CN109829888A - Wild animal monitoring analysis system and method based on depth convolutional neural networks - Google Patents
Wild animal monitoring analysis system and method based on depth convolutional neural networks Download PDFInfo
- Publication number
- CN109829888A CN109829888A CN201811610173.XA CN201811610173A CN109829888A CN 109829888 A CN109829888 A CN 109829888A CN 201811610173 A CN201811610173 A CN 201811610173A CN 109829888 A CN109829888 A CN 109829888A
- Authority
- CN
- China
- Prior art keywords
- wild animal
- module
- mobile object
- convolutional neural
- neural networks
- 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
Landscapes
- Image Analysis (AREA)
Abstract
The invention discloses a kind of wild animal wireless monitor analysis systems and method based on depth convolutional neural networks, by obtaining mobile object image data and establishing database, to image calibration and pre-process, in conjunction with Adam optimization algorithm training depth convolutional neural networks, carry out the detection and identification of wild animal, the time occurred to wild animal, place is for statistical analysis and establishes model, overcome large labor intensity in conventional method, time-consuming for manual analysis, and there are a large amount of redundancy and misrecognition information, it is difficult to the problem of analyzing the accurate picture for obtaining wildlife resource status and dynamic change, it can be long-range for researcher, the personal feature and Species structure situation of monitored region wild animal are obtained in real time, it improves work efficiency and precision of analysis.
Description
Technical field
The present invention relates to the protections of ecological environment and animal and plant and monitoring field, particularly relate to a kind of based on depth convolution
The wild animal wireless monitor analysis system and method for neural network.
Background technique
As a ring indispensable in ecological chain, wild animal plays important role in bio-diversity, protects
Wild animal is protected not only to the balance and stability important role for maintaining the ecosystem, is also had to the survival and development of the mankind
Important meaning.The monitoring of scientific and effective wild animal is the primary premise of resource of saving the wild animals, and this requires wild
When animal monitoring, pictorial information, the geographical location information of wild animal are obtained in real time, and obtain the region by these information
Information, quantity information and the habitat situation of wild animal show so that researcher grasps wildlife resource in time
Shape and dynamic change provide necessary guarantee for effective protection, sustainable utilization, scientific management wildlife resource.
However compared with other monitoring projects, wild animal monitoring project faces woodland environment complexity and every kind of Animal behaviour
The problems such as difference.Wild animal long-term dynamics monitor the difficult point and hot issue always studied both at home and abroad.Currently, domestic related
The research of wild animal is mainly by the way of the shooting of artificial field investigation, global positioning system necklace and infrared camera.This
A little traditional approach not only large labor intensity, time-consuming for manual analysis, and has a large amount of redundancy and misrecognition information, it is difficult to point
Analysis obtains the accurate picture of wildlife resource status and dynamic change.
On the other hand, image is caused to obtain lag, and be unfavorable for open country since network signal is weak in traditional wild animal monitoring
External environment increases the demand of node according to landform.
Summary of the invention
In view of this, it is an object of the invention to propose a kind of wild animal analysis prison based on depth convolutional neural networks
Survey analysis system and method, solve large labor intensity in traditional approach, time-consuming for manual analysis, and have a large amount of redundancy and
Misidentify information, it is difficult to the problem of analyzing the accurate picture for obtaining wildlife resource status and dynamic change.
System is analyzed based on the above-mentioned purpose wild animal wireless monitor provided by the invention based on depth convolutional neural networks
System, including power supply module, further include mobile image acquisition module, mobile image acquisition module includes several terminal nodes, right
The image data of mobile object is acquired；
Cloud service center, comprising: memory module receives the image data of mobile object and stored；
Database module establishes mobile object image data base using collected mobile object image data；
Preprocessing module, according in mobile object image animal individual feature and animal group feature to mobile object figure
It as database progress image calibration and is pre-processed, constitutes the training set in the monitoring region；
Detection module carries out wild animal type using the model in target detection network as target detection model
Calibrated training set is distinguished, model is completed to train in conjunction with Adam optimization algorithm, updates model parameter, obtains distinguishing not of the same race
The ideal model of the wild animal of class, to determine the classification and and location information of wild animal；
Analysis module, the time occurred to wild animal, place are for statistical analysis, determine the kind of the region wild animal
Group's size, feature and Migratory Regularity, obtain analysis result；
Collected mobile object image data is transferred to cloud service center, will analyzed by wireless image transmission network
As a result it is transferred to user terminal.
Preferably, the network structure of terminal node is mesh network, and mobile image acquisition module further includes several coordinations
Node, coordinator node can give out information to each terminal node, and message content includes the shooting of controlling terminal node and upper communication
Cease cloud service center.
Preferably, it is equipped with network handover module in wireless image transmission network, is believed according to the position of terminal node and 4G
Number intensity select suitable network to carry out the transmission of mobile object image data.
Preferably, it is equipped with moving Object Detection module in mobile image acquisition module, mobile object is detected, according to
Testing result controlling terminal node carries out Image Acquisition to mobile object.
Preferably, infrared compensating lamp and light detection module, light detection module root are equipped in mobile image acquisition module
Infrared compensating lamp work is controlled according to light luminance.
Preferably, power supply module uses wind light mutual complementing power generation module, and wind light mutual complementing power generation module includes power analysis module
And battery, power analysis module adjust the work shape of battery according to current intensity of sunshine, wind scale and payload size
State.
The present invention also provides a kind of wild animal wireless monitoring methods based on depth convolutional neural networks, including following step
It is rapid:
A. the mobile object image data of terminal node acquisition is obtained by wireless network；
B. mobile object image data base is established using collected mobile object data image；
C. according in mobile object image animal individual feature and animal group feature to mobile object image data base
It carries out image calibration and is pre-processed, constitute the training set in the monitoring region；
D. using the object module in target detection network, by distinguishing calibrated training to wild animal type
Collection is completed to train, be updated to model, obtains the reason for distinguishing variety classes wild animal in conjunction with Adam optimization algorithm to model
Model is thought, using updated model as detection model, to determine the classification and and location information of wild animal；
E. for statistical analysis to the time of wild animal appearance, place, determine the Population Size of the region wild animal,
Feature and Migratory Regularity obtain analysis result.
F. analysis result is transferred to user terminal.
Preferably, the network structure of terminal node is mesh network, is taken a step forward in step a including according to terminal node
The intensity of position and node signal selects suitable network node.
Preferably, it takes a step forward in step a including passing through object in background subtraction or optical flow method judgement monitoring region
It is mobile, acquisition identification message is issued to terminal node.
Preferably, periodically database is updated and is supplemented.
From the above it can be seen that the wild animal wireless monitor provided by the invention based on depth convolutional neural networks
Analysis system and method to image calibration and are pre-processed, are tied by obtaining mobile object image data and establishing database
Adam optimization algorithm training depth convolutional neural networks are closed, the detection and identification of wild animal is carried out, occurs to wild animal
Time, place is for statistical analysis and establishes model, overcomes large labor intensity in conventional method, and time-consuming for manual analysis, and
There are a large amount of redundancy and misrecognition information, it is difficult to which analysis obtains the accurate picture of wildlife resource status and dynamic change
The problem of, long-range for researcher, real-time it can obtain the personal feature and Species structure feelings of monitored region wild animal
Condition, improves work efficiency and precision of analysis；
On the other hand, terminal node network structure is set as mesh network, has the features such as wireless Ad Hoc, multi-hop, has
Increase node demand conducive to field environment is met according to landform, reduces the long-range delay and packet loss problem for obtaining image, pass through
The cooperation of light detection module and infrared compensating lamp overcomes asking for rainy weather and nighttime image discrimination difference in conventional method
Topic, further increases accuracy rate.
Detailed description of the invention
Fig. 1 is the modular structure schematic diagram of wild animal monitoring analysis system provided by the invention；
Fig. 2 is the terminal node mesh schematic network structure of wild animal monitoring analysis system provided by the invention；
Fig. 3 is the flow diagram of wild animal method for monitoring and analyzing provided by the invention；
Specific embodiment
To make the objectives, technical solutions, and advantages of the present invention clearer, below in conjunction with specific embodiment, and reference
Attached drawing, the present invention is described in more detail.
For wild animal wireless monitor analysis system labor intensive and time, the low problem of result accuracy rate is analyzed, this
The first aspect of inventive embodiments provides a kind of wild animal wireless monitor analysis system based on depth convolutional neural networks
System mainly includes following part, and the relationship between each section is as shown in Figure 1:
Mobile image acquisition module, including several terminal nodes are separately positioned in monitoring region, to mobile object
Image data is acquired；
Optionally, the network structure of terminal node is set as mesh network, has the characteristics that wireless Ad Hoc, multi-hop, can
To adapt to dispose the complexity of monitoring node in field environment, can be in communication with each other between each terminal node, mobile image
It is additionally provided with several coordinator nodes in acquisition module, as shown in Fig. 2, coordinator node is deployed in geographical location with respect to high point, passes through
Hair is worked as by infrared triggering system identification mobile object to institute's monitoring range all standing using high-resolution video capture device
Existing mobile object passes through monitored region, issues acquisition identification message, each terminal node to the terminal node in the region immediately
It can also be in communication with each other between coordinator node；
Optionally, infrared compensating lamp and light detection module, the inspection of light detection module are equipped in mobile image acquisition module
Survey current light value, when illumination value be lower than fixed threshold when, open infrared compensating lamp, to adapt to monitoring range dark and
Night environment；
Cloud service center, comprising: memory module is arranged at each terminal node, to the collected movement of terminal node
Object image data is locally stored；
Database module establishes mobile object image data base using collected mobile object data image, it is preferable that
Periodically the content of database is updated and is supplemented in conjunction with the image data of acquisition, it is accurate with the identification for improving wild animal
Rate, it is ensured that detecting different wild animals to different zones has good adaptability；
Preprocessing module, according in mobile object image animal individual feature and animal group feature to mobile object figure
It as database progress image calibration and is pre-processed, specifically, animal individual feature includes hair color, relative in region
Size and movement speed of fixed reference etc., animal group feature include that size of animal in group, the individual between animal are big
Small scale and group's movement law etc. distinguish calibration and pretreatment to wild animal and non-animal, to different types of open country
Lively object distinguishes calibration and pretreatment, constitutes the training set in the monitoring region.
Detection module, using the model in target detection network as target detection model, optionally, using in YOLOv3
Parameter initial weight is the model of YOLOv3-416 as target detection model, by distinguishing calibration to wild animal type
Training set afterwards is completed to train, updates model parameter, obtain distinguishing different types of wild in conjunction with Adam optimization algorithm to model
The ideal model of animal compares the wild animal image in ideal model and mobile object image to determine the classification of wild animal
With and location information；
Analysis module, using the data of training set, the place of the time and terminal node that obtain in conjunction with image are obtained wild
Information, quantity information and the habitat situation of animal, determine the Population Size of the region wild animal, feature and migrate rule
Rule, obtains analysis result；
Collected mobile object image data is transferred to database module by wireless image transmission network, and analysis is tied
Fruit is transferred to user terminal, it is preferable that network handover module is equipped in wireless image transmission network, according to the position of each terminal node
It sets and transmission that the intensity of node signal selects suitable network to carry out mobile object image data, when needing to upload picture number
According to meshed network intensity it is weak when, other terminal nodes of Automatic-searching complete image data upload function as relay node；
Power supply module provides electric power for the electrical equipment in system, it is preferable that wind light mutual complementing power generation module is used, including
Power analysis module and battery, power analysis module are stored according to current intensity of sunshine, wind scale and payload size, adjustment
The working condition of battery guarantees the permanently effective supply of systematic electricity.
For wild animal wireless monitor analysis system labor intensive and time, the low problem of result accuracy rate is analyzed, this
The second aspect of inventive embodiments, propose it is a kind of raising working efficiency and analyze accuracy rate based on depth convolutional Neural net
One embodiment of the wild animal wireless monitor analysis method of network, as described in Figure 1, be wild animal provided by the invention without
The flow diagram of one embodiment of line method for monitoring and analyzing.
The wild animal wireless monitor analysis method, as shown in Figure 3, comprising:
Step a. obtains the mobile object image data of terminal node acquisition by wireless network；
Optionally, by infrared triggering system detection mobile object image, before background subtraction or optical flow method comparison
Image frame data afterwards, it is determined whether there is mobile object to enter institute's detection zone, if it is determined that false triggering then without operation, when
Determining has mobile object to enter monitoring region, starts video record and stores into terminal node, terminal node is avoided to acquire
The redundancy for being largely free of mobile object image is generated when image, in the data for obtaining terminal node by wireless network
When, it selects suitable network to carry out image transmitting according to the intensity of node location and node signal, reduces image transmitting process
The delay and switch of middle generation.
Step b. establishes mobile object image data base using collected mobile object data image；
Optionally, database is updated and is supplemented every year, to improve the recognition accuracy of wild animal, it is ensured that not
There is good adaptability with region detection difference wild animal.
Step c. according in mobile object image animal individual feature and animal group feature to mobile object picture number
Image calibration is carried out according to library and is pre-processed, specifically, animal individual feature includes hair color, relative to fixed in region
Size and movement speed of object of reference etc., animal group feature include the size of animal in group, the ratio of the Individual Size between animal
Example and group's movement law etc., distinguish calibration and pretreatment to wild animal and non-animal, constitute the instruction in the monitoring region
Practice collection, provides foundation for further identification wild animal image.
Step d. is optionally using parameter initial weight in YOLOv3 using the model in target detection network
Original Image Adjusting is point by distinguishing the training set of calibration to wild animal type by the model of YOLOv3-416
Resolution 416*416, port number are 3 image, and will adjust image in different resolution as the input of feature extraction network, the model
Using Darknet-53 as feature extraction network, this feature network is formed using residual error folded structures, to reduce due to network
The more bring gradient problems of the number of plies reduce parameter amount, reduce trained difficulty, obtain the ideal mould of different types of wild animal
Type.The wild animal different levels feature being input in model can be effectively extracted by character network.Then, not by 3
YOLO characteristic spectrum with size predicted, is realized and is predicted multiple dimensioned wild animal, determine wild animal classification and
And location information.When training, using cross entropy as loss function, use Logistic model as classifier.
Step e. utilizes the data of training set, the place of the time and terminal node that obtain in conjunction with image, obtains wild dynamic
Information, quantity information and the habitat situation of object, determine the Population Size of the region wild animal, feature and migrate rule
Rule, obtains analysis result；
Step f. will analyze result by wireless network transmissions to user terminal, check analysis result for user.
Specifically, being illustrated so that red deer detects as an example to the above method:
1. collecting the mobile image including red deer image by infrared triggering system detection mobile object image, building
The vertical moving image data library including red deer image.
2. the feature skeleton coordinate position of red deer is fixed through rower in pair database, and it is red deer that animal category, which is arranged,.
3. being trained using the red deer picture that target detection model completes calibration, will obtain that weight file is trained to pass through
Cloud server downloads to each detection terminal.
4. detecting terminal is loaded into target detection model, when there is infrared triggering to generate, occur by target detection model inspection
The similarity degree of animal and red deer, the picture by similarity degree greater than 90% carries out local preservation, and is added currently to the picture
The geographical location information of node.
5. returning current shooting photo to cloud by wireless sensor network, cloud system creates current month data
Collection, data set before being different from, the data set were compiled by 1 year by the expansion as legacy data collection, as next year
Training set promotes the individual identification ability to red deer by continuous iteration.
6. the content in pair collected red deer data set first progress chronological classification, determines the red deer activity time, then
Place classification is carried out, determines activity space range.It is that red deer is for statistical analysis to label, determines the pass of Species structure
System.
7. all about red deer analysis as a result, eventually show in the user interface.
It should be understood by those ordinary skilled in the art that: the discussion of any of the above embodiment is exemplary only, not
It is intended to imply that the scope of the present disclosure (including claim) is limited to these examples；Under thinking of the invention, above embodiments
Or can also be combined between the technical characteristic in different embodiments, step can be realized with random order, and be existed such as
Many other variations of the upper different aspect of the invention, for simplicity, they are not provided in details.
In addition, to simplify explanation and discussing, and in order not to obscure the invention, it can in provided attached drawing
It is connect with showing or can not show with the well known power ground of integrated circuit (IC) chip and other components.Furthermore, it is possible to
Device is shown in block diagram form, to avoid obscuring the invention, and this has also contemplated following facts, i.e., about this
The details of the embodiment of a little block diagram arrangements be height depend on will implementing platform of the invention (that is, these details should
It is completely within the scope of the understanding of those skilled in the art).Elaborating that detail (for example, circuit) is of the invention to describe
In the case where exemplary embodiment, it will be apparent to those skilled in the art that can be in these no details
In the case where or implement the present invention in the case that these details change.Therefore, these descriptions should be considered as explanation
Property rather than it is restrictive.
Although having been incorporated with specific embodiments of the present invention, invention has been described, according to retouching for front
It states, many replacements of these embodiments, modifications and variations will be apparent for those of ordinary skills.Example
Such as, discussed embodiment can be used in other memory architectures (for example, dynamic ram (DRAM)).
The embodiment of the present invention be intended to cover fall into all such replacements within the broad range of appended claims,
Modifications and variations.Therefore, all within the spirits and principles of the present invention, any omission, modification, equivalent replacement, the improvement made
Deng should all be included in the protection scope of the present invention.
Claims (10)
1. the wild animal wireless monitor analysis system based on depth convolutional neural networks, including power supply module, it is characterised in that:
It further include mobile image acquisition module, the mobile image acquisition module includes several terminal nodes, to the figure of mobile object
As data are acquired；
Cloud service center, comprising: memory module receives the image data of mobile object and stored；
Database module establishes mobile object image data base using collected mobile object image data；
Preprocessing module, according in mobile object image animal individual feature and animal group feature to mobile object picture number
Image calibration is carried out according to library and is pre-processed, and the training set in the monitoring region is constituted；
Detection module distinguishes wild animal type using the model in target detection network as target detection model
Calibrated training set is completed to train, updates model parameter, obtain distinguishing different types of in conjunction with Adam optimization algorithm to model
The ideal model of wild animal, to determine the classification and and location information of wild animal；
Analysis module, the time occurred to wild animal, place are for statistical analysis, determine that the population of the region wild animal is big
Small, feature and Migratory Regularity obtain analysis result；
Collected mobile object image data is transferred to cloud service center, will analyze result by wireless image transmission network
It is transferred to user terminal.
2. the wild animal wireless monitor analysis system according to claim 1 based on depth convolutional neural networks, special
Sign is: the network structure of the terminal node is mesh network, and the mobile image acquisition module further includes several coordinations
Node, coordinator node can give out information to each terminal node, and message content includes the shooting of controlling terminal node and upper communication
Cease cloud service center.
3. the wild animal wireless monitor analysis system according to claim 1 based on depth convolutional neural networks, special
Sign is: being equipped with network handover module in the wireless image transmission network, is believed according to the position of the terminal node and 4G
Number intensity select suitable network to carry out the transmission of mobile object image data.
4. the wild animal wireless monitor analysis system according to claim 1 based on depth convolutional neural networks, special
Sign is: being equipped with moving Object Detection module in the mobile image acquisition module, detects to mobile object, according to detection
As a result controlling terminal node carries out Image Acquisition to mobile object.
5. the wild animal wireless monitor analysis system according to claim 1 based on depth convolutional neural networks, special
Sign is: infrared compensating lamp and light detection module is equipped in the mobile image acquisition module, light detection module is according to light
Line brightness controls infrared compensating lamp work.
6. the wild animal wireless monitor analysis system according to claim 1 based on depth convolutional neural networks, special
Sign is: the power supply module uses wind light mutual complementing power generation module, and wind light mutual complementing power generation module includes power analysis module and storage
Battery, power analysis module adjust the working condition of battery according to current intensity of sunshine, wind scale and payload size.
7. the wild animal wireless monitoring method based on depth convolutional neural networks, which comprises the following steps:
A. the mobile object image data of terminal node acquisition is obtained by wireless network；
B. mobile object image data base is established using collected mobile object data image；
C. according in mobile object image animal individual feature and animal group feature to mobile object image data base carry out
Image calibration is simultaneously pre-processed, and the training set in the monitoring region is constituted；
D. using the object module in target detection network, by distinguishing calibrated training set to wild animal type,
Model is completed to train in conjunction with Adam optimization algorithm, model is updated, obtains the ideal for distinguishing variety classes wild animal
Model, using updated model as detection model, to determine the classification and and location information of wild animal；
E. for statistical analysis to the time of wild animal appearance, place, determine the Population Size of the region wild animal, feature
And Migratory Regularity, obtain analysis result.
F. analysis result is transferred to user terminal.
8. the wild animal wireless monitoring method according to claim 1 based on depth convolutional neural networks, feature exist
Be mesh network in: the network structure of the terminal node, take a step forward including according to the position of terminal node in step a and
The intensity of node signal selects suitable network node.
9. the wild animal wireless monitoring method according to claim 1 based on depth convolutional neural networks, feature exist
In: it takes a step forward in step a including by the movement of object in background subtraction or optical flow method judgement monitoring region, to terminal
Node issues acquisition identification message.
10. the wild animal wireless monitoring method according to claim 1 based on depth convolutional neural networks, feature exist
In: periodically the database is updated and is supplemented.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811610173.XA CN109829888A (en) | 2018-12-27 | 2018-12-27 | Wild animal monitoring analysis system and method based on depth convolutional neural networks |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811610173.XA CN109829888A (en) | 2018-12-27 | 2018-12-27 | Wild animal monitoring analysis system and method based on depth convolutional neural networks |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109829888A true CN109829888A (en) | 2019-05-31 |
Family
ID=66861301
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811610173.XA Pending CN109829888A (en) | 2018-12-27 | 2018-12-27 | Wild animal monitoring analysis system and method based on depth convolutional neural networks |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109829888A (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110672162A (en) * | 2019-12-04 | 2020-01-10 | 中科星图(深圳)数字技术产业研发中心有限公司 | Animal environment monitoring method for natural protected area |
| CN111835834A (en) * | 2020-06-24 | 2020-10-27 | 清科优能(深圳)技术有限公司 | High-granularity data extraction and storage device of intelligent Internet of things terminal considering communication abnormity |
| CN112069972A (en) * | 2020-09-01 | 2020-12-11 | 安徽天立泰科技股份有限公司 | Artificial intelligence-based ounce recognition algorithm and recognition monitoring platform |
| CN112418223A (en) * | 2020-12-11 | 2021-02-26 | 互助土族自治县北山林场 | Wild animal image significance target detection method based on improved optimization |
| CN112579802A (en) * | 2020-10-28 | 2021-03-30 | 深圳市农产品质量安全检验检测中心(深圳市动物疫病预防控制中心) | Agricultural product type model base establishing method |
| CN112614148A (en) * | 2020-12-11 | 2021-04-06 | 北京林业大学 | Wild animal monitoring system based on edge cloud intelligent framework |
| CN112906511A (en) * | 2021-02-02 | 2021-06-04 | 东北林业大学 | Wild animal intelligent monitoring method combining individual image and footprint image |
| CN113435425A (en) * | 2021-08-26 | 2021-09-24 | 绵阳职业技术学院 | Wild animal emergence and emergence detection method based on recursive multi-feature fusion |
| CN113531423A (en) * | 2021-07-13 | 2021-10-22 | 读书郎教育科技有限公司 | Interactive intelligent projection table lamp and method |
| CN113591707A (en) * | 2021-07-30 | 2021-11-02 | 浙江兆晟科技股份有限公司 | Group hunting method based on thermal imaging search |
| CN114332849A (en) * | 2022-03-16 | 2022-04-12 | 科大天工智能装备技术（天津）有限公司 | Crop growth state combined monitoring method and device and storage medium |
| CN114743128A (en) * | 2022-03-09 | 2022-07-12 | 华侨大学 | Multimode northeast tiger re-identification method and device based on heterogeneous neural network |
| CN117765480A (en) * | 2024-02-20 | 2024-03-26 | 天科院环境科技发展（天津）有限公司 | Method and system for early warning migration of wild animals along road |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102186228A (en) * | 2011-04-27 | 2011-09-14 | 南京农业大学 | Wireless multimedia sensor network-based system and method for monitoring information of livestock and poultry facility welfare breeding environment |
| CN102547247A (en) * | 2012-01-31 | 2012-07-04 | 天津大学 | Field intelligent monitoring system based on long-distance wireless mesh network |
| CN102608571A (en) * | 2012-03-08 | 2012-07-25 | 西北大学 | Target location method for wireless sensor network for wild animals |
| CN103561242A (en) * | 2013-11-14 | 2014-02-05 | 北京林业大学 | Wild animal monitoring system based on wireless image sensor network |
| CN106897716A (en) * | 2017-04-27 | 2017-06-27 | 广东工业大学 | A kind of dormitory safety monitoring system and method |
| CN106934451A (en) * | 2015-12-29 | 2017-07-07 | 扬州大学 | A kind of BP neural network WSN forest fire crime prevention systems based on ant group optimization |
| CN107302654A (en) * | 2016-04-15 | 2017-10-27 | 中兴通讯股份有限公司 | A kind of image pickup method and device |
| US20170311574A1 (en) * | 2015-03-13 | 2017-11-02 | Michael W. Swan | Animal movement mapping and movement prediction method and device |
-
2018
- 2018-12-27 CN CN201811610173.XA patent/CN109829888A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102186228A (en) * | 2011-04-27 | 2011-09-14 | 南京农业大学 | Wireless multimedia sensor network-based system and method for monitoring information of livestock and poultry facility welfare breeding environment |
| CN102547247A (en) * | 2012-01-31 | 2012-07-04 | 天津大学 | Field intelligent monitoring system based on long-distance wireless mesh network |
| CN102608571A (en) * | 2012-03-08 | 2012-07-25 | 西北大学 | Target location method for wireless sensor network for wild animals |
| CN103561242A (en) * | 2013-11-14 | 2014-02-05 | 北京林业大学 | Wild animal monitoring system based on wireless image sensor network |
| US20170311574A1 (en) * | 2015-03-13 | 2017-11-02 | Michael W. Swan | Animal movement mapping and movement prediction method and device |
| CN106934451A (en) * | 2015-12-29 | 2017-07-07 | 扬州大学 | A kind of BP neural network WSN forest fire crime prevention systems based on ant group optimization |
| CN107302654A (en) * | 2016-04-15 | 2017-10-27 | 中兴通讯股份有限公司 | A kind of image pickup method and device |
| CN106897716A (en) * | 2017-04-27 | 2017-06-27 | 广东工业大学 | A kind of dormitory safety monitoring system and method |
Non-Patent Citations (4)
| Title |
|---|
| 刘文定， 李安琪，张军国，谢将剑，鲍伟东: "基于ROI-CNN的赛罕乌拉国家级自然保护区陆生野生动物自动识别", 《北京林业大学学报》 * |
| 赵谦，侯媛彬，郑茂全: "《智能视频图像处理技术与应用》", 30 September 2016 * |
| 陈善安，胡春鹤，王远，张军国: "基于无线传感器网络的野生动物图像监测***设计", 《现代制造技术与装备》 * |
| 陈家旭: "《中医诊断学研究》", 30 April 2008 * |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110672162B (en) * | 2019-12-04 | 2020-04-10 | 中科星图(深圳)数字技术产业研发中心有限公司 | Animal environment monitoring method for natural protected area |
| CN110672162A (en) * | 2019-12-04 | 2020-01-10 | 中科星图(深圳)数字技术产业研发中心有限公司 | Animal environment monitoring method for natural protected area |
| CN111835834B (en) * | 2020-06-24 | 2023-03-21 | 清科优能(深圳)技术有限公司 | High-granularity data extraction and storage device of intelligent Internet of things terminal considering communication abnormity |
| CN111835834A (en) * | 2020-06-24 | 2020-10-27 | 清科优能(深圳)技术有限公司 | High-granularity data extraction and storage device of intelligent Internet of things terminal considering communication abnormity |
| CN112069972A (en) * | 2020-09-01 | 2020-12-11 | 安徽天立泰科技股份有限公司 | Artificial intelligence-based ounce recognition algorithm and recognition monitoring platform |
| CN112579802A (en) * | 2020-10-28 | 2021-03-30 | 深圳市农产品质量安全检验检测中心(深圳市动物疫病预防控制中心) | Agricultural product type model base establishing method |
| CN112418223A (en) * | 2020-12-11 | 2021-02-26 | 互助土族自治县北山林场 | Wild animal image significance target detection method based on improved optimization |
| CN112614148A (en) * | 2020-12-11 | 2021-04-06 | 北京林业大学 | Wild animal monitoring system based on edge cloud intelligent framework |
| CN112614148B (en) * | 2020-12-11 | 2023-07-21 | 北京林业大学 | Wild animal monitoring system based on cloud intelligent architecture |
| CN112906511A (en) * | 2021-02-02 | 2021-06-04 | 东北林业大学 | Wild animal intelligent monitoring method combining individual image and footprint image |
| CN113531423A (en) * | 2021-07-13 | 2021-10-22 | 读书郎教育科技有限公司 | Interactive intelligent projection table lamp and method |
| CN113591707A (en) * | 2021-07-30 | 2021-11-02 | 浙江兆晟科技股份有限公司 | Group hunting method based on thermal imaging search |
| CN113435425A (en) * | 2021-08-26 | 2021-09-24 | 绵阳职业技术学院 | Wild animal emergence and emergence detection method based on recursive multi-feature fusion |
| CN113435425B (en) * | 2021-08-26 | 2021-12-07 | 绵阳职业技术学院 | Wild animal emergence and emergence detection method based on recursive multi-feature fusion |
| CN114743128A (en) * | 2022-03-09 | 2022-07-12 | 华侨大学 | Multimode northeast tiger re-identification method and device based on heterogeneous neural network |
| CN114332849B (en) * | 2022-03-16 | 2022-08-16 | 科大天工智能装备技术（天津）有限公司 | Crop growth state combined monitoring method and device and storage medium |
| CN114332849A (en) * | 2022-03-16 | 2022-04-12 | 科大天工智能装备技术（天津）有限公司 | Crop growth state combined monitoring method and device and storage medium |
| CN117765480A (en) * | 2024-02-20 | 2024-03-26 | 天科院环境科技发展（天津）有限公司 | Method and system for early warning migration of wild animals along road |
| CN117765480B (en) * | 2024-02-20 | 2024-05-10 | 天科院环境科技发展（天津）有限公司 | Method and system for early warning migration of wild animals along road |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109829888A (en) | Wild animal monitoring analysis system and method based on depth convolutional neural networks | |
| KR102300799B1 (en) | Crop state judgment apparatus and crop cultivation information service platform including the same | |
| CN111582234B (en) | Large-scale oil tea tree forest fruit intelligent detection and counting method based on UAV and deep learning | |
| CN112990262B (en) | Integrated solution system for monitoring and intelligent decision of grassland ecological data | |
| CN112422783A (en) | Unmanned aerial vehicle intelligent patrol system based on parking apron cluster | |
| CN108900576A (en) | A kind of intelligence community application service system based on 5G | |
| CN108710126A (en) | Automation detection expulsion goal approach and its system | |
| CN110135476A (en) | A kind of detection method of personal safety equipment, device, equipment and system | |
| CN104463168B (en) | A kind of useless house site information automation extraction method of sky based on remote sensing image | |
| CN109711348A (en) | Intelligent monitoring method and system based on the long-term real-time architecture against regulations in hollow panel | |
| CN109325495A (en) | A kind of crop image segmentation system and method based on deep neural network modeling | |
| Marín et al. | Urban lawn monitoring in smart city environments | |
| CN106776675A (en) | A kind of plants identification method and system based on mobile terminal | |
| CN110321774B (en) | Crop disaster situation evaluation method, device, equipment and computer readable storage medium | |
| CN109034177A (en) | A kind of mobile intelligent terminal indoor and outdoor recognition methods | |
| CN109241902A (en) | A kind of landslide detection method based on multi-scale feature fusion | |
| CN112613438A (en) | Portable online citrus yield measuring instrument | |
| CN114912707A (en) | Air quality prediction system and method based on multi-mode fusion | |
| CN114898238A (en) | Wild animal remote sensing identification method and device | |
| CN110532918A (en) | Method is determined based on the offshore wind electric field time-space attribute of time series remote sensing image | |
| CN116846059A (en) | Edge detection system for power grid inspection and monitoring | |
| CN110517440A (en) | Intelligent monitoring early warning system and method based on satellite remote sensing system | |
| CN117197595A (en) | Fruit tree growth period identification method, device and management platform based on edge calculation | |
| CN112509285A (en) | CNN-based global typhoon message collection method and system | |
| Yan et al. | A farmland-microclimate monitoring system based on the internet of things |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190531 |
|
| RJ01 | Rejection of invention patent application after publication |