CN108829762A - The Small object recognition methods of view-based access control model and device - Google Patents
The Small object recognition methods of view-based access control model and device Download PDFInfo
- Publication number
- CN108829762A CN108829762A CN201810523284.0A CN201810523284A CN108829762A CN 108829762 A CN108829762 A CN 108829762A CN 201810523284 A CN201810523284 A CN 201810523284A CN 108829762 A CN108829762 A CN 108829762A
- Authority
- CN
- China
- Prior art keywords
- image
- database
- small object
- data
- stored
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
Abstract
Small object recognition methods and device this application discloses a kind of view-based access control model.This method includes:Acquire the image in some region;Specific Small object in image is identified;In the case where there is the Small object in the picture, store the image in first database;According to the acquisition time of image, data cleansing is carried out to the image being stored in first database and is stored in the second database;With image and the information relevant to the image shown in the first database and/or the second database.This method realizes identification, detection to mouse in specified region using vision technique, the data of acquisition are stored in first database, according to the acquisition time of described image, a part of picture is stored in the second database, to which user can quickly check the picture being stored in the second database, due to saving complete data in first database, user can also check the picture of other moment shootings.
Description
Technical field
This application involves image identifying and processing technical fields, more particularly to a kind of Small object identification side of view-based access control model
Method and device.
Background technique
The mode of tional identification mouse feelings is generally by traditional mode, such as pulvis method, Mousetrap capture, and mouse sticking plate, ocular estimate, this
The original mode of sample identifies mouse feelings, and there are many deficiencies, and pulvis method often spreads powder agent in specified region, according to biology above
By being left a trace later to determine whether there is biology to pass through, judge whether biology is mouse by footprint, but it is such
Identification method can only often identify whether have mouse to pass through, but cannot count specific time and the rule of mouse appearance, and this
The mode of sample is limited by factors such as environment again, such as:Weather, humidity, wind-force, if having other pets or biology broken
It is bad etc..Mousetrap capture and mouse sticking plate mode are only applicable to kill rats or kill mouse, but if the mousetrap or mouse sticking plate capture mouse after if
If not finding cleaning in time by owner, mouse is because of " stranded " and after death, but the germ with mouse is not because of mouse
Death and disappear, exactly because the corpse of mouse because do not cleared up and moldy metamorphism in time, germ can be with geometric progression
Rate increases and sprawling, kills mouse not only without eliminating germ, more germ is quickly spread instead, while the corpse of mouse may be used also
Secondary pollution can be understood to the biology of mouse food chain upper end, such as:Cat, snake, the animals such as dog.Obtained with ocular estimate mouse information without
Different from trusting to chance and strokes of luck, often even more lost more than gain with special messenger come " keeping an eye on " mouse.Although existing utilize image in the prior art
The method that identification technology identifies small-sized biological, but if can be generated a large amount of when being taken pictures incessantly using video camera
Image data, since data volume is big, leads to check slowing, user cannot be fast when by the storage of these data in the database
Speed obtains the detection picture about current time mouse.
Summary of the invention
Aiming to overcome that the above problem or at least being partially solved or extenuate for the application solves the above problems.
According to the one aspect of the application, a kind of Small object recognition methods of view-based access control model is provided, including:
Image acquisition step:Acquire the image in some region;
Small object identification step:Specific Small object in described image is identified;
Image storing step:There are in the case where the Small object in described image, described image is stored in first
In database;
Image cleaning step:According to the acquisition time of described image, to the described image being stored in first database into
Row data cleansing is simultaneously stored in the second database;With
Data displaying:Show the first database and/or image in the second database and related to the image
Information.
This method realizes identification, detection to mouse in specified region using vision technique, and the data of acquisition are stored in
In first database, according to the acquisition time of described image, a part of picture is stored in the second database, thus user's energy
It is enough quickly to check the picture being stored in the second database, due to saving complete data, user in first database
It can check the picture of other moment shootings.
Optionally, after the Small object identification step, this method further includes:
Quantity statistics step:There are in the case where the Small object in detecting described image, the Small object is counted
Quantitative value, temporarily store the quantitative value, to the subsequent image of the image carry out Small object identification, adjusted according to recognition result
The quantitative value, when cannot recognize that the Small object from subsequent image, using current number magnitude as Small object quantity
Statistical value increment.
