CN106781167A - The method and apparatus of monitoring object motion state - Google Patents
The method and apparatus of monitoring object motion state Download PDFInfo
- Publication number
- CN106781167A CN106781167A CN201611247277.XA CN201611247277A CN106781167A CN 106781167 A CN106781167 A CN 106781167A CN 201611247277 A CN201611247277 A CN 201611247277A CN 106781167 A CN106781167 A CN 106781167A
- Authority
- CN
- China
- Prior art keywords
- difference value
- picture
- video
- monitoring
- analyzed
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- 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
- G08B13/19608—Tracking movement of a target, e.g. by detecting an object predefined as a target, using target direction and or velocity to predict its new position
-
- 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
- G06V20/48—Matching video sequences
Abstract
Present invention is disclosed a kind of method and apparatus of monitoring object motion state, wherein, the method comprising the steps of：Carry out the selection of the first analysis mode and the second analysis mode；When the first analysis mode of selection is analyzed to the live real-time video recorded, monitoring process is divided into by multiple constant time ranges according to monitoring cycle；Each constant time range is analyzed according to characteristic set compares acquisition difference value, be organized into difference value set；The characteristic set is analyzed by REF video and drawn；Being more than setting value more than the difference value sum of setting value or difference value set when there is a difference value in difference value set, sending a warning.Periodic basis set is extended to by by reference pictures so that when moving object is monitored, judges more accurate, situations such as reducing wrong report, fail to report, improve monitoring accuracy, reduce artificial secondary judgement, reduce cost of labor.
Description
Technical field
The present invention relates to electronic monitoring field, especially relate to a kind of method of monitoring object motion state and set
It is standby.
Background technology
Mobile monitoring (Motion detection technology), is also motion detection, is usually used in unattended monitoring
Video recording and automatic alarm.Can be by central processing unit according to certain algorithm according to the picture that different frame per second are collected by camera
Calculated and compared, when picture is changed, such as someone is passed by, camera lens is moved, calculated the digital meeting that comparative result draws
More than threshold value and the system is indicated automatically to make corresponding treatment.Mobile monitoring allows that the change of picture can be recognized in designated area
Change, detect the presence of moving object and avoid the interference brought by light change.However, due to the dynamic change of background frame,
Such as influence of weather, illumination, shadow and chaotic interference so that motion detection turns into an extremely difficult job.
In actual application environment, can be by adjusting threshold values to indoor conventional environment, outdoor environment obtains relatively good
Monitoring effect.Mainly the major class implementation method of software and hardware two can also be carried out according to the classification of functional realiey
Qualitatively analyze.
Employ hardware to realize monitoring function, be not take up central processing unit (CPU), possess processing speed faster, thus
Some complex algorithms can be used to obtain more accurate monitoring result, and have good real-time.It is integrated to regard
Can be stored up for tableaux first by frequency mobile detector, detection circuit, afterwards, if it find that the variable quantity of picture exceedes
Value set in advance, system will send alarm signal, to remind security personnel or start video recorder.But hardware is realized also anticipating
Taste cost higher, once and system dynamic monitoring function is proposed and updates requirement higher, then hardware originally
System can only abandon and without, it is necessary to purchase new hardware, cause waste.
The monitoring function realized with software, takes the central processing unit (CPU) of main frame to complete numerical computations, and algorithm can not
It is too complicated, and amount of calculation should not be too big, the easily realization of influence monitoring system other functions (such as show, record a video).
Both the above scheme can accurately monitor stationary object according to picture analysis algorithm and hardware processing capability
Situation of movement, and carry out the treatment of follow-up data.But when object is that when being constantly in motion state, the monitoring method of the above is just
Accurate monitoring can not be made, situations such as wrong report usually occur, fail to report, causes to increase unnecessary artificial monitoring cost, seriously
When, the running of monitor area may be influenceed, cause huge loss.
The content of the invention
The main object of the present invention is to provide a kind of method and apparatus of monitoring object motion state, reduces monitoring wrong report.
In order to realize the goal of the invention, present invention firstly provides a kind of method of monitoring object motion state, including step
Suddenly：
Carry out the selection of the first analysis mode and the second analysis mode；
When the first analysis mode of selection is analyzed to the live real-time video recorded, will be monitored according to monitoring cycle
Journey is divided into multiple constant time ranges；
Each constant time range is analyzed according to characteristic set compares acquisition difference value, be organized into difference value set；It is described
Characteristic set is analyzed by REF video and drawn；
Set when being more than more than the difference value sum of setting value or difference value set in the presence of a difference value in difference value set
Definite value, sends a warning.
Further, the method for the monitoring object motion state, is analyzed by REF video in feature description and drawn,
Including step：
Record REF video；
The REF video duration is set as monitoring cycle, and the first frame picture of the REF video is calculated
Analysis, obtains the pictorial feature point；
The picture composition reference period picture group of specific frame number in the REF video is chosen, and calculates reference period picture
The characteristic point of each picture in group, constitutes periodic characteristic set.
