CN111343426B - Control method of anti-theft monitoring equipment for basement - Google Patents
Control method of anti-theft monitoring equipment for basement Download PDFInfo
- Publication number
- CN111343426B CN111343426B CN202010131336.7A CN202010131336A CN111343426B CN 111343426 B CN111343426 B CN 111343426B CN 202010131336 A CN202010131336 A CN 202010131336A CN 111343426 B CN111343426 B CN 111343426B
- Authority
- CN
- China
- Prior art keywords
- theft
- image
- monitoring
- bat
- support vector
- 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.)
- Active
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 33
- 238000000034 method Methods 0.000 title claims abstract description 17
- 238000012806 monitoring device Methods 0.000 claims abstract description 36
- 238000012706 support-vector machine Methods 0.000 claims abstract description 35
- 238000012549 training Methods 0.000 claims abstract description 8
- 229920000742 Cotton Polymers 0.000 claims description 19
- 238000004140 cleaning Methods 0.000 claims description 18
- 230000001965 increasing effect Effects 0.000 claims description 8
- 238000012360 testing method Methods 0.000 claims description 8
- 238000011156 evaluation Methods 0.000 claims description 6
- 238000006243 chemical reaction Methods 0.000 claims description 5
- 241000288673 Chiroptera Species 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 230000009467 reduction Effects 0.000 claims description 3
- 230000006399 behavior Effects 0.000 abstract description 18
- 238000005259 measurement Methods 0.000 abstract 1
- 230000007306 turnover Effects 0.000 abstract 1
- 230000006870 function Effects 0.000 description 10
- 230000008901 benefit Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
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
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- 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
-
- 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
- G08B13/19613—Recognition of a predetermined image pattern or behaviour pattern indicating theft or intrusion
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B13/00—Burglar, theft or intruder alarms
- G08B13/22—Electrical actuation
- G08B13/24—Electrical actuation by interference with electromagnetic field distribution
- G08B13/2491—Intrusion detection systems, i.e. where the body of an intruder causes the interference with the electromagnetic field
Abstract
The invention relates to the technical field of measurement turnover cabinet identification, and particularly discloses a control method of anti-theft monitoring equipment for a basement, which comprises the following steps: the monitoring equipment acquires the theft images of the electric vehicles, marks the theft characteristics of the theft images to form a theft image sample library, and establishes a support vector machine and optimizes the support vector machine. And training the support vector machine by using the stolen image sample. The method comprises the steps that a plurality of monitoring devices are arranged in a basement, continuous frame monitoring images of the basement are obtained in real time, a support vector machine capable of predicting theft is adopted to predict theft of the continuous frame monitoring images, if one or more monitoring devices obtain the theft images within a continuous period of time, the monitoring devices are controlled to send the theft images to a monitoring center, and meanwhile an alarm unit of the monitoring center is controlled to give an alarm. And identifying the theft behavior in real time by using a robot learning algorithm and a monitoring device, and rapidly stopping the theft behavior according to the theft behavior of the found theft image.
Description
Technical Field
The invention belongs to the technical field of monitoring equipment control, and particularly relates to a control method for basement anti-theft monitoring equipment.
Background
The existing basement is generally a place for placing vehicles, and comprises a trolley or an electric car, the basement is placed in a house where the basement does not occupy a house, and the basement is placed and obtained very conveniently, however, basement personnel are generally few, and no supervision personnel exist, so that theft is easy to occur, the electric car is small in size and has no anti-theft measures, and therefore, the probability that the electric car placed in the basement is stolen is very high.
At present, the basement is basically provided with anti-theft measures, namely, a monitoring camera is arranged on the basement, the camera collects videos at fixed positions in real time, meanwhile, the videos are stored in a local image, when a theft action occurs, the stored local videos are replayed and inquired to inquire the stolen videos, and therefore the stolen electric car is searched according to the stolen videos.
Disclosure of Invention
The invention aims to provide a control method of anti-theft monitoring equipment for a basement, so as to overcome the defect that the efficiency of the conventional video-recording rechecking stolen electric vehicle is low.
In order to achieve the above object, the present invention provides a control method for a basement anti-theft monitoring device, comprising:
and S1, the monitoring equipment acquires the theft images of the electric vehicles, marks the theft characteristics of the theft images and forms a theft image sample library.
S2, establishing a support vector machine, and optimizing the network parameters of the support vector machine by using a bat algorithm to form the support vector machine with the optimal network parameters.
S3, dividing the marked theft image sample into a training set and a testing set.
S4, arranging a plurality of monitoring devices in the basement, acquiring continuous frame monitoring images of the basement in real time, predicting the theft of the continuous frame monitoring images by using a support vector machine capable of predicting the theft, judging the car theft if one or more monitoring devices acquire the theft images (for example, images for opening a cushion of the electric car are continuously acquired in 15) within a continuous period of time, namely controlling the monitoring devices to send the theft images to a monitoring center, and simultaneously controlling an alarm unit of the monitoring center to give an alarm.
Preferably, in the above technical solution, step S2 specifically includes:
s201, setting parameters of a support vector machine: the ranges of the parameters of the penalty parameter C, the RBF nuclear parameter delta and the loss function epsilon are as follows; initializing bat group related parameters: setting initial population number n and pulse loudness A0Pulse emissivity r0A bat pulse emission rate increasing coefficient gamma, a pulse loudness attenuation coefficient alpha, and bat search pulse frequency upper and lower limits fmin,fmaxMaximum number of iterations tmaxAnd the search precision;
s202, initializing the bat position xiAnd velocity vi；
S203, determining a fitness evaluation function f (x), x ═ x1,…xd)TEvaluating the fitness value of each bat according to the fitness evaluation function to find the current optimal solution x*；
S204, adjusting the bat search pulse frequency, and updating the speed and the position of the bat according to the formulas (1), (2) and (3):
fi＝fmin+(fmax-fmin)β (1)
in the formula: beta is [0,1 ]]A randomly generated uniform random number; f. ofiRepresents a frequency of the acoustic wave; x is the number of*Representing a current global optimal solution;
indicating the position of the ith bat at time t,
representing the speed at that moment;
s205, generating uniformly distributed random number rand, if rand > riS206 is entered, otherwise S207 is entered, wherein riThe pulse emissivity of the ith bat;
s206, randomly disturbing the current optimal solution to generate a new solution, and carrying out border-crossing processing on the new solution, namely searching a local solution near the currently selected optimal solution and recording the current optimal solution;
s207, generating a new solution through random flight if rand is less than AiAnd f (x)i) F (x), then go to S208, otherwise go to S209, where aiThe pulse loudness of the ith bat;
s208, recording the new solution, and updating r by using the formulas (4) and (5)iAnd Ai；
In the formula, ri t+1Represents the pulse emissivity of the ith bat in the t +1 generation, ri 0Represents the maximum pulse emissivity of the ith bat, gamma is a pulse emissivity increasing coefficient, wherein gamma is more than 0,
respectively indicates the ith bat at t +1 and tThe loudness of the generation pulses, a ∈ [0,1 ]]Is the pulse loudness attenuation coefficient;
s209, sorting the fitness values of all bats in the bat group, and finding out the current optimal solution and the optimal fitness value;
s210, if the preset search precision is met or the maximum search times are reached, turning to the step S211, otherwise, returning to the step S204;
and S211, outputting the current global optimal solution, and based on the currently selected optimal support vector machine model and the parameters thereof.
Preferably, in the above technical solution, after the monitoring device acquires the image in step S1 and step S4, performing enhanced noise reduction on the image specifically includes:
s301, converting the image from the rgb channel to the hsv channel for contrast enhancement, and then converting the image back to the rgb channel;
s302, performing gray value conversion on the image;
and S303, denoising the image after the gray value conversion by using a Gaussian filter.
Preferably, in the above technical solution, each monitoring device is connected to an intelligent gateway, a plurality of intelligent gateways are respectively deployed as a block chain node to form a block chain network, and each block chain node is used for transmitting and storing a theft behavior image.
Preferably, in the above technical solution, the monitoring device includes: a shell, a camera, a reflector, a light source, a controller, a cleaning cotton brush, a first motor, a second motor, a timer and a cover body, the casing is in a hexahedral cylinder shape, each cylindrical surface of the hexahedral cylinder shaped casing is provided with a camera, the camera of the camera extends out of each cylindrical surface, a reflecting plate is arranged above each camera, a light source is arranged between the reflector and the camera, one end of the cleaning cotton brush is hinged on one side of the camera, one end of the cover body is hinged with one side of the cleaning cotton brush, the first motor is used for driving the cover body to cover the cleaning cotton brush, the second motor is used for driving the cleaning cotton brush to swing back and forth towards the surface of the camera, the camera, the light source, the first motor, the second motor and the timer are respectively connected with the controller, and the controller is connected with the monitoring center.
Preferably, in the above technical solution, the initial state of the light source of the monitoring device is not bright, when the monitoring device acquires a theft image, the light source is turned on, the monitoring device predicts an image acquired after the image is predicted again by using a support vector machine capable of predicting a theft, and determines that the theft is a theft if the image is predicted again, and determines that the theft is not a theft if the image is not predicted again.
Compared with the prior art, the invention has the following beneficial effects:
1. the invention utilizes the robot learning algorithm to combine with the monitoring equipment to identify the theft in real time, when the theft of the electric vehicle is identified, the theft image is stored and the alarm is given, thereby improving the monitoring efficiency, directly obtaining the theft image without reviewing the recording image, and further rapidly preventing the theft according to the found theft of the theft image.
2. Because the basement is darker, the invention introduces related equipment and monitoring method, namely when identifying the theft, the equipment and method can obtain the theft image with increased brightness again, thus improve the recognition rate of the theft.
Drawings
Fig. 1 is a flow chart of a control method of the anti-theft monitoring device for the basement.
Fig. 2 is a flow chart of the BA optimization support vector machine parameters of the present invention.
Fig. 3 is a top view of the monitoring device of the present invention.
Fig. 4 is a front view of the monitoring device of the present invention.
FIG. 5 is a structural diagram of the MerkleTree of the present invention.
The device comprises a shell 1, a camera 2, a light source 3, a reflector 4, a cover 5, a first motor 6, a cleaning cotton brush 7, a second motor 8 and a controller 9.
Detailed Description
The following detailed description of the present invention is provided in conjunction with the accompanying drawings, but it should be understood that the scope of the present invention is not limited to the specific embodiments.
As shown in fig. 1, the control method of the anti-theft monitoring device for the basement in this embodiment includes:
and S1, the monitoring equipment acquires the theft images of the electric vehicles, marks the theft characteristics of the theft images, namely, extracts the characteristics, and forms a theft image sample library. The theft image includes: opening images of various tram cushions, images of a car parking lock, images of a tram battery, images of a tram pushing and the like. And marking key features of various images.
S2, establishing a support vector machine, and optimizing the network parameters of the support vector machine by using a bat algorithm to form the support vector machine with the optimal network parameters; the input of the stolen image is x, the output is y, and y is f (x) which is a nonlinear relation, and an inverse function of y is f (x) can be obtained by using a BA-SVM support vector machine. In the embodiment, for simplifying the operation, the theft image is used as input by utilizing an SVM (support vector machine), and the marked theft behavior is used as output.
And S3, dividing the marked theft image sample into a training set and a testing set, randomly extracting the first 90% of the sample data as the training set and the last 10% as the testing set, inputting the training set into a support vector machine with the optimal network parameters to train the theft characteristic data, and testing the trained support vector machine by using the testing set to obtain the support vector machine capable of predicting the theft behavior.
S4, arranging a plurality of monitoring devices in the basement, acquiring continuous frame monitoring images of the basement in real time, predicting the theft of the continuous frame monitoring images by using a support vector machine capable of predicting the theft, and controlling the monitoring devices to send the theft images to a monitoring center and simultaneously control an alarm unit of the monitoring center to give an alarm if one or more monitoring devices acquire the theft images (such as images for opening a cushion of an electric car in 15) within a continuous period of time.
As shown in fig. 2, step S2 specifically includes:
s201, setting parameters of the support vector machine: punishmentThe parameter C has a parameter range of [1, 1000%]RBF nuclear parameter delta in the range of 0.1, 100]The parameter range of the loss function epsilon is [0.001, 1 ]](ii) a Initializing bat group related parameters: setting initial population number n and pulse loudness A0Pulse emissivity r0A bat pulse emission rate increasing coefficient gamma, a pulse loudness attenuation coefficient alpha, and bat search pulse frequency upper and lower limits fmin,fmaxMaximum number of iterations tmaxAnd the search precision;
s202, initializing the bat position xiAnd velocity vi；
S203, determining a fitness evaluation function f (x), x ═ x1,…xd)TEvaluating the fitness value of each bat according to the fitness evaluation function to find the current optimal solution x*；
S204, adjusting the bat search pulse frequency, and updating the speed and the position of the bat according to the formulas (1), (2) and (3):
fi＝fmin+(fmax-fmin)β (1)
in the formula: beta is [0,1 ]]A randomly generated uniform random number; f. ofiRepresents a frequency of the acoustic wave; x is the number of*Representing a current global optimal solution;
indicating the position of the ith bat at time t,
representing the speed at that moment;
s205, generating uniformly distributed random number rand, if rand > riS206 is entered, otherwise S207 is entered, wherein riThe pulse emissivity of the ith bat;
s206, randomly disturbing the current optimal solution to generate a new solution, and carrying out border-crossing processing on the new solution, namely searching a local solution near the currently selected optimal solution and recording the current optimal solution; s207, generating a new solution through random flight if rand is less than AiAnd f (x)i) F (x), then go to S208, otherwise go to S209, where aiThe pulse loudness of the ith bat;
s208, recording the new solution, and updating r by using the formulas (4) and (5)iAnd Ai；
In the formula, ri t+1Represents the pulse emissivity of the ith bat in the t +1 generation, ri 0Represents the maximum pulse emissivity of the ith bat, gamma is a pulse emissivity increasing coefficient, wherein gamma is more than 0,
respectively represents the pulse loudness of the ith bat in t +1 and t generations, and a is equal to 0,1]Is the pulse loudness attenuation coefficient;
s209, sorting the fitness values of all bats in the bat group, and finding out the current optimal solution and the optimal fitness value;
s210, if the preset search precision is met or the maximum search times are reached, turning to the step S211, otherwise, returning to the step S204;
s211, outputting the current global optimal solution, and based on the currently selected optimal support vector machine model and the parameters thereof, including: training parameters (including penalty factor C, radial basis kernel function parameters, etc.), type of model, kernel function type, loss function and its parameters.
Further, the enhancing and denoising of the image after the monitoring device acquires the image in step S1 and step S4 specifically includes:
and S11, positioning the theft image to be identified, and cutting and zooming the corresponding area.
S12, an image contrast enhancement algorithm is adopted to convert the image from the rgb channel to the hsv channel for contrast enhancement, and then the image is converted back to the rgb channel.
And S13, converting the gray value of the image. Specifically, the algorithms B3 and F3 extract rgb values, and calculate the grayscale values according to the formula r 0.299+ g 0.587+ B0.144 (r, g, B are red, green, and blue luminance values).
And S14, denoising the image after the gray value conversion by using a Gaussian filter. The filtering principle is to calculate the weighted average value of the values of the pixel points and the surrounding pixel points, see formula (6), and the method has the advantages of less variables and strong control capability on the smoothness of the noise-reduced image. Considering that the actual detection situation is complex, the size of the Gaussian filter can be reduced when the noise reduction algorithm is optimized in the actual part, and therefore the operation efficiency is improved.
Wherein i and j are matrix row and column numbers, k is any positive integer, and sigma is variable.
Further, as shown in fig. 3 to 4, the monitoring apparatus includes: the camera comprises a shell 1, a camera 2, a reflector 4, a light source 3, a controller 9, a cleaning cotton brush 7, a first motor 6, a timer, a second motor 8 and a cover body 5, wherein the shell 1 is in a hexahedral cylinder shape, each cylindrical surface of the hexahedral cylinder shaped shell 1 is provided with one camera 2, a camera of the camera 2 extends out of each cylindrical surface, the reflector 4 is arranged above each camera 2, the light source 3 is arranged between the reflector 4 and the camera 2, one end of the cleaning cotton brush 7 is hinged to one side of the camera 2, one end of the cover body 5 is hinged to one side of the cleaning cotton brush 7, the first motor 6 is used for driving the cover body 5 to cover the cleaning cotton brush 7, the second motor 8 is used for driving the cleaning cotton brush 7 to swing back and forth towards the surface of the camera 2, and the camera 2, the light source 3, the first motor 6, the second motor 8 and the timer are respectively connected with, the controller 9 is connected with the monitoring center.
When the monitoring equipment works, the camera acquires the video in real time, transmits the video to the controller, and transmits the video to the monitoring center after being processed by the controller. Further, in the normality, the cover body covers on cleaning the cotton brush, timing through the timer, start once every 5 hours if regularly, when scheduled time, then start first motor and rotate and drive the cover body and upwards turn up, then the second motor drives and cleans the cotton brush toward the camera surface swing back and forth of camera, thereby clean the camera, be located the height all the year round with solving the camera, can't be by the mesh of cleaning, after the swing back and forth is several times, the second motor drives and cleans the cotton brush and resets, first motor drives the cover body and resets, thereby accomplish the camera and clean work.
Further, the basement is in a dark state all the year round, in order to prevent misjudgment, the initial state of a light source of the monitoring equipment is not bright, when the monitoring equipment acquires a theft behavior image, the controller of the monitoring equipment controls the turning on of the light source, the monitoring equipment predicts the image collected after the image is predicted again by using the support vector machine capable of predicting the theft behavior, at the moment, if the image is predicted again as the theft behavior image, the theft behavior is judged, and if the image is not predicted, the theft behavior is judged. When the theft is identified, the device and the method can acquire the theft image with increased brightness again, thereby improving the identification rate of the theft.
In the embodiment, a block chain storage technology is introduced, a controller of each monitoring device is connected with an intelligent gateway, the intelligent gateways are connected with one another, the intelligent gateways are respectively deployed as a block chain node, one controller is deployed as a tentacle node of a block chain network to form the block chain network, and each block chain node is used for transmitting, storing and updating the acquired theft behavior image; the controller is deployed as a block chain child node, so that each theft behavior image and log data are redundantly stored in each controller in a block chain mode; when any one controller node is damaged, all data of the damaged controller can be recovered from the adjacent controllers.
The specific construction method comprises the following steps:
the intelligent gateway has the advantages that the network is complete, the boundary calculation and the dirty data screening are realized, the data collected by the controller are sent to the intelligent gateway through the MQTT internet of things protocol, and the intelligent gateway has certain calculation capacity. The intelligent gateway may be deployed as a peer node of a blockchain. And the theft behavior image acquired each time is written into the block chain as block data and informs the adjacent nodes and the monitoring center.
And deploying the block link point module in the intelligent gateway. The gateway is automatically registered to a monitoring center after being started, and the unique identifier Hash (the original ID of each component of the intelligent gateway, which is generated by the SHA256 algorithm after being summarized) of the hardware of the industrial intelligent gateway is used as the attribute of the intelligent gateway and submitted to the monitoring center together. After receiving the registration of the intelligent gateway, the monitoring center adopts a block chain block sending algorithm to carry out block sending, and the block structure of the block chain is as shown in table 1:
TABLE 1 Block chain Structure Table diagram
After the intelligent gateway receives the acquired data, the acquired data are stored by adopting a block chain technology:
and inserting the Data parameters into the Collections field to generate a new Data List Data set. And then calculating the Merkle Tree root node Hash of all data nodes in the whole Collection field for checking the correctness of the collected data.
The MerkleTree structure is shown in fig. 5, and it should be noted that since the MerkleTree must be an even number of nodes, when the number of datalists is odd, the last node uses its own Hash to generate the Hash of the previous layer by two calculations. And after the data are inserted and the MerkLeTreeRootHash is updated successfully, the data added in the nodes are announced to all the online block nodes. All nodes (intelligent gateway nodes/upper monitoring equipment nodes of the network) receiving the message use PBFT (physical Byzantine factory Tolerance Practical Byzantine Fault-tolerant algorithm) to verify the validity of the message, and then update the acquisition information of the corresponding nodes on the chain.
The benefit of using blockchains to store industrial data is apparent. Data collected by the monitoring equipment are written into the block chain to become an electronic evidence which cannot be tampered, the data are completely decoupled from hardware, and the collected data chain can be recovered through the online node regardless of the state of the hardware.
In conclusion, the invention utilizes the robot learning algorithm and the monitoring equipment to identify the theft behavior in real time, when the theft behavior of the electric vehicle is identified, the theft image is stored and the alarm is given, thereby improving the monitoring efficiency, directly obtaining the theft image without reviewing and recording the image, and further rapidly stopping the theft behavior according to the found theft behavior of the theft image.
The foregoing descriptions of specific exemplary embodiments of the present invention have been presented for purposes of illustration and description. It is not intended to limit the invention to the precise form disclosed, and obviously many modifications and variations are possible in light of the above teaching. The exemplary embodiments were chosen and described in order to explain certain principles of the invention and its practical application to enable one skilled in the art to make and use various exemplary embodiments of the invention and various alternatives and modifications as are suited to the particular use contemplated. It is intended that the scope of the invention be defined by the claims and their equivalents.
Claims (4)
1. A control method of anti-theft monitoring equipment for a basement is characterized by comprising the following steps:
s1, the monitoring equipment acquires a plurality of theft images of the electric vehicle, and marks the theft characteristics of the theft images to form a theft image sample library;
s2, establishing a support vector machine, and optimizing the network parameters of the support vector machine by using a bat algorithm to form the support vector machine with the optimal network parameters;
s3, dividing the marked theft image sample into a training set and a testing set, inputting the training set into a support vector machine with optimal network parameters to train theft characteristic data, and testing the trained support vector machine by using the testing set to obtain the support vector machine capable of predicting theft;
s4, arranging a plurality of monitoring devices in a basement, acquiring continuous frame monitoring images of the basement in real time, predicting the theft of the continuous frame monitoring images by adopting a support vector machine capable of predicting the theft, controlling the monitoring devices to send the theft images to a monitoring center if one or more monitoring devices acquire the theft images within a continuous period of time, and controlling an alarm unit of the monitoring center to give an alarm;
the monitoring device includes: a shell, a camera, a reflector, a light source, a controller, a cleaning cotton brush, a first motor, a second motor, a timer and a cover body, the casing is in a hexahedral cylinder shape, each cylindrical surface of the hexahedral cylinder shaped casing is provided with a camera, the camera of the camera extends out of each cylindrical surface, a reflecting plate is arranged above each camera, a light source is arranged between the reflector and the camera, one end of the cleaning cotton brush is hinged on one side of the camera, one end of the cover body is hinged with one side of the cleaning cotton brush, the first motor is used for driving the cover body to cover the cleaning cotton brush, the second motor is used for driving the cleaning cotton brush to swing back and forth towards the surface of the camera, the camera, the light source, the first motor, the second motor and the timer are respectively connected with the controller, and the controller is connected with the monitoring center; the monitoring device is characterized in that the initial state of a light source of the monitoring device is not bright, when the monitoring device acquires a theft behavior image, the light source is turned on, the monitoring device predicts the acquired image again by using a support vector machine capable of predicting the theft behavior, if the image is predicted as the theft behavior image again, the theft behavior is judged, and if the image is not predicted as the theft behavior image, the theft behavior is judged.
2. The control method of the anti-theft monitoring device for the basement according to claim 1, wherein the step S2 specifically includes:
s201, setting parameters of the support vector machine: penalty parameter C, RBF kernel parameter δ, parameter range of loss function ε; initializing bat group related parameters: setting initial population number n and pulse loudness A0Pulse emissivity r0A bat pulse emission rate increasing coefficient gamma, a pulse loudness attenuation coefficient alpha, and bat search pulse frequency upper and lower limits fmin,fmaxMaximum number of iterations tmaxAnd the search precision;
s202, initializing the bat position xiAnd velocity vi；
S203, determining a fitness evaluation function f (x), x ═ x1,…xd)TEvaluating the fitness value of each bat according to the fitness evaluation function to find the current optimal solution x*；
S204, adjusting the bat search pulse frequency, and updating the speed and the position of the bat according to the formulas (1), (2) and (3):
fi＝fmin+(fmax-fmin)β (1)
in the formula: beta is [0,1 ]]A randomly generated uniform random number; f. ofiRepresents a frequency of the acoustic wave; x is the number of*Representing a current global optimal solution;
indicating the position of the ith bat at time t,
representing the speed at that moment;
s205, generating uniformly distributed random number rand, if rand > riS206 is entered, otherwise S207 is entered, wherein riThe pulse emissivity of the ith bat;
s206, randomly disturbing the current optimal solution to generate a new solution, and carrying out border-crossing processing on the new solution, namely searching a local solution near the currently selected optimal solution and recording the current optimal solution;
s207, generating a new solution through random flight if rand is less than AiAnd f (x)i) F (x), then go to S208, otherwise go to S209, where aiThe pulse loudness of the ith bat;
s208, recording the new solution, and updating r by using the formulas (4) and (5)iAnd Ai；
ri t+1＝ri 0[1-exp(-γ*t)] (4)
In the formula, ri t+1Represents the pulse emissivity of the ith bat in the t +1 generation, ri 0Represents the maximum pulse emissivity of the ith bat, gamma is a pulse emissivity increasing coefficient, wherein gamma is more than 0,
respectively represents the pulse loudness of the ith bat in t +1 and t generations, and a is equal to 0,1]Is the pulse loudness attenuation coefficient;
s209, sorting the fitness values of all bats in the bat group, and finding out the current optimal solution and the optimal fitness value;
s210, if the preset search precision is met or the maximum search times are reached, turning to the step S211, otherwise, returning to the step S204;
and S211, outputting the current global optimal solution, and based on the currently selected optimal support vector machine model and the parameters thereof.
3. The method for controlling the anti-theft monitoring device for the basement according to claim 1, wherein the monitoring device performs enhanced noise reduction on the image after acquiring the image in the steps S1 and S4, and specifically comprises:
s301, converting the image from the rgb channel to the hsv channel for contrast enhancement, and then converting the image back to the rgb channel;
s302, performing gray value conversion on the image;
and S303, denoising the image after the gray value conversion by using a Gaussian filter.
4. The control method of the anti-theft monitoring device for the basement according to claim 1, wherein each monitoring device is connected with an intelligent gateway, a plurality of intelligent gateways are respectively deployed as a block chain node to form a block chain network, and each block chain node is used for transmitting and storing a theft behavior image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010131336.7A CN111343426B (en) | 2020-02-28 | 2020-02-28 | Control method of anti-theft monitoring equipment for basement |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010131336.7A CN111343426B (en) | 2020-02-28 | 2020-02-28 | Control method of anti-theft monitoring equipment for basement |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111343426A CN111343426A (en) | 2020-06-26 |
| CN111343426B true CN111343426B (en) | 2021-05-28 |
Family
ID=71185747
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010131336.7A Active CN111343426B (en) | 2020-02-28 | 2020-02-28 | Control method of anti-theft monitoring equipment for basement |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111343426B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101635835A (en) * | 2008-07-25 | 2010-01-27 | 深圳市信义科技有限公司 | Intelligent video monitoring method and system thereof |
| CN105391768A (en) * | 2015-10-15 | 2016-03-09 | 桂林电子科技大学 | Monitoring system for logistics vehicles and monitoring method thereof |
| CN205179248U (en) * | 2015-11-23 | 2016-04-20 | 中核（天津）机械有限公司 | Production line spare part leak protection dress device |
| US9402061B2 (en) * | 2013-05-08 | 2016-07-26 | Smart-I, S.R.L. | Smart optical sensor for adaptive, predictive, and on-demand control of public lighting |
| CN108121999A (en) * | 2017-12-10 | 2018-06-05 | 北京工业大学 | Support vector machines parameter selection method based on mixing bat algorithm |
| CN110110690A (en) * | 2019-05-16 | 2019-08-09 | 廊坊鑫良基科技有限公司 | A kind of target pedestrian method for tracing, device, equipment and storage medium |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR102433931B1 (en) * | 2015-07-30 | 2022-08-19 | 삼성전자주식회사 | Method and device for recognizing motion |
| CN109377616B (en) * | 2018-10-30 | 2021-02-12 | 南京邮电大学 | Access control system based on two-dimensional face recognition |
-
2020
- 2020-02-28 CN CN202010131336.7A patent/CN111343426B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101635835A (en) * | 2008-07-25 | 2010-01-27 | 深圳市信义科技有限公司 | Intelligent video monitoring method and system thereof |
| US9402061B2 (en) * | 2013-05-08 | 2016-07-26 | Smart-I, S.R.L. | Smart optical sensor for adaptive, predictive, and on-demand control of public lighting |
| CN105391768A (en) * | 2015-10-15 | 2016-03-09 | 桂林电子科技大学 | Monitoring system for logistics vehicles and monitoring method thereof |
| CN205179248U (en) * | 2015-11-23 | 2016-04-20 | 中核（天津）机械有限公司 | Production line spare part leak protection dress device |
| CN108121999A (en) * | 2017-12-10 | 2018-06-05 | 北京工业大学 | Support vector machines parameter selection method based on mixing bat algorithm |
| CN110110690A (en) * | 2019-05-16 | 2019-08-09 | 廊坊鑫良基科技有限公司 | A kind of target pedestrian method for tracing, device, equipment and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 基于多参量联合特征的水下小目标分类识别;赵国贵 等;《水下无人系统学报》;20191215(第6期);第644-650页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111343426A (en) | 2020-06-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR101607224B1 (en) | Dynamic object classification | |
| CN102891960B (en) | Method and camera for determining an image adjustment parameter | |
| CN102445640B (en) | GIS device intelligent recognition method based on vector machine and artificial fish swarm optimization | |
| CN111126152A (en) | Video-based multi-target pedestrian detection and tracking method | |
| CN111680535B (en) | Method and system for real-time prediction of one or more potential threats in video surveillance | |
| CN110084130B (en) | Face screening method, device, equipment and storage medium based on multi-target tracking | |
| CN108805008A (en) | A kind of community's vehicle security system based on deep learning | |
| WO2020174459A1 (en) | A distributed-acoustic-sensing (das) analysis system using a generative-adversarial-network (gan) | |
| CN111931713B (en) | Abnormal behavior detection method and device, electronic equipment and storage medium | |
| KR102391853B1 (en) | System and Method for Processing Image Informaion | |
| CN112883782B (en) | Method, device, equipment and storage medium for identifying putting behaviors | |
| CN110826577A (en) | High-voltage isolating switch state tracking identification method based on target tracking | |
| CN106846304A (en) | Electrical equipment detection method and device based on infrared detection | |
| CN113112521A (en) | Motion detection method based on pulse array | |
| CN114445461A (en) | Visible light infrared target tracking training method and device based on non-paired data | |
| CN111343426B (en) | Control method of anti-theft monitoring equipment for basement | |
| CN116386081A (en) | Pedestrian detection method and system based on multi-mode images | |
| CN109101884B (en) | Pulse array prediction method | |
| CN112487877A (en) | Monitoring method, system, device and medium for standard operation of kitchen waste | |
| CN116935108A (en) | Method, device, equipment and medium for monitoring abnormal garbage throwing behavior | |
| CN112085096A (en) | Method for detecting local abnormal heating of object based on transfer learning | |
| CN111222370A (en) | Case studying and judging method, system and device | |
| CN113469013A (en) | Motor fault prediction method and system based on transfer learning and time sequence | |
| CN111199306A (en) | Prediction device, prediction method, and storage medium | |
| CN116579337B (en) | False news detection method integrating evidence credibility |
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 | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20200626 Assignee: Guangxi Jiulong Electronic Technology Co.,Ltd. Assignor: GUILIN University OF ELECTRONIC TECHNOLOGY Contract record no.: X2023980045660 Denomination of invention: A control method for basement anti-theft monitoring equipment Granted publication date: 20210528 License type: Common License Record date: 20231105 |
|
| EE01 | Entry into force of recordation of patent licensing contract |