Optionally, first database described in the first database includes one or more of following data library:Big number
According to platform, data warehouse, real-time data base, time series database, distributed relation database, distributed non-relational database
With the database stored based on distributed document;Second database is relevant database.
Optionally, the data displaying includes:
Acquired image is divided into different groups according to time tag, is shown and is corresponded to according to the time tag of user's selection
Image data list;
According to the querying condition of user, is inquired, will be looked into the first database and/or second database
It askes as the result is shown to the user.
Optionally, the quantity of the photographic device is two or more, and the different location in the same region is arranged in or sets
It sets in different zones.
Optionally, the Small object identification step includes:By artificial intelligence to the specific Small object in described image
It is identified.
According to further aspect of the application, a kind of Small object identification device of view-based access control model is additionally provided, including:
Image capture module is disposed for acquiring the image in some region;
Small object identification module is disposed for identifying the specific Small object in described image;
Image storage module is disposed in described image there are in the case where the Small object, by the figure
As being stored in first database;
Image cleaning module is disposed for the acquisition time according to described image, to being stored in first database
Described image data cleansing and be stored in the second database;With
Data disaply moudle is disposed for relevant to the described image information of display, so that described in user checks
Information and/or described image.
The device realizes identification, detection to mouse in specified region using vision technique, and the data of acquisition are stored in
In first database, according to the acquisition time of described image, a part of picture is stored in the second database, thus user's energy
It is enough quickly to check the picture being stored in the second database, due to saving complete data, user in first database
It can check the picture of other moment shootings.
Optionally, which further includes quantity statistics module, the Small object identification module also with the quantity statistics mould
Block connection, the quantity statistics module are disposed in detecting described image uniting there are in the case where the Small object
The quantitative value of the Small object is counted, the quantitative value is temporarily stored, Small object identification is carried out to the subsequent image of the image, according to
Recognition result adjusts the quantitative value and makees current number magnitude when cannot recognize that the Small object from subsequent image
For the increment of the statistical value of Small object quantity.
Optionally, the data disaply moudle is used for:Acquired image is divided into different groups, root according to time tag
Corresponding image data list is shown according to the time tag that user selects;According to the querying condition of user, in first data
It is inquired in library and/or second database, query result is shown to the user.
Optionally, described image acquisition module is connected with the Small object identification module by Internet of Things.
According to the accompanying drawings to the detailed description of the specific embodiment of the application, those skilled in the art will be more
Above-mentioned and other purposes, the advantages and features of the application are illustrated.
Detailed description of the invention
Some specific embodiments of the application are described in detail by way of example and not limitation with reference to the accompanying drawings hereinafter.
Identical appended drawing reference denotes same or similar part or part in attached drawing.It should be appreciated by those skilled in the art that these
What attached drawing was not necessarily drawn to scale.In attached drawing:
Fig. 1 is the schematic flow chart according to one embodiment of the Small object recognition methods of the view-based access control model of the application;
Fig. 2, Fig. 3 and Fig. 4 are the schematic diagrames of the recognition result obtained according to the present processes;
Fig. 5 is the schematic flow according to another embodiment of the Small object recognition methods of the view-based access control model of the application
Figure;
Fig. 6 is the schematic diagram of one embodiment of data display interface;
Fig. 7 is the schematic diagram of another embodiment of data display interface;
Fig. 8 is the schematic diagram of another embodiment of data display interface;
Fig. 9 is the schematic flow chart according to one embodiment of the Small object identification device of the view-based access control model of the application;
Figure 10 is the schematic flow according to another embodiment of the Small object identification device of the view-based access control model of the application
Figure;
Figure 11 is the block diagram of one embodiment of the calculating equipment of the application;
Figure 12 is the block diagram of one embodiment of the computer readable storage medium of the application.
Specific embodiment
According to the accompanying drawings to the detailed description of the specific embodiment of the application, those skilled in the art will be more
Above-mentioned and other purposes, the advantages and features of the application are illustrated.