Further, the method for the monitoring object motion state, it is described to carry out the first analysis mode and the second analysis side
The selection of formula, it is further comprising the steps of：
Floor reference pictures；
When the second analysis mode of selection is analyzed to the live real-time video recorded, and reference pictures are analyzed
Treatment, the picture after treatment is used as background frame；
The live real-time video picture of recording is made the difference with background frame, variance analysis is carried out to picture, extracted difference and draw
Face, and analyze difference value；
When difference value is more than setting value, send a warning.
Further, the method for the monitoring object motion state, it is described when in difference value set exist a difference value
Difference value sum more than setting value or difference value set is more than setting value, sends a warning or when difference value is more than setting value
When, after sending a warning, also include：
Record alarm real-time scene video；
Alarm real-time scene video is sent to default mailbox.
Further, the method for the monitoring object motion state, it is described to carry out the first analysis mode and the second analysis side
After the selection of formula, also include：
Alarm difference value is set；
Monitored area size and location parameter in monitoring picture is set；
Set and record the recording duration of alarm real-time scene video, record image quality and video format；
The transmission mailbox of alarm real-time scene video is set.
The present invention also provides a kind of equipment of monitoring object motion state, including：
Input module, the selection for carrying out the first analysis mode and the second analysis mode；
Processing module, for when the first analysis mode of selection is analyzed to the live real-time video recorded, according to prison
Monitoring process is divided into multiple constant time ranges by the survey cycle；
Processing module, acquisition difference value is compared for being analyzed to each constant time range according to characteristic set, is organized into difference
Different value set；The characteristic set is analyzed by REF video and drawn；
Alarm module, for when the difference for being more than setting value or difference value set in difference value set in the presence of a difference value
Value sum is more than setting value, sends a warning.
Further, the equipment of the monitoring object motion state, also includes：
Monitoring modular, for recording REF video；
Processing module, for the REF video duration to be set as into monitoring cycle, and by the first of the REF video
Frame picture carries out calculating analysis, obtains the pictorial feature point；
Processing module, the picture for choosing specific frame number in the REF video constitutes reference period picture group, and calculates
Go out the characteristic point of each picture in reference period picture group, constitute periodic characteristic set.
Further, the equipment of the monitoring object motion state, also includes：
Monitoring modular, for floor reference pictures；
Processing module, for when the second analysis mode of selection is analyzed to the live real-time video recorded, by benchmark
Picture is analyzed treatment, and the picture after treatment is used as background frame；
Processing module, for the live real-time video picture of recording to be made the difference with background frame, difference point is carried out to picture
Analysis, extracts difference picture, and analyze difference value；
Alarm module, for when difference value is more than setting value, sending a warning.
Further, the equipment of the monitoring object motion state, also includes：
Monitoring modular, for recording alarm real-time scene video；
Sending module, for alarm real-time scene video to be sent to default mailbox.
Further, the equipment of the monitoring object motion state, also includes：
Input module, for setting alarm difference value；
Input module, for setting monitored area size and location parameter in monitoring picture；
Input module, the recording duration of alarm real-time scene video is recorded for setting, image quality and video format is recorded.
Input module, the transmission mailbox for setting alarm real-time scene video.
The method and apparatus of monitoring object motion state of the invention, picture extension into periodic basis is monitored by by benchmark
Set so that when moving object is monitored, judges more accurate, situations such as reducing wrong report, fail to report, improves monitoring accuracy, reduces
The quantity of artificial secondary judgement, reduces cost of labor.
Brief description of the drawings
Fig. 1 is the analysis process figure in first analysis mode of one embodiment of the invention；
Fig. 2 is the first analysis mode flow chart of one embodiment of the invention；
Fig. 3 is the first analysis mode flow chart of another embodiment of the present invention；
Fig. 4 is the second analysis mode flow chart of one embodiment of the invention；
Fig. 5 is the recording alarm real-time scene video flow chart of one embodiment of the invention；
Fig. 6 is the predetermined registration operation flow chart of one embodiment of the invention；
Fig. 7 is the module annexation schematic diagram of a specific embodiment of the invention.
The realization of the object of the invention, functional characteristics and advantage will be described further referring to the drawings in conjunction with the embodiments.
Specific embodiment
It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not intended to limit the present invention.
Reference picture 1, the embodiment of the present invention proposes a kind of method of monitoring object motion state, including step：
S100, the selection for carrying out the first analysis mode and the second analysis mode；
S111, when choose the first analysis mode to record live real-time video be analyzed when, according to monitoring cycle will
Monitoring process is divided into multiple constant time ranges；
S112, each constant time range is analyzed according to characteristic set compares acquisition difference value, be organized into difference value set；
The characteristic set is analyzed by REF video and drawn；
It is S113, big more than the difference value sum of setting value or difference value set when there is difference value in difference value set
In setting value, send a warning.
