CN108124485A - For the alarm method of limbs conflict behavior, device, storage medium and server - Google Patents
For the alarm method of limbs conflict behavior, device, storage medium and server Download PDFInfo
- Publication number
- CN108124485A CN108124485A CN201780002312.8A CN201780002312A CN108124485A CN 108124485 A CN108124485 A CN 108124485A CN 201780002312 A CN201780002312 A CN 201780002312A CN 108124485 A CN108124485 A CN 108124485A
- Authority
- CN
- China
- Prior art keywords
- training
- video image
- convolutional neural
- conflict
- limbs
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/59—Context or environment of the image inside of a vehicle, e.g. relating to seat occupancy, driver state or inner lighting conditions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/20—Movements or behaviour, e.g. gesture recognition
- G06V40/28—Recognition of hand or arm movements, e.g. recognition of deaf sign language
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/02—Alarms for ensuring the safety of persons
Abstract
It is a kind of for the problem that the alarm method of limbs conflict behavior, for solving how to realize safely and effectively alarm in time in the case where limbs conflict occurs for driver.This includes for the alarm method of limbs conflict behavior：The video image of driver present position in real-time collection vehicle, the scene that the video image includes driver's upper part of the body are live；The video image is converted into specify data format；The convolutional neural networks that the video image after format transformation is completed as input input to pre-training obtain the output of the convolutional neural networks as a result, the output result is there are limbs conflict behavior or there is no limbs conflict behaviors；If the output result is there are limbs conflict behavior, warning information is sent.
Description
Technical field
Alarm method, dress the present invention relates to video information process technical field more particularly to for limbs conflict behavior
It puts, storage medium and server.
Background technology
With vehicle-mounted industry, towards in digitlization and the striding forward and drive a vehicle of intelligent direction, there are some similar to hitting department
The body conflict events such as machine, people thirst for can have it is a set of can pair with the relevant limbs conflict behavior of driver carry out real time monitoring and
The device of alarm so that society develops towards more harmonious safe direction, reduces worried driver's trip, property worry and life
It threatens.
There is no perfect administrative mechanism for the alert process of driver's limbs conflict on the market at present, typically to hitting department
The involved party of machine makes push and block or electric shock, this method easily to some because own bodies reason cannot bear push and block energetically and
The involved party of electric shock causes serious actual bodily harm or even has life threat so that this method is not suitable for the popularization of society
And application.
As it can be seen that how to realize that safely and effectively alarm becomes ability in time in the case where limbs conflict occurs for driver
The problem of field technique personnel's urgent need to resolve.
The content of the invention
An embodiment of the present invention provides a kind of for the alarm method of limbs conflict behavior, device, storage medium and service
Device can realize that the automatic identification of limbs conflict behavior judges and sends warning information, any action is actively made without driver,
Reduce worried driver's trip, property worry and life threat；Meanwhile without implementing transition behavior to involved party in order to avoid causing to hinder
Evil, suitable for the promotion and application of the whole society.
In a first aspect, a kind of alarm method for limbs conflict behavior is provided, including：
The video image of driver present position, the video image include the scene of driver's upper part of the body in real-time collection vehicle
It is live；
The video image is converted into specify data format；
The convolutional neural networks that the video image after format transformation is completed as input input to pre-training, obtain
The output of the convolutional neural networks is as a result, the output result is there are limbs conflict behavior or there is no limbs conflict rows
For；
If the output result is there are limbs conflict behavior, warning information is sent.
Optionally, by following steps, training obtains the convolutional neural networks in advance：
Training group sample is collected in advance, and the training group sample includes multiple first video images for training；
The corresponding standard recognition result of each first video image in the training group sample is marked in advance, and the standard is known
Other result includes the conflict numerical value of characterization limbs conflict behavior degree, wherein, when the numerical value that conflicts is minimum value, represent the mark
Quasi- recognition result is there is no limbs conflict behaviors；
First video image is converted into specify data format；
Using first video image after format transformation as input input to convolutional neural networks, the convolution is obtained
The training output result of neutral net；
The training is exported into result as target, the hidden layer parameter of the convolutional neural networks is adjusted, to minimize
Error between the training output result standard recognition result corresponding with the training group sample arrived；
If the error meets preset condition, it is determined that the convolutional neural networks training is completed.
Optionally, further include：
Test group sample is collected in advance, and the test group sample includes multiple second video images for test；
The corresponding standard recognition result of each second video image in the test group sample is marked in advance；
Before determining that the convolutional neural networks training is completed, the alarm method for limbs conflict behavior also wraps
It includes：
Second video image is converted into specify data format；
Using second video image after format transformation as input input to the convolutional neural networks, obtain described
The test output result of convolutional neural networks；
Calculate the test error between test output result standard recognition result corresponding with the test group sample；
If the test error is greater than or equal to default error threshold, it is determined that the convolutional neural networks have not been trained
Into next time beginning is trained；
If the test error is less than default error threshold, performs the definite convolutional neural networks and trained
Into the step of.
Optionally, the warning information that sends includes：
Whether the conflict numerical value for judging the output result is more than default conflict threshold；
If the conflict numerical value of the output result is more than default conflict threshold, alarm to police law execution department；
If the conflict numerical value of the output result gives a warning not less than default conflict threshold to the vehicle scene
Information.
Optionally, the warning information that sends includes：
Obtain the real-time positioning information of the vehicle；
Default warning message, the real-time positioning information and the video image that gathers in real time are sent to what is specified
Alert terminal.
Second aspect provides a kind of alarm device for limbs conflict behavior, including：
Video image acquisition module, the video image of driver present position in real-time collection vehicle, the video figure
As the scene for including driver's upper part of the body is live；
Format converting module, for being converted into the video image to specify data format；
Neural network module, for complete the video image after format transformation as input input to pre-training
Convolutional neural networks, obtain the output of the convolutional neural networks as a result, the output result be there are limbs conflict behavior or
Limbs conflict behavior is not present in person；
Alarm module if being there are limbs conflict behavior for the output result, sends warning information.
