-22 | 国家海洋环境监测中心 | Island coastal wetland waterfowl identification method, distribution query system and medium |
| CN115481680A (en) * | 2022-09-02 | 2022-12-16 | 武汉大学苏州研究院 | Exogenous radar-based flying bird and unmanned aerial vehicle track target classification method and equipment |
| CN116229518A (en) * | 2023-03-17 | 2023-06-06 | 百鸟数据科技(北京)有限责任公司 | Bird species observation method and system based on machine learning |
| CN116912880A (en) * | 2023-07-13 | 2023-10-20 | 百鸟数据科技(北京)有限责任公司 | Bird recognition quality assessment method and system based on bird key point detection |
| CN116630726A (en) * | 2023-07-26 | 2023-08-22 | 成都大熊猫繁育研究基地 | Multi-mode-based bird classification method and system |
Non-Patent Citations (1)
| Title |
|---|
| 赵毅力;徐丹;: "联合语义部件的鸟类图像细粒度识别", 计算机辅助设计与图形学学报, no. 08, 15 August 2018 (2018-08-15) * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3375282B1 (en) | Method, information processing apparatus and program for determining activity quantities of animals | |
| CN109886928B (en) | Target cell marking method, device, storage medium and terminal equipment | |
| CN105144239B (en) | Image processing apparatus, image processing method | |
| US11017207B2 (en) | System for automatic tumor detection and classification | |
| CN107992903A (en) | Livestock personal identification method, device, storage medium and electronic equipment | |
| US9940545B2 (en) | Method and apparatus for detecting anatomical elements | |
| CN111709421A (en) | Bird identification method and device, computer equipment and storage medium | |
| CN112183212B (en) | Weed identification method, device, terminal equipment and readable storage medium | |
| US9442565B2 (en) | System and method for determining distracting features in a visual display | |
| CN111325217B (en) | Data processing method, device, system and medium | |
| CN109460713A (en) | Recognition methods, device and the equipment of animal childbirth | |
| CN112115803B (en) | Mask state reminding method and device and mobile terminal | |
| CN110599520B (en) | Open field experiment data analysis method, system and terminal equipment | |
| CN113269752A (en) | Image detection method, device terminal equipment and storage medium | |
| CN117809099A (en) | Method and system for predicting bird category by means of key part prediction network | |
| CN115734072A (en) | Internet of things centralized monitoring method and device for industrial automation equipment | |
| CN112819885A (en) | Animal identification method, device and equipment based on deep learning and storage medium | |
| CN112153892B (en) | Device for fly management | |
| US20200042794A1 (en) | Information processing method and information processing system | |
| CN112884761A (en) | Aquatic economic animal head identification method, terminal device and readable storage medium | |
| CN114764833A (en) | Plant growth curve determination method and device, electronic equipment and medium | |
| US20190236268A1 (en) | Behavior determining method, behavior determining apparatus, and non-transitory computer readable medium | |
| US20240144633A1 (en) | Image recognition method, electronic device and storage medium | |
| CN117830731B (en) | Multidimensional parallel scheduling method | |
| CN111625739A (en) | Positioning method and device and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |