CN111447410A - Dog state identification monitoring system and method - Google Patents
Dog state identification monitoring system and method Download PDFInfo
- Publication number
- CN111447410A CN111447410A CN202010214069.XA CN202010214069A CN111447410A CN 111447410 A CN111447410 A CN 111447410A CN 202010214069 A CN202010214069 A CN 202010214069A CN 111447410 A CN111447410 A CN 111447410A
- Authority
- CN
- China
- Prior art keywords
- dog
- state
- monitoring
- data
- alarm
- 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
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
- H04N7/181—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast for receiving images from a plurality of remote sources
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B13/00—Burglar, theft or intruder alarms
- G08B13/18—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength
- G08B13/189—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems
- G08B13/194—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems
- G08B13/196—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems using television cameras
- G08B13/19602—Image analysis to detect motion of the intruder, e.g. by frame subtraction
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Multimedia (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Computational Linguistics (AREA)
- Human Computer Interaction (AREA)
- Evolutionary Biology (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Biophysics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Signal Processing (AREA)
- Alarm Systems (AREA)
Abstract
The invention discloses a dog state identification monitoring system which comprises a plurality of monitoring alarm nodes, wherein each monitoring alarm node comprises a high-definition camera for shooting image data, a voice loudspeaker for sending an alarm voice signal and a control panel; the control panel drives and controls the voice loudspeaker to send out an alarm signal according to the state data of the dog. The invention collects and identifies the type and the dog-restraining state of the dog through the nodes arranged in public places so as to give monitoring alarm and send data to the client of the supervision personnel in a cloud mode, thereby effectively monitoring and alarming the dog state in some places, and facilitating the management and monitoring of dogs, the voice alarm reminding and the monitoring and management of dog ropes which are not tied.
Description
Technical Field
The invention relates to the field of computer identification control, in particular to a dog state identification alarm system and method.
Background
In the existing life, the phenomenon of uncertainties such as dog leasing, feeding of a strong large dog and the like often occurs, the dog is likely to attack citizens under the condition of no dog leasing, and especially children and old people with weak resistance have potential safety hazards to the children and the old people. Every year, there is news that citizens are bitten by dogs not drawn according to regulations in public places, and the safety of citizens, especially children and old people, is seriously threatened. Furthermore, the method is simple. Therefore, in some public places, parks and the like, a dog state monitoring and identifying system is needed to monitor, identify and alarm the state of the dog.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides a dog state identification and monitoring system and a dog state identification and monitoring method, which are used for identifying dogs, identifying the state of a dog, monitoring and alarming and the like and identifying strong dog species or dogs without pulling ropes.
In order to achieve the purpose, the invention adopts the technical scheme that: the utility model provides a dog state identification monitored control system which characterized in that: the monitoring alarm system comprises a plurality of monitoring alarm nodes, wherein each monitoring alarm node comprises a high-definition camera for shooting image data, a voice loudspeaker for sending an alarm voice signal and a control panel, the high-definition camera is connected with the control panel and used for sending the collected video image data to the control panel, and the control panel is used for identifying the state of a dog in an image; the control panel drives and controls the voice loudspeaker to send out an alarm signal according to the state data of the dog.
The control panel is connected with the cloud data platform through a network and used for sending the identified dog state data and the position and time of the monitoring alarm node to the cloud data platform for monitoring, storing and forwarding.
The cloud data platform is connected with a client of a monitoring manager and used for acquiring dog state information stored by the cloud data platform and corresponding position and time information.
The client comprises a computer terminal and a mobile handheld terminal.
The control panel comprises a control unit for data processing and voice loudspeaker drive control and a communication unit for network communication; the control unit is connected with the cloud data platform through the communication unit.
And the control unit pre-calibrates and trains a neural network model for identifying the state of the dog to identify and judge the dog restraining state in the image.
The neural network model is used for identifying the type of the dog and the dog restraining state.
A monitoring method of a dog state identification monitoring system comprises the following steps: step 1: acquiring video image data through a high-definition camera and sending the data to a development board;
step 2: the neural network model after training and calibration is preset in the development board and is used for identifying the canine type of a canine and/or the canine status of the canine in a video image;
and step 3: and driving a voice loudspeaker to give an alarm prompt according to the recognized dog type and/or the dog-restraining state of the dogs.