Optionally, the convolutional neural networks are by the way that with lower module, training obtains in advance：
Training sample collection module, for collecting training group sample in advance, the training group sample is included for training
Multiple first video images；
Training sample mark module, for marking the corresponding mark of each first video image in the training group sample in advance
Quasi- recognition result, the standard recognition result include the conflict numerical value of characterization limbs conflict behavior degree, wherein, when conflict numerical value
For minimum value when, represent the standard recognition result as there is no limbs conflict behaviors；
First format converting module, for being converted into first video image to specify data format；
Network training module, for using first video image after format transformation as input input to convolutional Neural
Network obtains the training output result of the convolutional neural networks；
Parameter adjustment module for the training to be exported result as target, adjusts the hidden of the convolutional neural networks
Layer parameter, the training obtained with minimum are exported between result standard recognition result corresponding with the training group sample
Error；
Module is completed in training, if meeting preset condition for the error, it is determined that the convolutional neural networks have been trained
Into.
Optionally, further include：
Test sample collection module, for collecting test group sample in advance, the test group sample is included for test
Multiple second video images；
Test sample mark module, for marking the corresponding mark of each second video image in the test group sample in advance
Quasi- recognition result；
Before the training completion module determines that the convolutional neural networks training is completed, also trigger with lower module：
Second format converting module, for being converted into second video image to specify data format；
Network test module, for using second video image after format transformation as input input to the convolution
Neutral net obtains the test output result of the convolutional neural networks；
Test error computing module is known for calculating test output result standard corresponding with the test group sample
Test error between other result；
Training module again, if being greater than or equal to default error threshold for the test error, it is determined that the volume
Product neutral net not complete by training, starts to train next time；
Determining module is completed in training, if being less than default error threshold for the test error, triggers the training
It completes module and determines that the convolutional neural networks training is completed.
The third aspect provides a kind of alarm server for limbs conflict behavior, including memory, processor and
The computer program that can be run in the memory and on the processor is stored in, the processor performs the computer
The step of above-mentioned alarm method for limbs conflict behavior is realized during program.
Fourth aspect, provides a kind of computer readable storage medium, and the computer-readable recording medium storage has meter
Calculation machine program, the computer program realize the step of the above-mentioned alarm method for limbs conflict behavior when being executed by processor
Suddenly.
As can be seen from the above technical solutions, the embodiment of the present invention has the following advantages：
In the embodiment of the present invention, first, the video image of driver present position in real-time collection vehicle, the video image
Scene including driver's upper part of the body is live；Then, the video image is converted into specifying data format；Then, lattice will be converted
The convolutional neural networks that the video image after formula is completed as input input to pre-training, obtain the convolutional neural networks
Output as a result, the output result is there are limbs conflict behavior or there is no limbs conflict behaviors；If the output knot
Fruit is there are limbs conflict behavior, then sends warning information.In embodiments of the present invention, when generation is for the limbs conflict of driver
When, by the video image of driver present position in collection vehicle, the scene fact of driver's upper part of the body is obtained, by these video figures
It is identified as putting into convolutional neural networks, realizes that the automatic identification of limbs conflict behavior judges and sends warning information,
Any action is actively made without driver, reduces worried driver's trip, property worry and life threat；Meanwhile without to behavior
People implements transition behavior in order to avoid damaging, suitable for the promotion and application of the whole society.
Description of the drawings
It to describe the technical solutions in the embodiments of the present invention more clearly, below will be to embodiment or description of the prior art
Needed in attached drawing be briefly described, it should be apparent that, the accompanying drawings in the following description be only the present invention some
Embodiment, for those of ordinary skill in the art, without having to pay creative labor, can also be according to these
Attached drawing obtains other attached drawings.
Fig. 1 is a kind of alarm method one embodiment flow chart for limbs conflict behavior in the embodiment of the present invention；
Fig. 2 is that a kind of alarm method for limbs conflict behavior is advance under an application scenarios in the embodiment of the present invention
The flow diagram of training convolutional neural networks；
Fig. 3 is that a kind of alarm method for limbs conflict behavior is tested under an application scenarios in the embodiment of the present invention
The flow diagram of convolutional neural networks；
Fig. 4 is that a kind of alarm method for limbs conflict behavior is marking the application of conflict numerical value in the embodiment of the present invention
The flow diagram to send a warning message under scene；
Fig. 5 is a kind of alarm device one embodiment structure chart for limbs conflict behavior in the embodiment of the present invention；
Fig. 6 is the schematic diagram for the alarm server for limbs conflict behavior that one embodiment of the invention provides.
Specific embodiment
An embodiment of the present invention provides a kind of for the alarm method of limbs conflict behavior, device, storage medium and service
Device, for solving the problems, such as how alarm in time is realized safely and effectively in the case where limbs conflict occurs for driver.
Goal of the invention, feature, advantage to enable the present invention is more apparent and understandable, below in conjunction with the present invention
Attached drawing in embodiment is clearly and completely described the technical solution in the embodiment of the present invention, it is clear that disclosed below
Embodiment be only part of the embodiment of the present invention, and not all embodiment.Based on the embodiments of the present invention, this field
All other embodiment that those of ordinary skill is obtained without making creative work, belongs to protection of the present invention
Scope.
Referring to Fig. 1, a kind of alarm method one embodiment for limbs conflict behavior includes in the embodiment of the present invention：
101st, in real-time collection vehicle driver present position video image, the video image includes driver's upper part of the body
Scene is live；
In the present embodiment, the front of driver present position or side surface direction camera can be installed on vehicle, takes the photograph
Video capture is carried out as head is directed at driver position, the Video stream information of the scene fact of driver's upper part of the body can be collected in real time,
Frame segmentation and sampling are carried out to these Video stream informations, you can obtain video image.
It, will be in the executive agent of transmission of video images to this programme so that hold after camera collects video image
Row main body can get the video image and audio-frequency information of these acquisitions in real time.
It should be noted that the executive agent of this programme can specifically be mounted in terminal, system or remote on vehicle
Journey server, for ease of description, below unified presentation be executive agent.
102nd, the video image is converted into specifying data format；
It is understood that, it is necessary to be these video image numbers before video image is put into convolutional neural networks
According to a selected suitable data entry mode, to improve the recognition efficiency of convolutional neural networks and effect.In general, instructing
Before practicing convolutional neural networks, the just specified data format of selected good video image, and with selected specified data lattice
Formula carrys out the data format of adjusting training sample, so as to be trained to convolutional neural networks.Therefore, trained convolution is being used
When neutral net is identified, similarly need video image to be identified being converted into this specified data form.