Embodiments herein provides a kind of Small object recognition methods of view-based access control model, and Fig. 1 is the base according to the application
In the schematic flow chart of one embodiment of the Small object recognition methods of vision.This method includes:
S100 image acquisition step:Acquire the image in some region;
S200 Small object identification step:Specific Small object in described image is identified;
S400 image storing step:There are in the case where the Small object in described image, described image is stored in
In first database;
S500 image cleaning step:According to the acquisition time of described image, to the figure being stored in first database
As carrying out data cleansing and being stored in the second database;With
S600 data displaying:Show the first database and/or image in the second database and with the image
Relevant information.
This method realizes identification, detection to mouse in specified region using vision technique, and the data of acquisition are stored in
In first database, according to the acquisition time of described image, a part of picture is stored in the second database, thus user's energy
It is enough quickly to check the picture being stored in the second database, due to saving complete data, user in first database
It can check the picture of other moment shootings.
, can be by the image in web camera acquisition testing region, optionally in image acquisition step, which takes the photograph
Camera has night vision function, also to can be carried out shooting in the case where not having light source at night.The quantity of video camera can be one
It is a, be also possible to it is multiple, according to the range in region determine.The different location in the same region can be set in multiple video cameras,
It can be set in different regions, so as to carry out remote collection and processing simultaneously to multizone.The frequency for acquiring image can
To be set as needed, for example, it is primary to be set as acquisition per second.The picture of acquisition is uploaded in video data by Internet of Things
The heart is carried out the processing of image by video data center.For example, video data center is stored with Small object recognizer, it can be right
Small object in image is identified.Small object can be small-sized biological, for example, mouse etc..
In this step, Small object identification can be carried out by artificial intelligence etc..Artificial intelligence may include machine learning,
Deep learning and neural network.For example, can be carried out by particle swarm algorithm, genetic algorithm, greedy algorithm or ant group algorithm small
Target identification.
Optionally, in Small object identification step, the specific Small object in described image is carried out by artificial intelligence
Identification.For example, can be identified by trained deep learning model to the specific Small object in described image.It should
Step can be realized the automatic identification of single or more mouse.
In identification step, including Background Modeling step:Using the first image in image collection as background model,
Each pixel of the first image is established into original training set, the original training set includes closing on the pixel
The pixel value of pixel;Pixel classifying step:By the second sample set of the pixel in the second image of described image set
It is compared with the original training set of pixel corresponding in the first picture, the pixel of the second image is divided into prospect
Point or background dot, wherein the sample set of the pixel in second image includes the pixel closed on the pixel;Profile
Determine step:Determine profile that the pixel for being divided into prospect in second image is constituted to form foreground area figure
Piece;With target detection step:Classified using trained disaggregated model to the foreground area picture, so that it is determined that
The target of detection.Wherein, the second image can be the image acquired in image acquisition step.Using the first image as background mould
Type can greatly reduce a large amount of consumption for establishing background model to memory, accelerate arithmetic speed;Classified using artificial intelligence
The precision classified relative to traditional Feature Points Matching classification method is higher and more flexible.
Optionally, before the Background Modeling step, which further includes:Disaggregated model training step:
Disaggregated model is trained using background sample collection and target sample collection to be detected, is reached in the accuracy rate of the disaggregated model
In the case where first threshold, trained disaggregated model is obtained.The step can will acquire a certain number of to be detected small
Target sample collection and background negative sample collection establish high-precision classification device using artificial intelligence, instruct to the classified sample set of production
Practice and generate artificial intelligence model, target sample is effectively classified with negative sample.
Optionally, the pixel classifying step includes:By each of second sample set element and it is described just
It is poor that each of beginning sample set element is made, in the case where all differences are all larger than second threshold, by second sample
Collect corresponding pixel and be set as foreground point, otherwise set background dot for the pixel, according to foreground point and background dot to institute
The pixel for stating the second image carries out binary conversion treatment, and the pixel for being arranged to background dot is added in background model.
Optionally, the contours extract step includes:Discrete foreground point is eliminated by opening and closing operation, passes through integral operation
Foreground point is set by the pixel in region that foreground point surrounds.Using this method, continuous profile can be obtained, is eliminated
The interference of noise.
Optionally, after the target detection step, this method further includes;Disaggregated model updates step:Utilizing warp
Crossing the classification results that trained disaggregated model classifies to the foreground area picture is background, then utilizes the classification knot
Fruit reversely updates the background model.This method can be fed back and be updated to disaggregated model by result, so that classification mould
Type is more accurate, improves the speed and effect of succeeding target detection.
Fig. 2, Fig. 3 and Fig. 4 are the schematic diagrames of the recognition result obtained according to the present processes.Fig. 2 to Fig. 4 is will be at certain
The image shot when being tested in dining room, is identified by Small object, obtains recognition result, and described in the picture with contour line
The profile of Small object.As can be seen that the visual field, nearby testing result is a mouse and profile is than more complete in Fig. 2;Have in Fig. 3
Two mouse show the profile easily distinguished although also can detect that this two mouse in visual field distant place;There are three in Fig. 4
Mouse, the result of detection are also three.It is understood that this method can also detect more mouse simultaneously.It can
See, this method can accurately identify Small object.
Optionally, Fig. 5 is the signal according to another embodiment of the Small object recognition methods of the view-based access control model of the application
Property flow chart.After the S200 Small object identification step, this method further includes:
S300 quantity statistics step:There are in the case where the Small object in detecting described image, count described small
The quantitative value of target temporarily stores the quantitative value, Small object identification is carried out to the subsequent image of the image, according to recognition result
The quantitative value is adjusted, when cannot recognize that the Small object from subsequent image, using current number magnitude as Small object
The increment of the statistical value of quantity.
Picture comprising testing result is stored, sequentially in time, obtains historical data and latest data.History
Data can be one day, several days, a month even more prolonged data, and user is facilitated to extract the testing number of some period at any time
Amount.Latest data can be nearest several minutes, even more prolonged data of several hours.By these historical datas and most
New data, can verify the detection case of mouse, to calculate the omission factor and false detection rate of system.For example, certain mouse exists
Some day does not haunt within the scope of camera, then judge this day there are missing inspections.For example, certain mouse is just united
Statistical value is counted and has been included in, but this mouse disappeared from image later, the data of the mouse are for subsequent statistical data
For, belong to erroneous detection.
The application identifies image using artificial intelligence, is able to achieve jamproof function.Background is illuminated by the light, blocks
The interference of the factors such as object, surrounding enviroment can all impact testing result.By artificial intelligence, background variation can be excluded
And the interference that surrounding enviroment detect mouse, it is also avoided that a possibility that accidentally grabbing, guarantees the accuracy of mouse detection identification.
The time interval for shooting image can be with self-defining.According to specific environment, Image Acquisition interval is set, it is general every
Second generate three pictures, if current experiment place mouse haunts, frequency is higher, can will setting shooting image time between
Every being arranged smaller.
In S400 image storing step, by big data streaming processing technique, by the image real-time transmission of acquisition to
In one database, the first database is non-relational database.The first database may include in following data library
One or more:Big data platform, real-time data base, time series database, distributed relation database, is divided data warehouse
Cloth non-relational database and the database stored based on distributed document.The quantity of first database can be multiple.
According to the acquisition time of described image in S500 image cleaning step, described in being stored in first database
Image carries out data cleansing and is stored in the second database.It can be based on acquisition time, most freshly harvested data are stored in
In second database, second database is relevant database.If user is visited by website or application program (APP)
It asks historical data, then the inquiry of picture or information is carried out by the history report in access first database, if user thinks
It checks latest data, is then obtained into conventional relationship type database or cache database.It is understood that latest data
It can be obtained by relevant database.Due to there is pictures in the initial data for the magnanimity being stored in first database not
Completely, the case where storing inconsistent information, data exception, influences the inquiry and display of user at the terminal, so carrying out data
Cleaning is just particularly important, and data cleansing is mainly to pass through consistency check to carry out the data format of different first databases
It is unified, invalid value and missing values etc. are handled, to improve the quality of data.
Due to the limitation of relevant database storage capacity, there are storage bottleneck problems.It, will be newest using this kind of mode
Data are stored in relevant database, and by history data store in big data warehouse or distributed data base, it can either
Guarantee that the spilling of data or computer resource is not lost, prevented to all data, and the access speed of latest data can be provided.
Optionally, the S600 data displaying includes:Acquired image is divided into according to time tag different
Group shows corresponding image data list according to the time tag that user selects;
According to the querying condition of user, is inquired, will be looked into the first database and/or second database
It askes as the result is shown to the user.
Fig. 6 is the schematic diagram of one embodiment of data display interface.In order to make it easy to understand, being irised out in Fig. 6 with box
The list of each data, it is to be understood that in actual displayed, these boxes can be not present.
In the data display interface of Fig. 6, acquired image is divided into different groups, time tag according to time tag
Including:It is this day, this week, nearly January, nearly March, more, corresponding image data list is shown according to the time tag that user selects.
Fig. 7 is the schematic diagram of another embodiment of data display interface, and by image link, user can check corresponding moment acquisition
Picture.User input inquiry condition, querying condition can be the interval value of capture quantity, be looked into according to this in input frame
Inquiry condition is inquired in the first database and/or second database, and query result is shown to the user.
For example, user is indicating that capture quantity is to input 50 to 100 in input frame, the time tag that user selects is " nearly March ", second
The data that this week is stored in database, store all data in first database, then in the first database and described the
The data that two data base querying statistical values are 50 to 100, are shown to user for data list.
Fig. 8 is the schematic diagram of another embodiment of data display interface.The schematic diagram provides to multiple photographic devices
The picture of shooting carries out the result of Small object statistics.What horizontal axis indicated is the number of photographic device, and what the longitudinal axis indicated is certain
The statistical value of the Small object of period.This method can per diem, by week, monthly, per year counted, each can also be directed to
Photographic device, whole photographic devices, the photographic device difference of some regions are for statistical analysis.
Optionally, after data displaying, this method further includes:Warning step:When statistical value is more than given threshold,
Generate hint instructions.
The present processes pass through the image in web camera acquisition testing region, can promptly and accurately count mouse feelings
Information has accuracy rate height, the strong feature of timeliness.The image information obtained in this way is passed data by network mode
It is defeated arrive data storage device, big data technology is employed herein, with ensure mass data storage and calculate during will not go out
Existing hardware bottleneck.Whether there is mouse information in automatic identification image by artificial intelligence analysis after into data warehouse, if
Do not ignore the data then, if having recognized mouse information, it will by can only program, the mouse information captured is complete
It is recorded in the data table related of data warehouse, and retains relevant image data, in order to later examination and data statistics.
The present processes are suitable for various monitoring places, can be realized uninterrupted monitoring in whole day 24 hours, have visual statistics
Analytic function, obtained data are true and reliable, and can summarize automatically data, automatic to carry out early warning prompting, save big
The manpower and material resources cost of amount.
Embodiments herein additionally provides a kind of Small object identification device of view-based access control model.Fig. 9 is according to the application
The schematic flow of one embodiment of the Small object identification device of view-based access control model.The device includes:
Image capture module 100 is disposed for acquiring the image in some region;
Small object identification module 200 is disposed for identifying the specific Small object in described image;
Image storage module 400 is disposed in described image there are in the case where the Small object, will be described
Image is stored in first database;
Image cleaning module 500 is disposed for the acquisition time according to described image, to being stored in first database
In described image data cleansing and be stored in the second database;With
Data disaply moudle 600 is disposed for display information relevant to described image, so that user checks institute
State information and/or described image.
The device realizes identification, detection to mouse in specified region using vision technique, and the data of acquisition are stored in
In first database, according to the acquisition time of described image, a part of picture is stored in the second database, thus user's energy
It is enough quickly to check the picture being stored in the second database, due to saving complete data, user in first database
It can check the picture of other moment shootings.
Figure 10 is the schematic flow according to another embodiment of the Small object identification device of the view-based access control model of the application
Figure.Optionally, which further includes quantity statistics module 300, the Small object identification module also with the quantity statistics module
Connection, the quantity statistics module are disposed in detecting described image counting there are in the case where the Small object
The quantitative value of the Small object temporarily stores the quantitative value, is carried out to the subsequent image of the image by the artificial intelligence
Small object identification adjusts the quantitative value according to recognition result, when cannot recognize that the Small object from subsequent image,
Using current number magnitude as the increment of the statistical value of Small object quantity.
Optionally, the data disaply moudle 600 is used for:Acquired image is divided into according to time tag different
Group shows corresponding image data list according to the time tag that user selects;According to the querying condition of user, described first
It is inquired in database and/or second database, query result is shown to the user.