Such as above-mentioned step S100, the selection of the first analysis mode and the second analysis mode is carried out, in the present embodiment,
The analysis mode includes two kinds, first analysis mode and the step S120-S122 of respectively described step S111-S113
The second analysis mode；
Such as above-mentioned step S111, when the first analysis mode of selection is analyzed to the live real-time video recorded, according to
Monitoring process is divided into multiple constant time ranges by monitoring cycle, and monitoring cycle section is automatically generated, and monitoring cycle section is at least a cycle,
One cycle stopped monitoring to be believed that before terminating, last feature picture in cycle picture group on the basis of monitoring commencing signal, when
Monitoring objective thing is constantly in motion state, and monitors the starting point that starting point is generally defaulted as a period of motion, when monitoring sets
Standby to monitor that live view is identical with last frame feature picture in cycle picture group, that is, bring into operation monitoring function, similarly,
In each monitoring cycle, during the monitoring of the last frame feature picture in each cycle, the monitoring in a new cycle will be started；
Such as above-mentioned step S112, each constant time range is analyzed according to characteristic set compares acquisition difference value, be organized into
Difference value set, obtains the description of the feature in the picture of each specified frame number in each constant time range and is compared with periodic characteristic set
Compared with mainly compare is worth change to the plane coordinates of individual characteristic point, so as to calculate feature difference value, and is organized into difference value collection
Close, comparing difference value is generally comprised and compared twice：The difference value of each specified frame number picture is and setting value during analyzing
It is compared, when occurring in which that a difference value for specified frame number picture is more than setting value, sends a warning, is analyzing
Cheng Hou, the difference value superposition summation of all specified frame number pictures draws difference set in same period, when difference value set is more than
Setting value, sends a warning, the setting value compared with single picture difference value and with setting that difference value set compares
Definite value, big I is differed；The characteristic set is analyzed by REF video and drawn；
Such as above-mentioned step S113, when the difference for being more than setting value or difference value set in difference value set in the presence of a difference value
Different value sum is more than setting value, sends a warning, and the warning information is generally warning tones, and short message can also be according to demand
Add the forced termination signal to moving target.
Reference picture 2, in the present embodiment, the method for the monitoring object motion state passes through base in feature description
Quasi- video analysis draw, including step：
S101 records REF video；
The REF video duration is set as monitoring cycle by S102, and the first frame picture of the REF video is carried out
Analysis is calculated, the pictorial feature point is obtained；
S103 chooses the picture composition reference period picture group of specific frame number in the REF video, and calculates reference period
The characteristic point of each picture in picture group, constitutes periodic characteristic set.
Such as above-mentioned step S101, the next complete action cycle video of monitoring objective ordinary running condition is recorded, as benchmark
Video, the REF video acquiescence of length includes 24 frame pictures within each second, and picture frame number can adjust, each second video institute
The upper limit containing frame number is constant according to monitoring device own specifications, is usually no more than 100 frames；
Such as above-mentioned step S102, the REF video duration is set as monitoring cycle, and by the of the REF video
One frame picture carries out calculating analysis, obtains the pictorial feature point；The REF video duration is set as monitoring cycle, for example,
A length of two points 18 seconds during the REF video, then in monitoring process every two points judge week for once complete analysis within 18 seconds
Phase, each cycle is interior in the case where there are not abnormal conditions, the target trajectory and each movement locus in monitored area
Duration required for producing or continuing is extremely similar or identical, and similarity degree is not less than setting value, and by the REF video
First frame picture carries out calculating analysis, obtains the pictorial feature point, and pictorial feature point is generally comprised, the color protrusion in picture
Point, the profile sideline random point of monitoring objective in picture, it is believed that especially selected impact point, pictorial feature point quantity is general at least
Including 30；The gray value of characteristic point is combined into binary data string, and using binary data string as this feature point
Feature is described, and feature description also includes the plane coordinate value of this feature point, and same coordinate value can only be corresponded in same picture
A unique characteristic point, but multiple characteristic points of same grayscale value are may occur in which in same picture；
Such as above-mentioned step S103, picture composition the reference period picture group, example of specific frame number in the REF video are chosen
Such as, in REF video a length of ten seconds, it is per second including 30 frame pictures under conditions of, select the 1st, 6,12,18 ... 288,294,
300 frames totally five ten pictures used as reference period picture group, selected frame number can need and monitor according to the precision of monitoring objective
The length change in cycle, general each frame picture must be spaced no less than three frames, calculate each picture pair in reference period picture group
The plane coordinate value of the feature description answered, draws in the feature description point for same gray value occur is than step S101-S102
When the feature description point of the same gray value of reference pictures is more, several pictures before the picture in reference period picture group are recalled
Same grey value characteristics point in the region for newly-increased feature point occur is carried out path analysis by (at most 3), is drawn by several
The characteristic point moving curve that picture is constituted, chose with the closest characteristic point of curve as validity feature point, by benchmark week
The feature description of the characteristic point of each picture collects composition periodic characteristic set in phase picture group；
Reference picture 3, in another specific embodiment, the analysis mode can also monitor static target, static monitoring
During target, the step S103 is stopped, the step S102 continues to use part turns into step S141, step S111 described in function,
S112, S113 are replaced with the following steps S140, S142, S143, only need to be by the first frame picture of the REF video as benchmark
Picture, using the feature description in the characteristic point of the reference pictures as the analysis benchmark of monitoring process, comprises the following steps that：
S101, recording REF video；S140, the first frame picture of the REF video is carried out into calculating analysis, obtain the pictorial feature
Point；S141, the gray value of characteristic point is combined into binary string, and is described binary string as the feature of this feature point；
S142, video pictures are compared with feature description, calculate difference value；143rd, when difference value is more than setting value, send
Warning information.