Specifically, above-mentioned specified data format can be specifically the data matrix of n*n, and the size of data matrix can root
It is suitably adjusted according to actual conditions needs, is not especially limited herein.Selectable, this specifies data format to can also be figure
Picture, characteristic pattern, vector etc..
103rd, the convolutional neural networks for completing the video image after format transformation to pre-training as input input,
Obtain the output result of the convolutional neural networks；
After by video image format transformation, using video image as the convolutional Neural net of input input to pre-training completion
Network obtains the output of the convolutional neural networks as a result, wherein, the output result is there are limbs conflict behavior or does not deposit
In limbs conflict behavior.
It is understood that the convolutional neural networks be it is pre- first pass through the training of substantial amounts of training sample and complete to obtain,
Classification and Identification being carried out to the behavior act in video image, made to whether there is limbs conflict behavior in current video image
Go out real-time judge, and export result.For example, the convolutional neural networks by constantly learning, can distinguish some typical passes
In the scene of driver's limbs conflict, for example, driver pulls with behavior human arm, involved party is using hand or hand-held object is beaten
Driver etc..
Wherein, the pre-training process of above-mentioned convolutional neural networks will be described in detail in the following.
If the 104, the output result is there are limbs conflict behavior, warning information is sent.
In the present embodiment, if the output result is there are limbs conflict behavior, it may be considered that driver institute in vehicle
Limbs conflict just occurs in position, warning information should be sent in time at this time.Specifically, the form for sending warning information can have
It is a variety of, for example, sending suggestion voice " please stopping your behavior immediately " to involved party；Alternatively, any carry is not made to involved party
Show, to avoid stimulating involved party that behavior is caused to upgrade, executive agent is quietly alarmed to nearest public security system, waits police
Come to handle；Etc..
Preferably, GPS positioning module can be installed, so as to which executive agent can get the reality of vehicle in real time on vehicle
When location information；Then, when needing to send warning information, by default warning message, the real-time positioning information and in real time
The video image of acquisition is sent to the alarm terminal specified.Alarm terminal said herein can be the alert service of public security bureau
Device.Wherein, the video image sent together with warning message may be used as proving the evidence of limbs conflict behavior, in order to enforce the law
Personnel punish involved party.
In the following, it will describe in detail to the pre-training process of above-mentioned convolutional neural networks.As shown in Fig. 2, the convolution
Neutral net can training obtains in advance by following steps：
201st, training group sample is collected in advance, and the training group sample includes multiple first video images for training；
202nd, the corresponding standard recognition result of each first video image, the mark in the training group sample are marked in advance
Quasi- recognition result includes the conflict numerical value of characterization limbs conflict behavior degree, wherein, when the numerical value that conflicts is minimum value, represent institute
Standard recognition result is stated as there is no limbs conflict behaviors；
203rd, first video image is converted into specifying data format；
204th, using first video image after format transformation as input input to convolutional neural networks, obtain described
The training output result of convolutional neural networks；
205th, the training is exported into result as target, the hidden layer parameter of the convolutional neural networks is adjusted, with minimum
Change the error between obtained training output result standard recognition result corresponding with the training group sample；
If the 206th, the error meets preset condition, it is determined that the convolutional neural networks training is completed.
For above-mentioned steps 201 and step 202, it is necessary to collect to train in advance before training convolutional neural networks
Multiple video images, i.e., the first above-mentioned video image.The data volume of these the first video images is bigger, to convolutional Neural net
The training effect of network is better.
After these training group samples are collected into, it is also necessary to mark each first video image in these training group samples
Corresponding standard recognition result, i.e. which video image are acquisitions from there are the scene of limbs conflict behavior, which video images
It is the scene that limbs conflict behavior is not present in acquisition certainly.
It should be noted that in the present embodiment, limbs conflict behavior is identified in order to further enhance convolutional neural networks
Accuracy, in marker samples, the limbs conflict behavior of each first video image is marked different conflict numerical value, this
A little conflict numerical value represent the degree of limbs conflict behavior in video image.For example, can set conflict numerical value section as [0,
10], wherein 0 represents no limbs conflict behavior, conflict numerical value is minimum；1-10 is represented there are limbs conflict behavior, and 1-10 divides
The different degrees of of limbs conflict behavior is not represented, and numerical value is bigger, then limbs conflict is more serious.For example, specifically " it can will brandish fist
First tap hits driver " behavior conflict numerical value labeled as 4, the conflict numerical value of the behavior that " will hold knife attack driver " is labeled as 9, general
The conflict numerical value of the behavior of " armed attack driver " is labeled as 10.
Above-mentioned steps 203 are similar to the content of above-mentioned steps 102, and principle is essentially identical, and details are not described herein again.
For above-mentioned steps 204, in this training, the first video image after format transformation is inputted to convolutional Neural
Network, due to the convolutional neural networks, not yet training is completed at this time, and output trains what is exported result and mark in advance
Standard recognition result can there are certain deviation, errors.
For above-mentioned steps 205, after training output result is obtained, can calculate the training output result with it is described
Error between the corresponding standard recognition result of training group sample, and according to the error transfer factor calculated the convolutional neural networks
Hidden layer parameter, as far as possible so that minimizing the error between the training output result of follow-up training output and standard recognition result.
Wherein, which can be equipped with multiple levels, including convolutional layer, pond layer, full articulamentum etc., the convolutional Neural net
Network can be specifically not construed as limiting herein determines according to actual conditions equipped with how many a levels.Assuming that the convolution god in the present embodiment
X level is equipped with through network, then when adjusting hidden layer parameter, the parameters in x levels can be adjusted.
On the calculating of error, illustrate and be：Assuming that the conflict numerical value difference of the standard recognition result of wherein 3 samples
For 0,4,9；And the conflict numerical value that this 3 samples input after the convolutional neural networks training output result exported is respectively 0,5,
9, comparison is understood, in this training of 3 samples, error 33.3%.
For above-mentioned steps 206, the hidden layer parameter of convolutional neural networks is being adjusted repeatedly, is being carried out after repeatedly training, it is right
Error between standard recognition result more corresponding with training group sample than each training output result, if the error meet it is pre-
If condition, such as error are less than 5%, then it can determine that the convolutional neural networks training is completed.Wherein, the preset condition can
To be determined in the specific convolutional neural networks of training, for example setting error is less than specific threshold, which can be one
A percentages, specific threshold is smaller, then the convolutional neural networks that last training completion obtains are more stable, and accuracy of identification is got over
It is high.
It, can be with standard in order to further verify the training performance level of the convolutional neural networks for above-mentioned steps 206
Standby a set of test group sample different from training group sample is tested the convolutional neural networks, is examined.Before test, may be used
To collect test group sample in advance, the test group sample includes multiple second video images for test；Then, mark in advance
Remember the corresponding standard recognition result of each second video image in the test group sample.As shown in figure 3, in the definite convolution
Before neural metwork training is completed, the alarm method for limbs conflict behavior can also include：
301st, second video image is converted into specifying data format；
302nd, using second video image after format transformation as input input to the convolutional neural networks, obtain
The test output result of the convolutional neural networks；
303rd, the test between test output result standard recognition result corresponding with the test group sample is calculated
Error；
304th, judge whether the test error is less than default error threshold, if it is not, step 305 is then performed, if so,
Perform step 206；
305th, determine that the convolutional neural networks not complete by training, starts to train next time.
Above-mentioned steps 301~302 are similar to the content of above-mentioned steps 203~204, and principle is essentially identical, no longer superfluous herein
It states.
For above-mentioned steps 303, after test output result is obtained, calculate its standard corresponding with test group sample and know
Test error between other result assesses the training performance level of the convolutional neural networks by test error.Due to test
Test group sample is different from training group sample, more strange for the convolutional neural networks, therefore the effect assessed
Fruit can also be better than the Evaluated effect of training stage.
For above-mentioned steps 304~305, if the test error of this test illustrates not less than default error threshold
The convolutional neural networks do not meet the needs of actual use yet, and training remains unfulfilled, and may thereby determine that the convolutional Neural net
Network not complete by training, starts to train next time；If conversely, the test error is less than default error threshold, illustrate the volume
Product neutral net has met the needs of actual use, and training is completed, and performs step 206 and determines that the convolutional neural networks have been trained
Into.
Further, under the application scenarios in the training convolutional neural networks using the sample that conflict numerical value is marked,
Correspondingly, the output result that above-mentioned steps 103 obtain also includes the conflict numerical value of characterization limbs conflict degree, so as to such as Fig. 4 institutes
Show, above-mentioned steps 104 can include：
Whether the conflict numerical value for the 401st, judging the output result is more than default conflict threshold, if so, performing step
402, if it is not, then performing step 403；
402nd, alarm to police law execution department；
403rd, sent a warning message to the vehicle scene.
It is understood that whether above-mentioned steps 401~403, be more than default by the conflict numerical value for judging to export result
Conflict threshold judge whether current driver in the video image gathered in real time is faced with the threat of serious limbs conflict, if
More than the conflict threshold, then illustrate that the limbs conflict behavior degree in video image is serious, it is necessary to alert process；If not less than this
Conflict threshold then illustrates the limbs conflict behavior degree in video image still within general range, without wasting police strength processing,
Driver individual is remained to be handled.As it can be seen that the severity of limbs conflict is distinguished by the judgement to conflict numerical value,
While protecting driver safety, the social police strength resource of saving has been taken into full account, the harmony for being conducive to society is stablized.
Preferably, in the present embodiment, display, display driving letter used for vehicles can also be equipped on vehicle
Breath carries out real-time display, and display content can specifically be set according to different vehicle, for example, when driver is by limbs conflict, it can
The video image of acquisition is shown bulletin over the display, start the strength of the masses on vehicle that driver is helped to rescue.In addition,
When sending warning information, it can broadcast alarm sound by the reminding module installed on vehicle and the modes such as picture flash are warned
Show that involved party stops conflict behavior.
In the present embodiment, first, the video image of driver present position in real-time collection vehicle, the video image includes
The scene of driver's upper part of the body is live；Then, the video image is converted into specifying data format；It then, will be after format transformation
The video image as input input to pre-training complete convolutional neural networks, obtain the defeated of the convolutional neural networks
Go out as a result, the output result is there are limbs conflict behavior or there is no limbs conflict behaviors；If the output result is
There are limbs conflict behaviors, then send warning information.In the present embodiment, when the limbs conflict for driver occurs, pass through
The video image of driver present position in collection vehicle obtains the scene fact of driver's upper part of the body, these video images is put into
It is identified into convolutional neural networks, realizes that the automatic identification of limbs conflict behavior judges and sends warning information, without department
Owner, which moves, makes any action, reduces worried driver's trip, property worry and life threat；Meanwhile without implementing to involved party
Transition behavior is in order to avoid damage, suitable for the promotion and application of the whole society.
As it can be seen that the alarm method provided in this embodiment for limbs conflict behavior can pair limbs punching relevant with driver
The device that prominent behavior is monitored in real time and alarmed, to find such event in time, by alarming, warning is reduced because hitting department
Personal injury and interruption of train operation that machine is brought are the inconvenience that other passengers bring, without carrying out involved party with dangerous object
Reason attack, and retain the evidences such as video, picture, make truth reduction to spot, form complete management closed loop.
It is to be understood that the size of the sequence number of each step is not meant to the priority of execution sequence, each process in above-described embodiment
Execution sequence should determine that the implementation process without tackling the embodiment of the present invention forms any limit with its function and internal logic
It is fixed.
A kind of alarm method for limbs conflict behavior is essentially described above, will be directed to limbs conflict to one kind below
The alarm device of behavior is described in detail.
Fig. 5 shows a kind of alarm device one embodiment structure for limbs conflict behavior in the embodiment of the present invention
Figure.
In the present embodiment, a kind of alarm device for limbs conflict behavior includes：
Video image acquisition module 501, the video image of driver present position in real-time collection vehicle, the video
The scene that image includes driver's upper part of the body is live；
Format converting module 502, for being converted into the video image to specify data format；
Neural network module 503, for the video image after format transformation is complete to pre-training as input input
Into convolutional neural networks, obtain the output of the convolutional neural networks as a result, the output result is have limbs conflict rows
For or there is no limbs conflict behavior；
Alarm module 504 if being there are limbs conflict behavior for the output result, sends warning information.
Further, the convolutional neural networks can be by the way that with lower module, training obtains in advance：
Training sample collection module, for collecting training group sample in advance, the training group sample is included for training
Multiple first video images；
Training sample mark module, for marking the corresponding mark of each first video image in the training group sample in advance
Quasi- recognition result, the standard recognition result include the conflict numerical value of characterization limbs conflict behavior degree, wherein, when conflict numerical value
For minimum value when, represent the standard recognition result as there is no limbs conflict behaviors；
First format converting module, for being converted into first video image to specify data format；
Network training module, for using first video image after format transformation as input input to convolutional Neural
Network obtains the training output result of the convolutional neural networks；
Parameter adjustment module for the training to be exported result as target, adjusts the hidden of the convolutional neural networks
Layer parameter, the training obtained with minimum are exported between result standard recognition result corresponding with the training group sample
Error；
Module is completed in training, if meeting preset condition for the error, it is determined that the convolutional neural networks have been trained
Into.
Further, the alarm device for limbs conflict behavior can also include：
Test sample collection module, for collecting test group sample in advance, the test group sample is included for test
Multiple second video images；
Test sample mark module, for marking the corresponding mark of each second video image in the test group sample in advance
Quasi- recognition result；
Before the training completion module determines that the convolutional neural networks training is completed, it can also trigger with lower die
Block：
Second format converting module, for being converted into second video image to specify data format；
Network test module, for using second video image after format transformation as input input to the convolution
Neutral net obtains the test output result of the convolutional neural networks；
Test error computing module is known for calculating test output result standard corresponding with the test group sample
Test error between other result；
Training module again, if being greater than or equal to default error threshold for the test error, it is determined that the volume
Product neutral net not complete by training, starts to train next time；
Determining module is completed in training, if being less than default error threshold for the test error, triggers the training
It completes module and determines that the convolutional neural networks training is completed.
Further, the alarm module can include：
Conflict judging unit, for judging whether the conflict numerical value of the output result is more than default conflict threshold；
Alarm unit if the judging result for the conflict judging unit is yes, is alarmed to police law execution department；
Live Alarm Unit if being no for the judging result of the conflict judging unit, is sent out to vehicle scene
Go out warning message.
Further, the alarm module can include：
Location information acquiring unit, for obtaining the real-time positioning information of the vehicle；
Information transmitting unit, for regarding described in gathering default warning message, the real-time positioning information and in real time
Frequency image is sent to the alarm terminal specified.
Fig. 6 is the schematic diagram for the alarm server for limbs conflict behavior that one embodiment of the invention provides.Such as Fig. 6 institutes
Show, the alarm server 6 for limbs conflict behavior of the embodiment includes：Processor 60, memory 61 and it is stored in institute
The computer program 62 that can be run in memory 61 and on the processor 60 is stated, such as is performed above-mentioned for limbs conflict row
For alarm method program.The processor 60 is realized above-mentioned each for limbs conflict when performing the computer program 62
Step in the alarm method embodiment of behavior, such as step 101 shown in FIG. 1 is to 104.Alternatively, the processor 60 performs
The function of each module/unit in above-mentioned each device embodiment, such as module 501 shown in Fig. 5 are realized during the computer program 62
To 504 function.
Illustratively, the computer program 62 can be divided into one or more module/units, it is one or
Multiple module/units are stored in the memory 61, and are performed by the processor 60, to complete the present invention.Described one
A or multiple module/units can be the series of computation machine program instruction section that can complete specific function, which is used for
Implementation procedure of the computer program 62 in the alarm server 6 for limbs conflict behavior is described.
The alarm server 6 for limbs conflict behavior can be that the calculating such as home server, cloud server are set
It is standby.The alarm server for limbs conflict behavior may include, but be not limited only to, processor 60, memory 61.This field
Technical staff is appreciated that examples of the Fig. 6 only for the alarm server 6 of limbs conflict behavior, does not form to being directed to
The restriction of the alarm server 6 of limbs conflict behavior can be included than illustrating more or fewer components or combining some portions
Part or different components, for example, the alarm server for limbs conflict behavior can also include input-output equipment,
Network access equipment, bus etc..
The processor 60 can be central processing unit (Central Processing Unit, CPU), can also be
Other general processors, digital signal processor (Digital Signal Processor, DSP), application-specific integrated circuit
(Application Specific Integrated Circuit, ASIC), ready-made programmable gate array (Field-
Programmable Gate Array, FPGA) either other programmable logic device, discrete gate or transistor logic,
Discrete hardware components etc..General processor can be microprocessor or the processor can also be any conventional processor
Deng.
The memory 61 can be the internal storage unit of the alarm server 6 for limbs conflict behavior, example
Such as it is directed to the hard disk or memory of the alarm server 6 of limbs conflict behavior.The memory 61 can also be described for limbs
Match somebody with somebody on the External memory equipment of the alarm server 6 of conflict behavior, such as the alarm server 6 for limbs conflict behavior
Standby plug-in type hard disk, intelligent memory card (Smart Media Card, SMC), secure digital (Secure Digital, SD)
Card, flash card (Flash Card) etc..Further, the memory 61 can also both include described for limbs conflict behavior
Alarm server 6 internal storage unit also include External memory equipment.The memory 61 is used to store the computer
Other programs and data needed for program and the alarm server for limbs conflict behavior.The memory 61 may be used also
For temporarily storing the data that has exported or will export.
It is apparent to those skilled in the art that for convenience and simplicity of description, the system of foregoing description,
The specific work process of device and unit may be referred to the corresponding process in preceding method embodiment, and details are not described herein.
In the above-described embodiments, all emphasize particularly on different fields to the description of each embodiment, be not described in detail or remember in some embodiment
The part of load may refer to the associated description of other embodiments.
Those of ordinary skill in the art may realize that each embodiments described with reference to the embodiments described herein
Module, unit and/or method and step can be realized with the combination of electronic hardware or computer software and electronic hardware.This
A little functions are performed actually with hardware or software mode, specific application and design constraint depending on technical solution.Specially
Industry technical staff can realize described function to each specific application using distinct methods, but this realization is not
It is considered as beyond the scope of this invention.
In several embodiments provided herein, it should be understood that disclosed system, apparatus and method can be with
It realizes by another way.For example, the apparatus embodiments described above are merely exemplary, for example, the unit
Division is only a kind of division of logic function, can there is other dividing mode, such as multiple units or component in actual implementation
It may be combined or can be integrated into another system or some features can be ignored or does not perform.It is another, it is shown or
The mutual coupling, direct-coupling or communication connection discussed can be the indirect coupling by some interfaces, device or unit
It closes or communicates to connect, can be electrical, machinery or other forms.
The unit illustrated as separating component may or may not be physically separate, be shown as unit
The component shown may or may not be physical location, you can be located at a place or can also be distributed to multiple
In network element.Some or all of unit therein can be selected to realize the mesh of this embodiment scheme according to the actual needs
's.
In addition, each functional unit in each embodiment of the present invention can be integrated in a processing unit, it can also
That unit is individually physically present, can also two or more units integrate in a unit.Above-mentioned integrated list
The form that hardware had both may be employed in member is realized, can also be realized in the form of SFU software functional unit.
If the integrated unit is realized in the form of SFU software functional unit and is independent production marketing or use
When, it can be stored in a computer read/write memory medium.Based on such understanding, the present invention realizes above-described embodiment side
All or part of flow in method can also instruct relevant hardware to complete, the computer by computer program
Program can be stored in a computer readable storage medium, and the computer program is when being executed by processor, it can be achieved that above-mentioned each
The step of a embodiment of the method.Wherein, the computer program includes computer program code, and the computer program code can
Think source code form, object identification code form, executable file or some intermediate forms etc..The computer-readable medium can be with
Including：Any entity of the computer program code or device, recording medium, USB flash disk, mobile hard disk, magnetic disc, light can be carried
Disk, computer storage, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random
Access Memory), electric carrier signal, telecommunication signal and software distribution medium etc..It should be noted that the computer
The content that readable medium includes can carry out appropriate increase and decrease according to legislation in jurisdiction and the requirement of patent practice, such as
In some jurisdictions, according to legislation and patent practice, computer-readable medium does not include electric carrier signal and telecommunication signal.
The above, the above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations；Although with reference to before
Embodiment is stated the present invention is described in detail, it will be understood by those of ordinary skill in the art that：It still can be to preceding
The technical solution recorded in each embodiment is stated to modify or carry out equivalent substitution to which part technical characteristic；And these
Modification is replaced, and the essence of appropriate technical solution is not made to depart from the spirit and scope of various embodiments of the present invention technical solution.
Claims (10)
1. a kind of alarm method for limbs conflict behavior, which is characterized in that including：
The video image of driver present position in real-time collection vehicle, the scene that the video image includes driver's upper part of the body are real
Condition；
The video image is converted into specify data format；
The convolutional neural networks that the video image after format transformation is completed as input input to pre-training obtain described
The output of convolutional neural networks is as a result, the output result is there are limbs conflict behavior or there is no limbs conflict behaviors；
If the output result is there are limbs conflict behavior, warning information is sent.
2. the alarm method according to claim 1 for limbs conflict behavior, which is characterized in that the convolutional Neural net
By following steps, training obtains network in advance：
Training group sample is collected in advance, and the training group sample includes multiple first video images for training；
The corresponding standard recognition result of each first video image in the training group sample, the standard identification knot are marked in advance
Fruit includes the conflict numerical value of characterization limbs conflict behavior degree, wherein, when the numerical value that conflicts is minimum value, represents the standard and know
Other result is there is no limbs conflict behaviors；
First video image is converted into specify data format；
Using first video image after format transformation as input input to convolutional neural networks, the convolutional Neural is obtained
The training output result of network；
The training is exported into result as target, the hidden layer parameter of the convolutional neural networks is adjusted, is obtained with minimum
The training exports the error between result standard recognition result corresponding with the training group sample；
If the error meets preset condition, it is determined that the convolutional neural networks training is completed.
3. the alarm method according to claim 2 for limbs conflict behavior, which is characterized in that further include：
Test group sample is collected in advance, and the test group sample includes multiple second video images for test；
The corresponding standard recognition result of each second video image in the test group sample is marked in advance；
Before determining that the convolutional neural networks training is completed, the alarm method for limbs conflict behavior further includes：
Second video image is converted into specify data format；
Using second video image after format transformation as input input to the convolutional neural networks, the convolution is obtained
The test output result of neutral net；
Calculate the test error between test output result standard recognition result corresponding with the test group sample；
If the test error is greater than or equal to default error threshold, it is determined that the convolutional neural networks not complete by training,
Start to train next time；
If the test error is less than default error threshold, perform what the definite convolutional neural networks training was completed
Step.
4. the alarm method for limbs conflict behavior according to Claims 2 or 3, which is characterized in that described to send announcement
Alert information includes：
Whether the conflict numerical value for judging the output result is more than default conflict threshold；
If the conflict numerical value of the output result is more than default conflict threshold, alarm to police law execution department；
If the conflict numerical value of the output result gives a warning letter not less than default conflict threshold to the vehicle scene
Breath.
5. the alarm method according to any one of claim 1 to 3 for limbs conflict behavior, which is characterized in that institute
It states and sends warning information and include：
Obtain the real-time positioning information of the vehicle；
Default warning message, the real-time positioning information and the video image that gathers in real time are sent to the alarm specified
Terminal.
6. a kind of alarm device for limbs conflict behavior, which is characterized in that including：
Video image acquisition module, the video image of driver present position in real-time collection vehicle, the video image bag
The scene for including driver's upper part of the body is live；
Format converting module, for being converted into the video image to specify data format；
Neural network module, for the convolution for completing the video image after format transformation to pre-training as input input
Neutral net obtains the output of the convolutional neural networks as a result, the output result is there are limbs conflict behavior or not
There are limbs conflict behaviors；
Alarm module if being there are limbs conflict behavior for the output result, sends warning information.
7. the alarm device according to claim 6 for limbs conflict behavior, which is characterized in that the convolutional Neural net
Network is by the way that with lower module, training obtains in advance：
Training sample collection module, for collecting training group sample in advance, the training group sample is included for the multiple of training
First video image；
Training sample mark module, for the corresponding standard of each first video image in the training group sample to be marked to know in advance
Not as a result, the standard recognition result includes the conflict numerical value of characterization limbs conflict behavior degree, wherein, when conflict numerical value is most
During small value, the standard recognition result is represented as there is no limbs conflict behaviors；
First format converting module, for being converted into first video image to specify data format；
Network training module, for using first video image after format transformation as input input to convolutional Neural net
Network obtains the training output result of the convolutional neural networks；
Parameter adjustment module, for the training to be exported result as target, the hidden layer for adjusting the convolutional neural networks is joined
Number, to minimize the mistake between obtained training output result standard recognition result corresponding with the training group sample
Difference；
Module is completed in training, if meeting preset condition for the error, it is determined that the convolutional neural networks training is completed.
8. the alarm device according to claim 7 for limbs conflict behavior, which is characterized in that further include：
Test sample collection module, for collecting test group sample in advance, the test group sample is included for the multiple of test
Second video image；
Test sample mark module, for the corresponding standard of each second video image in the test group sample to be marked to know in advance
Other result；
Before the training completion module determines that the convolutional neural networks training is completed, also trigger with lower module：
Second format converting module, for being converted into second video image to specify data format；
Network test module, for using second video image after format transformation as input input to the convolutional Neural
Network obtains the test output result of the convolutional neural networks；
Test error computing module, for calculating test output result standard identification knot corresponding with the test group sample
Test error between fruit；
Training module again, if being greater than or equal to default error threshold for the test error, it is determined that the convolution god
Through network, training is not completed, and starts to train next time；
Determining module is completed in training, if being less than default error threshold for the test error, is triggered the training and is completed
Module determines that the convolutional neural networks training is completed.
9. a kind of alarm server for limbs conflict behavior including memory, processor and is stored in the memory
In and the computer program that can run on the processor, which is characterized in that the processor performs the computer program
Shi Shixian is directed to the step of alarm method of limbs conflict behavior as any one of claim 1 to 5.
10. a kind of computer readable storage medium, the computer-readable recording medium storage has computer program, and feature exists
In being realized when the computer program is executed by processor as any one of claim 1 to 5 for limbs conflict behavior
Alarm method the step of.
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| PCT/CN2017/119567 WO2019127271A1 (en) | 2017-12-28 | 2017-12-28 | Warning method, device, storage medium and server regarding physical conflict behavior |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108124485A true CN108124485A (en) | 2018-06-05 |
Family
ID=62234417
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201780002312.8A Pending CN108124485A (en) | 2017-12-28 | 2017-12-28 | For the alarm method of limbs conflict behavior, device, storage medium and server |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN108124485A (en) |
| WO (1) | WO2019127271A1 (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109003425A (en) * | 2018-08-10 | 2018-12-14 | 北京车和家信息技术有限公司 | A kind of method for early warning and relevant device |
| CN109685131A (en) * | 2018-12-20 | 2019-04-26 | 斑马网络技术有限公司 | Automobile vehicle device system exception recognition methods and device |
| CN109934119A (en) * | 2019-02-19 | 2019-06-25 | 平安科技（深圳）有限公司 | Adjust vehicle heading method, apparatus, computer equipment and storage medium |
| CN110135317A (en) * | 2019-05-08 | 2019-08-16 | 深圳达实智能股份有限公司 | Behavior monitoring and management system and method based on cooperated computing system |
| CN110276323A (en) * | 2019-06-26 | 2019-09-24 | 四川工商学院 | A kind of Activity recognition system and recognition methods held knife and kidnap driver |
| CN110288796A (en) * | 2019-06-21 | 2019-09-27 | 浙江大华技术股份有限公司 | Vehicle monitoring method and device, storage medium, electronic device |
| CN111008586A (en) * | 2019-11-29 | 2020-04-14 | 上海能塔智能科技有限公司 | Data processing method, device, equipment and storage medium for passenger car conflict detection |
| CN111428545A (en) * | 2019-01-10 | 2020-07-17 | 北京嘀嘀无限科技发展有限公司 | Behavior judgment method and device and electronic equipment |
| CN112165605A (en) * | 2020-09-14 | 2021-01-01 | 上海明略人工智能(集团)有限公司 | Behavior-limiting alarm method and device |
| CN113793441A (en) * | 2021-11-17 | 2021-12-14 | 兰和科技（深圳）有限公司 | Campus visitor safety analysis system based on thing networking |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112306220A (en) * | 2019-07-31 | 2021-02-02 | 北京字节跳动网络技术有限公司 | Control method and device based on limb identification, electronic equipment and storage medium |
| CN110458108B (en) * | 2019-08-13 | 2023-02-21 | 固安跃盛自动化设备有限公司 | Manual operation real-time monitoring method, system, terminal equipment and storage medium |
| CN111667175A (en) * | 2020-06-05 | 2020-09-15 | 易绍福 | Personal safety warning method, system, platform and storage medium |
| CN112836567A (en) * | 2020-12-08 | 2021-05-25 | 北京澎思科技有限公司 | Police force mobilization method and device, computer equipment and medium |
| CN112784706B (en) * | 2021-01-04 | 2023-11-14 | 中国石油天然气集团有限公司 | Oil testing test operation area safety control method based on image intelligent identification |
| CN113205008B (en) * | 2021-04-16 | 2023-11-17 | 深圳供电局有限公司 | Alarm control method for dynamic alarm window |
| CN113158937A (en) * | 2021-04-28 | 2021-07-23 | 合肥移瑞通信技术有限公司 | Sleep monitoring method, device, equipment and readable storage medium |
| CN113283978A (en) * | 2021-05-06 | 2021-08-20 | 北京思图场景数据科技服务有限公司 | Financial risk assessment method based on biological basis, behavior characteristics and business characteristics |
| CN113313908A (en) * | 2021-05-18 | 2021-08-27 | 许昌学院 | Safety monitoring system and method for network appointment |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090324010A1 (en) * | 2008-06-26 | 2009-12-31 | Billy Hou | Neural network-controlled automatic tracking and recognizing system and method |
| CN106373332A (en) * | 2016-09-30 | 2017-02-01 | 北京奇虎科技有限公司 | Vehicle-mounted intelligent alarm method and device |
| CN106407903A (en) * | 2016-08-31 | 2017-02-15 | 四川瞳知科技有限公司 | Multiple dimensioned convolution neural network-based real time human body abnormal behavior identification method |
| CN106571055A (en) * | 2016-10-28 | 2017-04-19 | 广州广电银通金融电子科技有限公司 | Internet-of-things-based intelligent vehicle management system and method |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101840632B (en) * | 2009-03-18 | 2013-05-29 | 深圳先进技术研究院 | Method and system for monitoring abnormal driving behavior in vehicle |
| CN104504400B (en) * | 2015-01-08 | 2018-07-17 | 中国科学院重庆绿色智能技术研究院 | A kind of driver's anomaly detection method based on online behavior modeling |
| CN106203626A (en) * | 2016-06-30 | 2016-12-07 | 北京奇虎科技有限公司 | Car steering behavioral value method and device, automobile |
-
2017
- 2017-12-28 WO PCT/CN2017/119567 patent/WO2019127271A1/en active Application Filing
- 2017-12-28 CN CN201780002312.8A patent/CN108124485A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090324010A1 (en) * | 2008-06-26 | 2009-12-31 | Billy Hou | Neural network-controlled automatic tracking and recognizing system and method |
| CN106407903A (en) * | 2016-08-31 | 2017-02-15 | 四川瞳知科技有限公司 | Multiple dimensioned convolution neural network-based real time human body abnormal behavior identification method |
| CN106373332A (en) * | 2016-09-30 | 2017-02-01 | 北京奇虎科技有限公司 | Vehicle-mounted intelligent alarm method and device |
| CN106571055A (en) * | 2016-10-28 | 2017-04-19 | 广州广电银通金融电子科技有限公司 | Internet-of-things-based intelligent vehicle management system and method |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109003425A (en) * | 2018-08-10 | 2018-12-14 | 北京车和家信息技术有限公司 | A kind of method for early warning and relevant device |
| CN109685131A (en) * | 2018-12-20 | 2019-04-26 | 斑马网络技术有限公司 | Automobile vehicle device system exception recognition methods and device |
| CN111428545A (en) * | 2019-01-10 | 2020-07-17 | 北京嘀嘀无限科技发展有限公司 | Behavior judgment method and device and electronic equipment |
| CN109934119A (en) * | 2019-02-19 | 2019-06-25 | 平安科技（深圳）有限公司 | Adjust vehicle heading method, apparatus, computer equipment and storage medium |
| CN109934119B (en) * | 2019-02-19 | 2023-10-31 | 平安科技（深圳）有限公司 | Method, device, computer equipment and storage medium for adjusting vehicle running direction |
| CN110135317A (en) * | 2019-05-08 | 2019-08-16 | 深圳达实智能股份有限公司 | Behavior monitoring and management system and method based on cooperated computing system |
| CN110288796A (en) * | 2019-06-21 | 2019-09-27 | 浙江大华技术股份有限公司 | Vehicle monitoring method and device, storage medium, electronic device |
| CN110276323A (en) * | 2019-06-26 | 2019-09-24 | 四川工商学院 | A kind of Activity recognition system and recognition methods held knife and kidnap driver |
| CN111008586A (en) * | 2019-11-29 | 2020-04-14 | 上海能塔智能科技有限公司 | Data processing method, device, equipment and storage medium for passenger car conflict detection |
| CN112165605A (en) * | 2020-09-14 | 2021-01-01 | 上海明略人工智能(集团)有限公司 | Behavior-limiting alarm method and device |
| CN113793441A (en) * | 2021-11-17 | 2021-12-14 | 兰和科技（深圳）有限公司 | Campus visitor safety analysis system based on thing networking |
| CN113793441B (en) * | 2021-11-17 | 2022-01-07 | 兰和科技（深圳）有限公司 | Campus visitor safety analysis system based on thing networking |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2019127271A1 (en) | 2019-07-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108124485A (en) | For the alarm method of limbs conflict behavior, device, storage medium and server | |
| CN104882001B (en) | Driving behavior monitoring method based on drive recorder, apparatus and system | |
| CN205091585U (en) | Intelligence sanitation car monitored control system based on wisdom sanitation cloud platform | |
| CN108351968B (en) | Alarming method, device, storage medium and server for criminal activities | |
| CN109219838B (en) | Traffic monitoring alarm method, device, terminal equipment and storage medium | |
| CN109657564A (en) | A kind of personnel detection method, device, storage medium and terminal device on duty | |
| CN109243181A (en) | Traffic accident method for early warning, device, terminal device and storage medium | |
| CN107945311A (en) | A kind of method for early warning of dangerous driving behavior, device, storage medium and server | |
| CN108647630A (en) | A kind of dangerous driving behavior measure of supervision and device based on video identification | |
| CN110046547A (en) | Report method, system, computer equipment and storage medium violating the regulations | |
| CN105469035A (en) | Driver's bad driving behavior detection system based on binocular video analysis | |
| CN106101628A (en) | The method of a kind of car accident process and terminal | |
| CN107220633A (en) | A kind of intelligent mobile enforcement system and method | |
| CN104392611B (en) | A kind of identification high price automobile method and system | |
| CN111191545B (en) | Real-time monitoring and analyzing system and method for driver behaviors | |
| CN111862656A (en) | Map updating method and device for intelligent carrier, storage medium and processor | |
| CN109767502A (en) | A kind of parking field control method and system based on recognition of face | |
| CN205211166U (en) | Vehicle information acquisition device that breaks rules and regulations based on on -vehicle driving recording apparatus | |
| CN112052878A (en) | Radar shielding identification method and device and storage medium | |
| CN111476102A (en) | Safety protection method, central control equipment and computer storage medium | |
| CN204929052U (en) | Internet security protection control intelligent management system | |
| CN109383516A (en) | A kind of anomaly analysis behavioral value system based on user behavior analysis | |
| CN115938128B (en) | Traffic accident prediction method, device, terminal and storage medium | |
| CN110473396A (en) | Traffic congestion data analysis method, device, electronic equipment and storage medium | |
| CN109410013A (en) | One kind obtaining VIN code based on mobile unit and identifies matched alarm method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180605 |