Optionally, which further includes warning module, is connect with data disaply moudle, for being more than setting threshold in statistical value
When value, hint instructions are generated.
The embodiment of the present application also provides a kind of calculating equipment, referring to Fig.1 1, which includes memory 1120, place
It manages device 1110 and is stored in the computer program that can be run in the memory 1120 and by the processor 1110, the computer
Program is stored in the space 1130 for program code in memory 1120, which executes by processor 1110
Shi Shixian is for executing any one steps of a method in accordance with the invention 1131.
The embodiment of the present application also provides a kind of computer readable storage mediums.Referring to Fig.1 2, the computer-readable storage
Medium includes the storage unit for program code, which is provided with for executing steps of a method in accordance with the invention
Program 1131 ', the program are executed by processor.
The embodiment of the present application also provides a kind of computer program products comprising instruction.When the computer program product exists
When being run on computer, so that computer is executed according to the present processes step.
In the above-described embodiments, can come wholly or partly by software, hardware, firmware or any combination thereof real
It is existing.When implemented in software, it can entirely or partly realize in the form of a computer program product.The computer program
Product includes one or more computer instructions.When computer loads and executes the computer program instructions, whole or portion
Ground is divided to generate according to process or function described in the embodiment of the present application.The computer can be general purpose computer, dedicated computing
Machine, computer network obtain other programmable devices.The computer instruction can store in computer readable storage medium
In, or from a computer readable storage medium to the transmission of another computer readable storage medium, for example, the computer
Instruction can pass through wired (such as coaxial cable, optical fiber, number from a web-site, computer, server or data center
User's line (DSL)) or wireless (such as infrared, wireless, microwave etc.) mode to another web-site, computer, server or
Data center is transmitted.The computer readable storage medium can be any usable medium that computer can access or
It is comprising data storage devices such as one or more usable mediums integrated server, data centers.The usable medium can be with
It is magnetic medium, (for example, floppy disk, hard disk, tape), optical medium (for example, DVD) or semiconductor medium (such as solid state hard disk
Solid State Disk (SSD)) etc..
Professional should further appreciate that, described in conjunction with the examples disclosed in the embodiments of the present disclosure
Unit and algorithm steps, can be realized with electronic hardware, computer software, or a combination of the two, hard in order to clearly demonstrate
The interchangeability of part and software generally describes each exemplary composition and step according to function in the above description.
These functions are implemented in hardware or software actually, the specific application and design constraint depending on technical solution.
Professional technician can use different methods to achieve the described function each specific application, but this realization
It is not considered that exceeding scope of the present application.
Those of ordinary skill in the art will appreciate that implement the method for the above embodiments be can be with
By program come instruction processing unit completion, the program be can store in computer readable storage medium, and the storage is situated between
Matter is non-transitory (English:Non-transitory) medium, such as random access memory, read-only memory, flash
Device, hard disk, solid state hard disk, tape (English:Magnetic tape), floppy disk (English:Floppy disk), CD (English:
Optical disc) and any combination thereof.
The preferable specific embodiment of the above, only the application, but the protection scope of the application is not limited thereto,
Within the technical scope of the present application, any changes or substitutions that can be easily thought of by anyone skilled in the art,
Should all it cover within the scope of protection of this application.Therefore, the protection scope of the application should be with scope of protection of the claims
Subject to.
Claims (10)
1. a kind of Small object recognition methods of view-based access control model, including:
Image acquisition step:Acquire the image in some region;
Small object identification step:Specific Small object in described image is identified;
Image storing step:There are in the case where the Small object in described image, described image is stored in the first data
In library;
Image cleaning step:According to the acquisition time of described image, the image being stored in the first database is counted
According to cleaning and be stored in the second database;With
Data displaying:Show the first database and/or image and letter relevant to the image in the second database
Breath.
2. the method according to claim 1, wherein this method further includes after the Small object identification step:
Quantity statistics step:There are in the case where the Small object in detecting described image, the number of the Small object is counted
Magnitude temporarily stores the quantitative value, carries out Small object identification to the subsequent image of the image, adjusted according to recognition result described in
Quantitative value, when cannot recognize that the Small object from subsequent image, using current number magnitude as the system of Small object quantity
The increment of evaluation.