Reference picture 4, in the present embodiment, the method for the monitoring object motion state is described to carry out the first analysis mode
And second analysis mode selection, it is further comprising the steps of：
S120, floor reference pictures；
S121, when choosing the second analysis mode the live real-time video recorded being analyzed, and reference pictures are entered
Row analyzing and processing, the picture after treatment is used as background frame；
S122, the live real-time video picture and background frame of recording are made the difference, variance analysis is carried out to picture, it is poor to extract
Different picture, and analyze difference value；
S123, when difference value be more than setting value when, send a warning.
Such as above-mentioned step S120, the picture under shooting and monitoring target normal rest state, as reference pictures；
Such as above-mentioned step S121, reference pictures are carried out into gray processing and filtering process, dispel it is variegated wait influence monitoring because
Element；Picture after gray processing is carried out into edge detection process, the edge picture of only monitoring objective and background is obtained, after treatment
Picture as background frame；
Such as above-mentioned step S122, real-time video picture and background frame are made the difference, draw real-time video picture and background painting
The discrepancy in face, and the picture is carried out into binaryzation, difference picture is drawn, difference picture is extracted, and according to the difference of difference picture
DRS degree analyzes difference value；
Such as above-mentioned step S123, when difference value is more than setting value, send a warning, the warning information is generally police
Sound is accused, short message can also according to demand add the forced termination signal to moving target.
Reference picture 5, in the present embodiment, the method for the monitoring object motion state, described when in difference value set
It is more than setting value more than the difference value sum of setting value or difference value set in the presence of a difference value, sends a warning or be on duty
When different value is more than setting value, after sending a warning, also include：
S130, recording alarm real-time scene video；
S131, the real-time scene video that will alarm are sent to default mailbox.
Such as above-mentioned step S130, alarm real-time scene video is recorded, when there is alarm condition, sending warning message
While, equipment will can occur from alarm condition after a period of time in situation got off by videograph, the length one of video
As be at most 10 minutes；
Such as above-mentioned step S131, after the alarm real-time scene video is recorded, alarm real-time scene video is passed through into nothing
Line or cable network are sent to default mailbox, and the mailbox is generally one, and multiple reception mailboxes can be set.
Reference picture 6, in the present embodiment, the method for the monitoring object motion state, carries out the first analysis at the S100
After the selection of mode and the second analysis mode, also include：
S10, setting alarm difference value；
S20, the monitored area size and location parameter set in monitoring picture；
S30, the recording duration that recording alarm real-time scene video is set, recording image quality and video format；
S40, the transmission mailbox that alarm real-time scene video is set.
Such as above-mentioned step S10, alarm difference value is set, the difference value is the percentages of 1-100, and the alarm is poor
The different smaller analysis precision of value numerical value is more accurate；
Such as above-mentioned step S20, the monitored area size and location parameter in monitoring picture are set, are set in video pictures
Monitored area, monitored area can be made up of multiple regions, or single region, the size and plane coordinates in each region
Parameter can according to circumstances be set；
Such as above-mentioned step S30, the length of the recording time for recording alarm real-time scene video is set, is recorded image quality and is regarded
Frequency form, the length of the recording time of video is generally at most 10 minutes, records image quality and generally comprises SD, high definition and super clear
Three alternative recording image quality, general default setting is SD image quality, video format one remove including MP4, MOV, RM and
WMV, general default setting is MP4 forms.
Such as above-mentioned step S40, the transmission mailbox of alarm real-time scene video is set, the mailbox is generally one, can set
Put multiple reception mailboxes.
Reference picture 7, the embodiment of the present invention provides a kind of equipment of monitoring object motion state, including：Input module 100,
Selection for carrying out the first analysis mode and the second analysis mode；Processing module 120, for when the first analysis mode of selection
When live real-time video to recording is analyzed, monitoring process is divided into by multiple constant time ranges according to monitoring cycle；Processing module
120, compare acquisition difference value for being analyzed to each constant time range according to characteristic set, it is organized into difference value set；It is described
Characteristic set is analyzed by REF video and drawn；Alarm module 130, for when being more than in the presence of a difference value in difference value set
The difference value sum of setting value or difference value set is more than setting value, sends a warning.
Above-mentioned input module 100, for selecting analysis mode, in the present embodiment, the analysis mode includes two kinds, point
Not Wei the step S100-S113 the first analysis mode and second analysis mode of the step S120-S123；
Above-mentioned processing module 120, for monitoring process to be divided into multiple constant time ranges according to monitoring cycle, monitoring cycle section is certainly
Dynamic to generate, monitoring cycle section is at least a cycle, and monitoring is stopped to be believed that before terminating in a cycle, and monitoring commencing signal is base
Last feature picture in paracycle picture group, when monitoring objective thing is constantly in motion state, and monitors starting point and generally writes from memory
The starting point of a period of motion is thought, when monitoring device monitors live view with last frame feature picture in cycle picture group
Identical, that is, bring into operation monitoring function, similarly, in each monitoring cycle, the prison of the last frame feature picture in each cycle
During survey, the monitoring in a new cycle will be started；
Above-mentioned processing module 120, for being analyzed to each constant time range according to periodic characteristic set, obtains each cycle
Feature in section in the picture of each specified frame number is described and is compared with periodic characteristic set, is mainly compared to individual characteristic point
Plane coordinates is worth change, so as to calculate feature difference value, and is organized into difference value set, and comparing difference value generally comprises two
It is secondary to compare：The difference value of each specified frame number picture is compared with setting value during analyzing, when occurring in which one
When specifying the difference value of frame number picture to be more than setting value, send a warning, it is after the analysis is complete, all in same period to specify
The difference value superposition summation of frame number picture draws difference set, when difference value set is more than setting value, sends a warning, described
The setting value compared with single picture difference value and the setting value compared with difference value set, big I are differed；
Above-mentioned alarm module 130, for being more than setting value or difference value collection when there is difference value in difference value set
The difference value sum of conjunction is more than setting value, sends a warning, and the warning information is generally warning tones, and short message can also root
The forced termination signal to moving target is added according to demand.
In the present embodiment, the equipment of the monitoring object motion state, also includes：Monitoring modular 110, for recording base
Quasi- video；Processing module 120, for the REF video duration to be set as into monitoring cycle, and by the of the REF video
One frame picture carries out calculating analysis, obtains the pictorial feature point；Processing module 120, it is specific in the REF video for choosing
The picture composition reference period picture group of frame number, and the characteristic point of each picture in reference period picture group is calculated, the composition cycle is special
Collection is closed.
Above-mentioned monitoring modular 110, for recording the next complete action cycle video of monitoring objective ordinary running condition, as
REF video, the REF video acquiescence of length includes 24 frame pictures within each second, and picture frame number can adjust, regard within each second
The contained frame number upper limit of frequency is constant according to the own specifications of the monitoring modular 110, is usually no more than 100 frames；
Above-mentioned processing module 120, for the REF video duration to be set as into monitoring cycle, for example, the benchmark is regarded
A length of two points 18 seconds during frequency, then in monitoring process every two points judge cycle, each week for once complete analysis within 18 seconds
In phase in the case where there are not abnormal conditions, the generation or continuity of target trajectory and each movement locus in monitored area
Required duration is extremely similar or identical, and similarity degree is not less than setting value, and by the first frame picture of the REF video
Calculating analysis is carried out, the pictorial feature point is obtained, pictorial feature point is generally comprised, the color protruding point in picture, supervised in picture
Survey the profile sideline random point of target, it is believed that especially selected impact point, pictorial feature point quantity typically at least includes 30,
The gray value of characteristic point is combined into binary data string, and is described binary data string as the feature of this feature point, it is special
The plane coordinate value that description also includes this feature point is levied, same coordinate value can only correspond to a unique spy in same picture
Levy a little, but multiple characteristic points of same grayscale value are may occur in which in same picture；
Above-mentioned processing module 120, the picture for choosing specific frame number in the REF video constitutes reference period picture
Group, for example, in REF video a length of ten seconds, it is per second including 30 frame pictures under conditions of, select the 1st, 6,12,18 ... 288,
294th, 300 frames totally five ten pictures as reference period picture group, selected frame number can be needed according to the precision of monitoring objective and
The length change of monitoring cycle, general each frame picture must be spaced no less than three frames, calculated and respectively draw in reference period picture group
The plane coordinate value of the corresponding feature description in face, the base drawn in the feature description point for same gray value occur is than step S102
When the feature description point of the same gray value of quasi- picture is more, several pictures in reference period picture group before the picture are recalled (extremely
Many 3) the same grey value characteristics point in the region for newly-increased feature point occur is carried out into path analysis, draw by several pictures
The characteristic point moving curve of composition, is chosen with the closest characteristic point of curve as validity feature point, and reference period is drawn
The feature description of the characteristic point of each picture collects composition periodic characteristic set in the group of face.
In another specific embodiment, the analysis mode can also monitor static target, when static target is monitored, institute
Stating monitoring modular 110 only need to be by the first frame picture of the REF video as reference pictures, and the processing module 120 is by the base
Feature in the characteristic point of quasi- picture is described as the analysis benchmark of monitoring process, specific as follows：Monitoring modular 110, for recording
REF video processed；Monitoring modular 110, for studio real-time video；Processing module 120, for by the REF video
First frame picture carries out calculating analysis, obtains the pictorial feature point；Processing module 120, for the gray value of characteristic point to be combined
Into binary string, and described binary string as the feature of this feature point；Processing module 120, for video pictures with it is described
Feature description is compared, and calculates difference value；Alarm module 130, for being more than setting value when difference value, sends alarm signal
Breath.
In the present embodiment, the equipment of the monitoring object motion state, also includes：Monitoring modular 110, it is existing for shooting
Field reference pictures；Processing module 120, for when choose the second analysis mode to record live real-time video be analyzed when,
Reference pictures are analyzed treatment, the picture after treatment is used as background frame；Processing module 120, for the scene that will be recorded
Real-time video picture is made the difference with background frame, and variance analysis is carried out to picture, extracts difference picture, and analyze difference value；Report
Alert module 130, for when difference value is more than setting value, sending a warning.
Above-mentioned monitoring modular 110, for the picture under shooting and monitoring target normal rest state, as reference pictures；
Above-mentioned processing module 120, for reference pictures to be carried out into gray processing and filtering process, dispelling variegated grade influences prison
Survey factor, and the picture after gray processing is carried out into edge detection process, the edge picture of only monitoring objective and background is obtained,
Picture after treatment is used as background frame；
Above-mentioned processing module 120, for real-time video picture and background frame to be made the difference, draws real-time video picture with the back of the body
The discrepancy of scape picture, and the picture is carried out into binaryzation, difference picture is drawn, difference picture is extracted, and according to difference picture
Difference degree analyze difference value；
Above-mentioned alarm module 130, for when difference value is more than setting value, sending a warning, the warning information one
As be warning tones, short message can also according to demand add the forced termination signal to moving target.
In the present embodiment, the equipment of the monitoring object motion state, also includes：Monitoring modular 110, for reality of alarming
When scene video；Sending module 140, for alarm real-time scene video to be sent to default mailbox.
Above-mentioned monitoring modular 110, is additionally operable to record alarm real-time scene video, when there is alarm condition, is sending
While warning message, the monitoring modular 110 can will occur from alarm condition after a period of time in situation by video remember
Record is got off, and the length of video is generally at most 10 minutes；
Above-mentioned sending module 140, for after the alarm real-time scene video is recorded, alarm real-time scene video being led to
Cross wirelessly or non-wirelessly network to send to default mailbox, the mailbox is generally one, multiple reception mailboxes can be set.
In the present embodiment, the equipment of the monitoring object motion state, also includes：Input module 100, reports for setting
Alert difference value；Input module 100, for setting monitored area size and location parameter in monitoring picture；Input module
100, the recording duration of alarm real-time scene video is recorded for setting, image quality and video format is recorded.Input module 100,
Transmission mailbox for setting alarm real-time scene video.
Above-mentioned input module 100, is used for；Alarm difference value is set, and the difference value is the percentages of 1-100, institute
State the alarm smaller analysis precision of difference value numerical value more accurate；
Above-mentioned input module 100, for setting monitored area size and location parameter in monitoring picture, video pictures
Middle setting monitored area, monitored area can be made up of multiple regions, or single region, the size peace in each region
Areal coordinate parameter can according to circumstances be set；
Above-mentioned input module 100, for setting the length of the recording time for recording alarm real-time scene video, recording image quality
And video format, the length of the recording time of video is generally at most 10 minutes, record image quality generally comprise SD, high definition with
And super clear three alternative recording image quality, general default setting be SD image quality, video format one remove including MP4, MOV,
RM and WMV, general default setting is MP4 forms.
Above-mentioned input module 100, is additionally operable to set the transmission mailbox of alarm real-time scene video, and the mailbox is generally one
It is individual, multiple reception mailboxes can be set.
The preferred embodiments of the present invention are the foregoing is only, the scope of the claims of the invention, every utilization is not thereby limited
Equivalent structure or equivalent flow conversion that description of the invention and accompanying drawing content are made, or directly or indirectly it is used in other correlations
Technical field, be included within the scope of the present invention.
Claims (10)
1. a kind of method of monitoring object motion state, it is characterised in that including step：
Carry out the selection of the first analysis mode and the second analysis mode；
When the first analysis mode of selection is analyzed to the live real-time video recorded, monitoring process is divided according to monitoring cycle
Into multiple constant time ranges；
Each constant time range is analyzed according to characteristic set compares acquisition difference value, be organized into difference value set；The feature
Set is analyzed by REF video and drawn；
It is more than setting value more than the difference value sum of setting value or difference value set when there is a difference value in difference value set,
Send a warning.
2. the method for monitoring object motion state according to claim 1, it is characterised in that pass through in feature description
REF video analysis draws, including step：
Record REF video；
The REF video duration is set as monitoring cycle, and the first frame picture of the REF video is carried out into calculating point
Analysis, obtains the pictorial feature point；
Choose the picture composition reference period picture group of specific frame number in the REF video, and calculate reference period picture group
The characteristic point of each picture, constitutes periodic characteristic set.
3. the method for monitoring object motion state according to claim 1, it is characterised in that described to carry out the first analysis side
The selection of formula and the second analysis mode, it is further comprising the steps of：
Floor reference pictures；
When the second analysis mode of selection is analyzed to the live real-time video recorded, and reference pictures are analyzed place
Reason, the picture after treatment is used as background frame；
The live real-time video picture of recording is made the difference with background frame, variance analysis is carried out to picture, extract difference picture, and
Analyze difference value；
When difference value is more than setting value, send a warning.
4. the method for the monitoring object motion state according to claim 1 or 3, it is characterised in that work as difference value described
There is a difference value in set and be more than setting value more than the difference value sum of setting value or difference value set, send a warning
Or when difference value is more than setting value, after sending a warning, also include：
Record alarm real-time scene video；
Alarm real-time scene video is sent to default mailbox.
5. the method for monitoring object motion state according to claim 1, it is characterised in that carry out the first analysis described
After the selection of mode and the second analysis mode, also include：
Alarm difference value is set；
Monitored area size and location parameter in monitoring picture is set；
Set and record the recording duration of alarm real-time scene video, record image quality and video format；
The transmission mailbox of alarm real-time scene video is set.
6. a kind of equipment of monitoring object motion state, it is characterised in that including：
Input module, the selection for carrying out the first analysis mode and the second analysis mode；
Processing module, for when the first analysis mode of selection is analyzed to the live real-time video recorded, according to monitoring week
Monitoring process is divided into multiple constant time ranges by the phase；
Processing module, acquisition difference value is compared for being analyzed to each constant time range according to characteristic set, is organized into difference value
Set；The characteristic set is analyzed by REF video and drawn；
Alarm module, for when exist in difference value set difference value more than setting value or difference value set difference value it
With more than setting value, send a warning.
7. the equipment of monitoring object motion state according to claim 6, it is characterised in that also include：
Monitoring modular, for recording REF video；
Processing module, for the REF video duration to be set as into monitoring cycle, and the first frame of the REF video is drawn
Face carries out calculating analysis, obtains the pictorial feature point；
Processing module, the picture for choosing specific frame number in the REF video constitutes reference period picture group, and calculates base
The characteristic point of each picture in paracycle picture group, constitutes periodic characteristic set.
8. the equipment of monitoring object motion state according to claim 6, it is characterised in that also include：
Monitoring modular, for floor reference pictures；
Processing module, for when the second analysis mode of selection is analyzed to the live real-time video recorded, by reference pictures
Treatment is analyzed, the picture after treatment is used as background frame；
Processing module, for the live real-time video picture of recording to be made the difference with background frame, variance analysis is carried out to picture, is carried
Difference picture is taken, and analyzes difference value；
Alarm module, for when difference value is more than setting value, sending a warning.
9. the equipment of the monitoring object motion state according to claim 6 or 8, it is characterised in that also include：
Monitoring modular, for recording alarm real-time scene video；
Sending module, for alarm real-time scene video to be sent to default mailbox.
10. the equipment of monitoring object motion state according to claim 6, it is characterised in that also include：
Input module, for setting alarm difference value；
Input module, for setting monitored area size and location parameter in monitoring picture；
Input module, the recording duration of alarm real-time scene video is recorded for setting, image quality and video format is recorded.
Input module, the transmission mailbox for setting alarm real-time scene video.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611247277.XA CN106781167B (en) | 2016-12-29 | 2016-12-29 | Method and device for monitoring motion state of object |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611247277.XA CN106781167B (en) | 2016-12-29 | 2016-12-29 | Method and device for monitoring motion state of object |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106781167A true CN106781167A (en) | 2017-05-31 |
| CN106781167B CN106781167B (en) | 2021-03-02 |
Family
ID=58927722
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611247277.XA Active CN106781167B (en) | 2016-12-29 | 2016-12-29 | Method and device for monitoring motion state of object |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106781167B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108492502A (en) * | 2018-03-16 | 2018-09-04 | 深圳勇艺达机器人有限公司 | Intelligent robot monocular wide-angle vision anti-theft alarming method |
| CN109064715A (en) * | 2018-08-08 | 2018-12-21 | 国网山东省电力公司惠民县供电公司 | A kind of monitoring system and monitoring method of for transformer |
| CN111263108A (en) * | 2019-12-20 | 2020-06-09 | 中译语通文娱科技(青岛)有限公司 | Intelligent production video monitoring method |
| CN113793366A (en) * | 2021-08-31 | 2021-12-14 | 北京达佳互联信息技术有限公司 | Image processing method, device, equipment and storage medium |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5351078A (en) * | 1954-12-24 | 1994-09-27 | Lemelson Medical, Education & Research Foundation Limited Partnership | Apparatus and methods for automated observation of objects |
| EP0700515B1 (en) * | 1993-05-28 | 2000-09-06 | Millennium Venture Holdings Ltd. | An automatic inspection apparatus |
| US20020150278A1 (en) * | 2000-10-10 | 2002-10-17 | Sick Ag | Apparatus and a method for the detection of objects |
| US7173666B1 (en) * | 2002-08-22 | 2007-02-06 | Smal Camera Technologies | System and method for displaying a non-standard aspect ratio image on a standard aspect ratio monitor |
| CN101720031A (en) * | 2009-12-22 | 2010-06-02 | 北京理工大学 | Difference method-based indoor video monitoring and alarming method |
| CN103286782A (en) * | 2013-06-07 | 2013-09-11 | 上海发那科机器人有限公司 | Flexible tracking and positioning system and flexible tracking and positioning method of robot |
| CN104516337A (en) * | 2013-10-01 | 2015-04-15 | 美国博通公司 | Gesture-Based Industrial Monitoring |
| CN105282521A (en) * | 2015-11-21 | 2016-01-27 | 浙江宇视科技有限公司 | Method and device for detecting motion during IP camera cruising process |
| CN106131375A (en) * | 2016-06-29 | 2016-11-16 | 联想(北京)有限公司 | Equipment and motion state detection method |
-
2016
- 2016-12-29 CN CN201611247277.XA patent/CN106781167B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5351078A (en) * | 1954-12-24 | 1994-09-27 | Lemelson Medical, Education & Research Foundation Limited Partnership | Apparatus and methods for automated observation of objects |
| EP0700515B1 (en) * | 1993-05-28 | 2000-09-06 | Millennium Venture Holdings Ltd. | An automatic inspection apparatus |
| US20020150278A1 (en) * | 2000-10-10 | 2002-10-17 | Sick Ag | Apparatus and a method for the detection of objects |
| US7173666B1 (en) * | 2002-08-22 | 2007-02-06 | Smal Camera Technologies | System and method for displaying a non-standard aspect ratio image on a standard aspect ratio monitor |
| CN101720031A (en) * | 2009-12-22 | 2010-06-02 | 北京理工大学 | Difference method-based indoor video monitoring and alarming method |
| CN103286782A (en) * | 2013-06-07 | 2013-09-11 | 上海发那科机器人有限公司 | Flexible tracking and positioning system and flexible tracking and positioning method of robot |
| CN104516337A (en) * | 2013-10-01 | 2015-04-15 | 美国博通公司 | Gesture-Based Industrial Monitoring |
| CN105282521A (en) * | 2015-11-21 | 2016-01-27 | 浙江宇视科技有限公司 | Method and device for detecting motion during IP camera cruising process |
| CN106131375A (en) * | 2016-06-29 | 2016-11-16 | 联想(北京)有限公司 | Equipment and motion state detection method |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108492502A (en) * | 2018-03-16 | 2018-09-04 | 深圳勇艺达机器人有限公司 | Intelligent robot monocular wide-angle vision anti-theft alarming method |
| CN109064715A (en) * | 2018-08-08 | 2018-12-21 | 国网山东省电力公司惠民县供电公司 | A kind of monitoring system and monitoring method of for transformer |
| CN111263108A (en) * | 2019-12-20 | 2020-06-09 | 中译语通文娱科技(青岛)有限公司 | Intelligent production video monitoring method |
| CN113793366A (en) * | 2021-08-31 | 2021-12-14 | 北京达佳互联信息技术有限公司 | Image processing method, device, equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106781167B (en) | 2021-03-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10070053B2 (en) | Method and camera for determining an image adjustment parameter | |
| CN110222640B (en) | Method, device and method for identifying suspect in monitoring site and storage medium | |
| CN108965826B (en) | Monitoring method, monitoring device, processing equipment and storage medium | |
| KR102058452B1 (en) | IoT Convergence Intelligent Video Analysis Platform System | |
| CN106781167A (en) | The method and apparatus of monitoring object motion state | |
| US20110025847A1 (en) | Service management using video processing | |
| US10937132B2 (en) | Spike signal-based display method and system | |
| JP5388829B2 (en) | Intruder detection device | |
| CN105120217A (en) | Intelligent camera motion detection alarm system and method based on big data analysis and user feedback | |
| US11893796B2 (en) | Methods and systems for detection of anomalous motion in a video stream and for creating a video summary | |
| US10152630B2 (en) | Methods and systems of performing blob filtering in video analytics | |
| CN102348128A (en) | Surveillance camera system having camera malfunction detection function | |
| CN107920223B (en) | Object behavior detection method and device | |
| WO2006118559A1 (en) | Automatic video quality monitoring for surveillance cameras | |
| KR20190038137A (en) | Image Analysis Method and Server Apparatus for Per-channel Optimization of Object Detection | |
| CN107330414A (en) | Act of violence monitoring method | |
| CN111228706A (en) | GIS-based building interior fire fighting equipment control method and device | |
| US10593047B2 (en) | Setting of a motion trigger level | |
| CN116133031A (en) | Building network quality assessment method, device, electronic equipment and medium | |
| CN115965894A (en) | Patrol robot control method and device, electronic equipment and storage medium | |
| CN114418388A (en) | Personnel management method and device based on image recognition and storage medium | |
| CN114584758A (en) | City-level monitoring video quality assessment method and system | |
| KR101169631B1 (en) | Video analysis camera, video analysis system having detection interface and detecting adjustment method of the same | |
| TWI784718B (en) | Method and system for processing alarm event in factory | |
| KR100950734B1 (en) | Automatic Recognition Method of Abnormal Status at Home Surveillance System and Internet Refrigerator |
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 |