The monitoring method further comprises the following step 4: the development board sends the image data and the recognized state data to the cloud data platform, and monitoring personnel access the cloud data platform through the client to obtain the monitoring data.
In step 2, the neural network model which is preset in the development board and calibrated by training adopts a convolutional neural network, and the neural network model is trained through a preset data set and then used for identifying and outputting the dog state and the dog type through the neural network model.
The invention has the advantages that: thereby through the node that sets up in public place gather and discern the kind that the dog was only and restraint the dog state and give the monitoring alarm and send data to supervisory personnel's customer end through high in the clouds mode, can effectively monitor the alarm to the dog state in some places, conveniently to the management control of dog and to not tying the dog rope carry out voice alarm and remind and monitor the management, avoid or reduced the attack incident that the dog probably takes place under the dog state of not tying.
Drawings
The contents of the expressions in the various figures of the present specification and the labels in the figures are briefly described as follows:
FIG. 1 is a schematic diagram of an identification monitoring system according to the present invention;
FIG. 2 is a flow chart of a monitoring method of the present invention.
Detailed Description
The following description of preferred embodiments of the invention will be made in further detail with reference to the accompanying drawings.
The application mainly aims to remind a dog owner of tying a dog rope in an alarm reminding mode and facilitate management of public management personnel by identifying whether a pet dog in a public place is in a tying state or not and identifying the type of the pet dog. The mode is very suitable for monitoring the state of the dog when the dog is stroked in public places such as parks and communities and alarming to remind a manager in the further expanded community to monitor and manage.
As shown in fig. 1, a dog state identification monitoring system comprises a plurality of monitoring alarm nodes, wherein the monitoring alarm nodes are arranged in a community, a park and other occasions needing dog rope state identification. The installation position can be selected to be installed at the trunk, telegraph pole and other positions of a community or a park, and can also be installed on lawn sounds in the park (lawn sounds used for playing music and the like on lawns of a plurality of park communities), so that the monitoring and alarming of each monitoring node are completed.
The control panel is realized by adopting a paddlePi or a UP2 development panel, the paddlePi is used as an embedded development panel, the cost is low and the paddlePi is stable, a tinyYO L O target detection model and a mobilenet classification model supported by the paddlePi are trained by using hundreds of aids, after model conversion, quantization and compilation, a K-Flash is adopted to be burnt into the paddlePi, so that the calibration and training work of the model development panel are completed, the trained development panel is applied to the reality, the development panel judges whether a dog target and a dog target exist in a video according to the video data collected by the camera and determines the dog state, only the type of the developed dog is recognized on the basis, the main function of the development panel is to recognize the dog target and determine the dog state, and the dog binding data is recognized by using a network bolt, and the model is recognized by using a network for recognizing the dog leash type and/or recognizing the dog target by using a network.
The development board is actually a PCB board integrating a control chip, a communication unit and a serial port circuit, the PCB board is used for realizing control and data processing, a serial port arranged on the development board can be connected with a high-definition camera, and the control chip can communicate with an external device through an integrated communication chip (such as a 2G chip). After the neural network model is trained in advance, the control chip identifies the required dog state (the dog restraining state and the dog type) through the two neural networks. The control unit drives the voice alarm loudspeaker to send out a voice reminding signal through the driving circuit, and when the remote control recognizes that the pet dog is not tied, the voice loudspeaker gives out voice reminding such as voice 'please tie the pet dog to a dog rope' and the like, so that only a dog owner beside the unbundled dog is reminded, and the purpose of reminding the dog owner to tie the pet dog rope is achieved.
Because this application is applicable to public places such as district, park, for the convenience of the managers' management in public places, the control chip of control panel passes through on-board integrated 2G communication chip and high in the clouds data platform communication, be used for behind drive control audio alert with video data of shooing and the identification data (dog class variety, whether tie the dog rope) of dog only data transmission to high in the clouds data platform, high in the clouds data platform is as a high in the clouds server, the monitoring managers visits the alarm data that the high in the clouds data platform was known in real time through the mode of customer end, make things convenient for managers in time to the field processing warning dog owner and tie the dog rope. The uploaded data refers to that after the dog is not tied, the control unit sends out an alarm signal and sends the state data of the dog at the moment and the position and time of the monitoring alarm node to the cloud data platform for monitoring, storing and forwarding. Management personnel access the server through a mobile phone client, a computer client and the like to obtain the dog state information and the corresponding position and time information in the platform, so that the on-site processing and warning are conveniently and timely carried out.
A monitoring method of a dog state identification monitoring system comprises the following steps:
step 1: acquiring video image data through a high-definition camera and sending the data to a development board;
step 2, presetting a training calibrated YO L O3 and a mobilenet neural network model in a development board for identifying the canine type of a dog and/or the canine status of the dog in a video image;
and step 3: and driving a voice loudspeaker to give an alarm prompt according to the recognized dog type and/or the dog-restraining state of the dogs. When the dog types belong to large malignant dogs, the voice can be given to remind dog owners of paying attention to safety risks existing in the dogs, if the dogs do not tie the dog leashes, the voice is also used for giving an alarm prompt to remind the dog owners of giving the alarm prompt, and the dog owners are reminded of tying the dog leashes when walking the dogs;
and 4, step 4: the development board sends the image data and the recognized state data to the cloud data platform, and monitoring personnel access the cloud data platform through the client to obtain the monitoring data. When identifying a dog leash which is not strong, a large dog and the like or is not bound, a voice alarm reminding signal is sent out, video data corresponding to the identified dog leash which is not bound, position data corresponding to a node, time data and the like are sent to a cloud data platform, and then monitoring personnel can conveniently acquire related data to monitor, alarm and manage.
The neural network model which is preset in the development board and calibrated by training adopts two convolutional neural networks, and the neural network model is trained through a preset data set and then used for identifying and outputting the state of the toe dog through the neural network model. The training mode comprises the following steps:
1. acquiring pictures containing dogs and tethers through on-site acquisition, network collection and other ways to perform data preprocessing and labeling work, and constructing training set sample image data;
2. training target detection network structures such as a convolutional neural network YO L OV3 and the like by adopting a training set to obtain a fasciculate recognition model;
3. and inputting the pictures to be identified into the dog binding identification model, acquiring the specific positions of the dogs and the dog ropes in each picture to be identified, counting and summing the number of the dogs and the ropes in all the pictures to be identified, and acquiring a dog position matrix M (M × 4) and a rope position matrix N (N × 4) for storing the positions.
4. We can use a matrix Y of m x n to represent the IOU values to facilitate subsequent operations, where m is the number of dogs and n is the number of tethers, and loop each tether groupMi(i＝1...m)And the IOU values of N all canine N-tuples, and storing the calculated IOU values in Y: (m×n)In the matrix, judging the position j of all columns with the value of 0 in the Y matrix, and returningNj(x1，y1，x2，y2)Can be based onThe returned coordinates depict the cord with an IOU of 0, i.e., the dog and tether in an unbundled state.
The application adopts hardware comprising paddlePi or up2 and other embedded development boards, a loudspeaker, a power amplifier and a camera, and identifies whether the pet dog ties the dog rope or not through an image identification technology, so as to alarm, monitor, remind and the like.
The general technical scheme can be divided into two parts, namely target detection and dog class classification, wherein the target detection part carries out model training based on the dog leasing rope and the picture data set of the dog to obtain a target identification model of the dog and the dog leasing rope, and further judges the constraint state of the dog on the basis. And a dog class classification part for training a discrimination model of a large dog and a malignant dog of a dog on the dog class data set, and further judging the class of the dog picture detected by the target.
About voice alarm equipment, can integrate and discern and warn in real time the restraint condition of dog at outdoor intelligent lawn audio amplifier, upload the video to servers such as cloud end data platform simultaneously through the network, will violate information such as the place, time, picture of dog restraint and upload the cloud platform through the network, the cloud platform is gathered and is visual to data. And the law enforcement manager controls the restriction condition of the dogs in the community through the cloud platform and dispatches patrol personnel. And (4) implementing dissuasion and punishment by patrol personnel according to the information and the historical records of the handheld equipment. The community staff and the property can adjust the broadcasting content of the system through the broadcasting system, and can also take certain propaganda or dissuasion measures through the information provided by the cloud platform.
In order to obtain required training data (image data of a dog), three data crawler, manual shooting and movie and television work extraction modes are adopted for acquiring data. In the process of collecting data, training pictures are obtained from the perspective of a 'dog' instead of a general cell monitoring overlook angle, and the reason for doing so is as follows:
1) the height of the lawn sound box is basically consistent with that of the dog. The training picture acquired from the dog visual angle is consistent with the installation visual angle of the camera of the future system, and the effectiveness of system identification is improved.
2) The future function expansion of the system is considered, the training picture is obtained from the view angle of the dog, the picture of the dog can be clear, and the large dog and the hard dog can be further identified.
For the target detection work, considering that the workload of image marking is very large, we first label 1000 pictures to generate Dog _ 1000. On the basis of continuous experiments, 2000 pictures are added, and a Dog _3000 data set is generated to verify the feasibility of the neural network algorithm. Wherein all pictures are normalized to 640 x 480 resolution. After the feasibility of the algorithm is preliminarily verified, according to the experimental results in the early stage, the toe dog data are further increased to 6800 by different means such as network, field shooting, movie and television work extraction and the like. The information of the data is shown in table 1.
TABLE 1 bundle Dog Dataset (Dog L eash Dataset) information
We manually labeled large and malignant dogs based on the species data set disclosed by Stanford dogs, Udacity _ dogs 2 to better support training of the species classification model. Considering that there are few pictures in some of the two data sets for the category of dog, we further use the crawler self-created Ahpu _ dogs data set as a supplement, and the three data sets are shown in table 2.
TABLE 2 Canine breed data set information
Based on the Dog _6800 data set, the experimental results of training the target detection model by using YO L OV3 (other target detection networks can be used as well) are shown in table 3, and it can be seen that the map value of the target identification of the toe Dog model can reach 0.83, and the performance is believed to be further improved with the increase of samples.
TABLE 3 target detection experiment results based on YO L OV3 model
The canine species are tried to be classified by a mobilenet classification network (other classification networks can also be used), and the classification network is trained through a data set and then embedded into a development board to be identified and classified in practice.
Considering the judgment requirement of the dog state of different dog bunches under the condition of multiple dogs. Based on dog and dog-restraining rope target identification, a dog restraining state discrimination algorithm based on a dog and dog restraining object image matrix is provided. The algorithm comprises the following steps
TABLE 5 dog state discrimination method
It is clear that the specific implementation of the invention is not restricted to the above-described embodiments, but that various insubstantial modifications of the inventive process concept and technical solutions are within the scope of protection of the invention.
Claims (10)
1. The utility model provides a dog state identification monitored control system which characterized in that: the monitoring alarm system comprises a plurality of monitoring alarm nodes, wherein each monitoring alarm node comprises a high-definition camera for shooting image data, a voice loudspeaker for sending an alarm voice signal and a control panel, the high-definition camera is connected with the control panel and used for sending the collected video image data to the control panel, and the control panel is used for identifying the state of a dog in an image; the control panel drives and controls the voice loudspeaker to send out an alarm signal according to the state data of the dog.
2. The dog state identification monitoring system of claim 1, wherein: the control panel is connected with the cloud data platform through a network and used for sending the identified dog state data and the position and time of the monitoring alarm node to the cloud data platform for monitoring, storing and forwarding.
3. The dog state identification monitoring system of claim 2, wherein: the cloud data platform is connected with a client of a monitoring manager and used for acquiring dog state information stored by the cloud data platform and corresponding position and time information.
4. The dog state identification monitoring system of claim 3, wherein: the client comprises a computer terminal and a mobile handheld terminal.
5. The dog state identification monitoring system according to any one of claims 1-4, wherein: the control panel comprises a control unit for data processing and voice loudspeaker drive control and a communication unit for network communication; the control unit is connected with the cloud data platform through the communication unit.
6. The dog state identification monitoring system according to any one of claims 1-4, wherein: and the control unit pre-calibrates and trains a neural network model for identifying the state of the dog to identify and judge the dog restraining state in the image.
7. The dog state identification monitoring system of claim 6, wherein: the neural network model is used for identifying the type of the dog and the dog restraining state.
8. The monitoring method of a dog state identification monitoring system as claimed in any one of claims 1-7, wherein:
step 1: acquiring video image data through a high-definition camera and sending the data to a development board;
step 2: the neural network model after training calibration is preset in the development board to identify the dog type and the dog restraining state of the dog in the video image;
and step 3: and driving a voice loudspeaker to give an alarm prompt according to the recognized dog type and/or the dog-restraining state of the dogs.
9. The monitoring method of the dog state identification monitoring system according to claim 8, wherein: the monitoring method further comprises the following step 4: the development board sends the image data and the recognized state data to the cloud data platform, and monitoring personnel access the cloud data platform through the client to obtain the monitoring data.
10. The monitoring method of the dog state identification monitoring system according to claim 8 or 9, wherein: in step 2, the neural network model which is preset in the development board and calibrated by training adopts a convolutional neural network, and the neural network model is trained through a preset data set and then used for identifying and outputting the dog state and the dog type through the neural network model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010214069.XA CN111447410A (en) | 2020-03-24 | 2020-03-24 | Dog state identification monitoring system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010214069.XA CN111447410A (en) | 2020-03-24 | 2020-03-24 | Dog state identification monitoring system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111447410A true CN111447410A (en) | 2020-07-24 |
Family
ID=71652410
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010214069.XA Pending CN111447410A (en) | 2020-03-24 | 2020-03-24 | Dog state identification monitoring system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111447410A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112163489A (en) * | 2020-09-21 | 2021-01-01 | 南京特殊教育师范学院 | Dangerous dog only identification early warning system based on deep learning technology |
| CN112784797A (en) * | 2021-01-29 | 2021-05-11 | 北京百度网讯科技有限公司 | Target image recognition method and device |
| CN112906678A (en) * | 2021-05-07 | 2021-06-04 | 南京甄视智能科技有限公司 | Illegal dog walking event detection method and device based on monitoring video |
| CN113011404A (en) * | 2021-05-25 | 2021-06-22 | 南京甄视智能科技有限公司 | Dog leash identification method and device based on time-space domain features |
| CN113498745A (en) * | 2021-07-16 | 2021-10-15 | 广州鼎飞航空科技有限公司 | Canine posture recognition method, device and equipment |
| DE102021005091A1 (en) | 2021-10-12 | 2022-01-13 | Daimler Ag | Method of interacting with an animal in a vehicle |
| CN113965722A (en) * | 2021-09-10 | 2022-01-21 | 浙江西谷数字技术股份有限公司 | Intelligent community security monitoring system based on Internet of things |
| CN115424211A (en) * | 2022-09-30 | 2022-12-02 | 星宠王国(北京)科技有限公司 | Civilized dog raising terminal operation method and device based on big data and terminal |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109582810A (en) * | 2018-12-03 | 2019-04-05 | 泰芯科技（杭州）有限公司 | A kind of dog dog management system and its implementation based on deep learning |
| US20190196698A1 (en) * | 2017-12-22 | 2019-06-27 | Adobe Inc. | Removing and Replacing Objects in Images According to a Directed User Conversation |
| CN110334593A (en) * | 2019-05-28 | 2019-10-15 | 浙江泽曦科技有限公司 | Pet recognition algorithms and system |
| CN110532888A (en) * | 2019-08-01 | 2019-12-03 | 悉地国际设计顾问（深圳）有限公司 | A kind of monitoring method, apparatus and system |
| CN110852198A (en) * | 2019-10-27 | 2020-02-28 | 恒大智慧科技有限公司 | Control method, equipment and storage medium for preventing pet dog attack in smart community |
-
2020
- 2020-03-24 CN CN202010214069.XA patent/CN111447410A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190196698A1 (en) * | 2017-12-22 | 2019-06-27 | Adobe Inc. | Removing and Replacing Objects in Images According to a Directed User Conversation |
| CN109582810A (en) * | 2018-12-03 | 2019-04-05 | 泰芯科技（杭州）有限公司 | A kind of dog dog management system and its implementation based on deep learning |
| CN110334593A (en) * | 2019-05-28 | 2019-10-15 | 浙江泽曦科技有限公司 | Pet recognition algorithms and system |
| CN110532888A (en) * | 2019-08-01 | 2019-12-03 | 悉地国际设计顾问（深圳）有限公司 | A kind of monitoring method, apparatus and system |
| CN110852198A (en) * | 2019-10-27 | 2020-02-28 | 恒大智慧科技有限公司 | Control method, equipment and storage medium for preventing pet dog attack in smart community |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112163489A (en) * | 2020-09-21 | 2021-01-01 | 南京特殊教育师范学院 | Dangerous dog only identification early warning system based on deep learning technology |
| CN112784797A (en) * | 2021-01-29 | 2021-05-11 | 北京百度网讯科技有限公司 | Target image recognition method and device |
| CN112906678A (en) * | 2021-05-07 | 2021-06-04 | 南京甄视智能科技有限公司 | Illegal dog walking event detection method and device based on monitoring video |
| CN113011404A (en) * | 2021-05-25 | 2021-06-22 | 南京甄视智能科技有限公司 | Dog leash identification method and device based on time-space domain features |
| CN113011404B (en) * | 2021-05-25 | 2021-08-24 | 南京甄视智能科技有限公司 | Dog leash identification method and device based on time-space domain features |
| CN113498745A (en) * | 2021-07-16 | 2021-10-15 | 广州鼎飞航空科技有限公司 | Canine posture recognition method, device and equipment |
| CN113498745B (en) * | 2021-07-16 | 2022-08-30 | 广州鼎飞航空科技有限公司 | Canine posture recognition method, device and equipment |
| CN113965722A (en) * | 2021-09-10 | 2022-01-21 | 浙江西谷数字技术股份有限公司 | Intelligent community security monitoring system based on Internet of things |
| DE102021005091A1 (en) | 2021-10-12 | 2022-01-13 | Daimler Ag | Method of interacting with an animal in a vehicle |
| CN115424211A (en) * | 2022-09-30 | 2022-12-02 | 星宠王国(北京)科技有限公司 | Civilized dog raising terminal operation method and device based on big data and terminal |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111447410A (en) | Dog state identification monitoring system and method | |
| CN207301750U (en) | Power grid construction intelligence control platform system | |
| US20090122144A1 (en) | Method for detecting events at a secured location | |
| CN204990695U (en) | Vehicle information collection system that breaks rules and regulations based on on -vehicle driving recording apparatus | |
| CN100527178C (en) | Automatic tracking display system for warn | |
| CN108205896A (en) | A kind of Campus transport safety intelligence monitoring and controlling manages system | |
| CN106254838A (en) | A kind of preventing damage to power transmission line caused by external force intelligent early-warning system | |
| CN107798781A (en) | A kind of shared bicycle management system and method based on community cloud monitoring system | |
| CN112542027A (en) | Construction site personnel safety warning system based on image recognition | |
| CN202584429U (en) | Traffic violation monitoring and processing system | |
| CN1665297A (en) | Electronic monitoring and management system for road traffic | |
| CN114358980A (en) | Intelligent community property management system and method based on Internet of things | |
| CN205211166U (en) | Vehicle information acquisition device that breaks rules and regulations based on on -vehicle driving recording apparatus | |
| CN112099109A (en) | Intelligent control system of meteorological observation station | |
| CN115661732A (en) | Garbage classification and supervision method based on AI intelligent recognition system | |
| CN201307339Y (en) | Public security bayonet system | |
| CN201886527U (en) | Electronic device for displaying school leaving of students and informing parents | |
| CN204375161U (en) | Parking offense automatic snapshot system | |
| CN107991953A (en) | One kind building cental system | |
| CN112001810A (en) | Intelligent forestry patrolling system and method based on machine vision | |
| CN210091231U (en) | Wisdom garden management system | |
| CN202352074U (en) | Comprehensive monitoring system with high-definition image recognition and video monitoring | |
| CN110260915A (en) | Environmental protection equipment on-line monitoring platform and monitoring method | |
| CN115314675A (en) | High-level monitoring management system and method for power area | |
| CN213338762U (en) | Edge calculation system based on animal identification |
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: 20200724 |
|
| RJ01 | Rejection of invention patent application after publication |