3. the method according to claim 1, wherein the first database includes one of following data library
Or it is several:Big data platform, data warehouse, real-time data base, time series database, distributed relation database, distribution are non-
Relevant database and the database stored based on distributed document;Second database is relevant database.
4. according to the method in any one of claims 1 to 3, which is characterized in that the data displaying includes:
Acquired image is divided into different groups according to time tag, corresponding figure is shown according to the time tag that user selects
As data list;
It according to the querying condition of user, is inquired in the first database and/or second database, inquiry is tied
Fruit is shown to the user.
5. according to the method described in claim 4, setting exists it is characterized in that, the quantity of the photographic device is two or more
The different location in the same region is arranged in different zones.
6. the method according to claim 1, wherein the Small object identification step includes:Pass through artificial intelligence
Specific Small object in described image is identified.
7. a kind of Small object identification device of view-based access control model, including:
Image capture module is disposed for acquiring the image in some region;
Small object identification module is disposed for identifying the specific Small object in described image;
Image storage module is disposed in described image depositing described image there are in the case where the Small object
Storage is in first database;
Image cleaning module is disposed for the acquisition time according to described image, to the institute being stored in first database
Image is stated to carry out data cleansing and be stored in the second database;With
Data disaply moudle is disposed for display information relevant to described image, so that user checks the information
And/or described image.
8. device according to claim 7, which is characterized in that the device further includes quantity statistics module, the Small object
Identification module is also connect with the quantity statistics module, and the quantity statistics module is disposed in detecting described image
There are in the case where the Small object, the quantitative value of the Small object is counted, temporarily stores the quantitative value, behind the image
Image carry out Small object identification, the quantitative value is adjusted according to recognition result, when cannot recognize that institute from subsequent image
When stating Small object, using current number magnitude as the increment of the statistical value of Small object quantity.
9. device according to claim 7 or 8, which is characterized in that the data disaply moudle is used for:By collected figure
As being divided into different groups according to time tag, corresponding image data list is shown according to the time tag that user selects;According to
The querying condition of user is inquired in the first database and/or second database, query result is shown to
The user.
10. device according to claim 7, which is characterized in that described image acquisition module and the Small object identify mould
Block is connected by Internet of Things.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810523284.0A CN108829762B (en) | 2018-05-28 | 2018-05-28 | Vision-based small target identification method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810523284.0A CN108829762B (en) | 2018-05-28 | 2018-05-28 | Vision-based small target identification method and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108829762A true CN108829762A (en) | 2018-11-16 |
CN108829762B CN108829762B (en) | 2020-09-04 |
Family
ID=64146224
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810523284.0A Active CN108829762B (en) | 2018-05-28 | 2018-05-28 | Vision-based small target identification method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108829762B (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109511642A (en) * | 2018-11-29 | 2019-03-26 | 陕西师范大学 | A kind of mouse killing in grasslands method and system based on image and clustering |
CN109886129A (en) * | 2019-01-24 | 2019-06-14 | 北京明略软件系统有限公司 | Prompt information generation method and device, storage medium and electronic device |
CN109934099A (en) * | 2019-01-24 | 2019-06-25 | 北京明略软件系统有限公司 | Reminding method and device, storage medium, the electronic device of placement location |
CN110472492A (en) * | 2019-07-05 | 2019-11-19 | 平安国际智慧城市科技股份有限公司 | Target organism detection method, device, computer equipment and storage medium |
CN110532893A (en) * | 2019-08-05 | 2019-12-03 | 西安电子科技大学 | Icon detection method in the competing small map image of electricity |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20040037450A1 (en) * | 2002-08-22 | 2004-02-26 | Bradski Gary R. | Method, apparatus and system for using computer vision to identify facial characteristics |
CN102317929A (en) * | 2009-02-18 | 2012-01-11 | A9.Com有限公司 | Method and system for image matching |
US20120308121A1 (en) * | 2011-06-03 | 2012-12-06 | International Business Machines Corporation | Image ranking based on attribute correlation |
CN104284162A (en) * | 2014-10-29 | 2015-01-14 | 广州中国科学院软件应用技术研究所 | Video retrieval method and system |
CN105373813A (en) * | 2015-12-24 | 2016-03-02 | 四川华雁信息产业股份有限公司 | Equipment state image monitoring method and device |
CN107358190A (en) * | 2017-07-07 | 2017-11-17 | 广东中星电子有限公司 | A kind of image key area management method and device |
-
2018
- 2018-05-28 CN CN201810523284.0A patent/CN108829762B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20040037450A1 (en) * | 2002-08-22 | 2004-02-26 | Bradski Gary R. | Method, apparatus and system for using computer vision to identify facial characteristics |
CN102317929A (en) * | 2009-02-18 | 2012-01-11 | A9.Com有限公司 | Method and system for image matching |
US20120308121A1 (en) * | 2011-06-03 | 2012-12-06 | International Business Machines Corporation | Image ranking based on attribute correlation |
CN104284162A (en) * | 2014-10-29 | 2015-01-14 | 广州中国科学院软件应用技术研究所 | Video retrieval method and system |
CN105373813A (en) * | 2015-12-24 | 2016-03-02 | 四川华雁信息产业股份有限公司 | Equipment state image monitoring method and device |
CN107358190A (en) * | 2017-07-07 | 2017-11-17 | 广东中星电子有限公司 | A kind of image key area management method and device |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109511642A (en) * | 2018-11-29 | 2019-03-26 | 陕西师范大学 | A kind of mouse killing in grasslands method and system based on image and clustering |
CN109511642B (en) * | 2018-11-29 | 2021-05-11 | 陕西师范大学 | Grassland deratization method and system based on image and cluster analysis |
CN109886129A (en) * | 2019-01-24 | 2019-06-14 | 北京明略软件系统有限公司 | Prompt information generation method and device, storage medium and electronic device |
CN109934099A (en) * | 2019-01-24 | 2019-06-25 | 北京明略软件系统有限公司 | Reminding method and device, storage medium, the electronic device of placement location |
CN110472492A (en) * | 2019-07-05 | 2019-11-19 | 平安国际智慧城市科技股份有限公司 | Target organism detection method, device, computer equipment and storage medium |
CN110532893A (en) * | 2019-08-05 | 2019-12-03 | 西安电子科技大学 | Icon detection method in the competing small map image of electricity |
Also Published As
Publication number | Publication date |
---|---|
CN108829762B (en) | 2020-09-04 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108829762A (en) | The Small object recognition methods of view-based access control model and device | |
Aquino et al. | Automated early yield prediction in vineyards from on-the-go image acquisition | |
CN109726700B (en) | Insect pest recognition early warning method and device based on multiple features | |
Wu et al. | Detection and counting of banana bunches by integrating deep learning and classic image-processing algorithms | |
CN110689519B (en) | Fog drop deposition image detection system and method based on yolo network | |
Boogaard et al. | Robust node detection and tracking in fruit-vegetable crops using deep learning and multi-view imaging | |
US10729117B2 (en) | Pest monitoring method based on machine vision | |
CN108874910B (en) | Vision-based small target recognition system | |
Whalley et al. | Applications of image processing in viticulture: A review | |
CN115482465A (en) | Crop disease and insect pest prediction method and system based on machine vision and storage medium | |
CN108460370B (en) | Fixed poultry life information alarm device | |
CN114818909A (en) | Weed detection method and device based on crop growth characteristics | |
CN116543347A (en) | Intelligent insect condition on-line monitoring system, method, device and medium | |
CN112580671A (en) | Automatic detection method and system for multiple development stages of rice ears based on deep learning | |
CN115661650A (en) | Farm management system based on data monitoring of Internet of things | |
CN110728810A (en) | Distributed target monitoring system and method | |
CN114723667A (en) | Agricultural fine planting and disaster prevention control system | |
CN115601585A (en) | Agricultural pest and disease diagnosis method and device based on picture analysis | |
CN111797831A (en) | BIM and artificial intelligence based parallel abnormality detection method for poultry feeding | |
CN112116206A (en) | Intelligent agricultural system based on big data | |
CN114898405A (en) | Portable broiler chicken abnormity monitoring system based on edge calculation | |
CN116300608A (en) | Intelligent agriculture remote monitoring system based on big data | |
CN110781865A (en) | Crop growth control system | |
WO2022137242A1 (en) | Deep learning system for estimating object volume, weight and crop yields based on remote imaging | |
CN113989538A (en) | Depth image-based chicken flock uniformity estimation method, device, system and 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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |