CN106781167B - Method and device for monitoring motion state of object - Google Patents
Method and device for monitoring motion state of object Download PDFInfo
- Publication number
- CN106781167B CN106781167B CN201611247277.XA CN201611247277A CN106781167B CN 106781167 B CN106781167 B CN 106781167B CN 201611247277 A CN201611247277 A CN 201611247277A CN 106781167 B CN106781167 B CN 106781167B
- Authority
- CN
- China
- Prior art keywords
- monitoring
- picture
- video
- difference value
- period
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Closed-Circuit Television Systems (AREA)
Abstract
The invention discloses a method and equipment for monitoring the motion state of an object, wherein the method comprises the following steps: selecting a first analysis mode and a second analysis mode; when a first analysis mode is selected to analyze the recorded live real-time video, dividing the monitoring process into a plurality of period segments according to the monitoring period; analyzing and comparing each period section according to the feature set to obtain a difference value, and sorting the difference value into a difference value set; the characteristic set is obtained through reference video analysis; and when one difference value in the difference value set is greater than the set value or the sum of the difference values in the difference value set is greater than the set value, sending alarm information. By expanding the reference picture into the periodic reference set, the judgment is more accurate when a moving object is monitored, the situations of false alarm, missing report and the like are reduced, the monitoring accuracy is improved, the artificial secondary judgment is reduced, and the labor cost is reduced.
Description
Technical Field
The invention relates to the field of electronic monitoring, in particular to a method and equipment for monitoring the motion state of an object.
Background
Motion detection (Motion detection technology), also called Motion detection, is commonly used for unattended surveillance video and automatic alarm. The pictures acquired by the camera according to different frame rates are calculated and compared by the central processing unit according to a certain algorithm, when the pictures are changed, if people walk, the lens is moved, the number obtained by calculating the comparison result exceeds the threshold value, and the system is indicated to automatically perform corresponding processing. Motion monitoring allows the change of the picture to be recognized in a designated area, the presence of moving objects to be detected and the interference caused by the change of light to be avoided. However, motion detection is a difficult task due to dynamic changes of the background picture, such as weather, illumination, shadows, and clutter.
In the practical application environment, the monitoring effect on the indoor common environment and the outdoor environment can be better through adjusting the valve value. The method can be classified according to the function realization method, and mainly carries out qualitative analysis on two main realization methods of software and hardware.
The hardware is adopted to realize the monitoring function, a Central Processing Unit (CPU) is not occupied, and the processing speed is higher, so that more complex algorithms can be adopted to obtain more accurate monitoring results, and the real-time performance is good. The integrated video mobile detector is characterized in that a detection circuit firstly stores a static picture, and then if the variable quantity of the picture exceeds a preset value, a system sends out an alarm signal to remind security personnel or start a video recorder. However, the hardware implementation also means higher cost, and once the system puts higher requirements on updating the dynamic monitoring function, the original hardware system can only be abandoned, and new hardware must be purchased, which causes waste.
The monitoring function realized by software occupies a Central Processing Unit (CPU) of a host to complete numerical calculation, the algorithm cannot be too complex, the calculated amount is not too large, and the realization of other functions (such as display, video recording and the like) of the monitoring system is easily influenced.
The two schemes can accurately monitor the moving condition of the static object according to the picture analysis algorithm and the hardware processing capacity, and process subsequent data. However, when the object is in motion state all the time, the above monitoring method cannot perform accurate monitoring, and situations such as false alarm, missing report and the like often occur, which increases unnecessary human monitoring cost, and in severe cases, the operation of the monitoring area may be affected, resulting in huge loss.
Disclosure of Invention
The invention mainly aims to provide a method and equipment for monitoring the motion state of an object, which can reduce the monitoring false alarm.
In order to achieve the object, the present invention firstly provides a method for monitoring the motion state of an object, comprising the steps of:
selecting a first analysis mode and a second analysis mode;
when a first analysis mode is selected to analyze the recorded live real-time video, dividing the monitoring process into a plurality of period segments according to the monitoring period;
analyzing and comparing each period section according to the feature set to obtain a difference value, and sorting the difference value into a difference value set; the characteristic set is obtained through reference video analysis;
and when one difference value in the difference value set is greater than the set value or the sum of the difference values in the difference value set is greater than the set value, sending alarm information.
Further, the method for monitoring the motion state of the object, wherein the characteristic description is obtained by reference video analysis, comprises the following steps:
recording a reference video;
setting the duration of the reference video as a monitoring period, and calculating and analyzing a first frame of picture of the reference video to obtain picture characteristic points;
and selecting pictures with specific frame numbers in the reference video to form a reference period picture group, and calculating the characteristic points of each picture in the reference period picture group to form a period characteristic set.
Further, the method for monitoring the motion state of the object, which selects the first analysis mode and the second analysis mode, further comprises the following steps:
shooting a site reference picture;
when a second analysis mode is selected to analyze the recorded live real-time video, analyzing and processing the reference picture, and taking the processed picture as a background picture;
the difference is made between the recorded live real-time video picture and the background picture, the difference analysis is carried out on the picture, the difference picture is extracted, and the difference value is analyzed;
and when the difference value is larger than the set value, sending alarm information.
Further, the method for monitoring the motion state of the object, when there is a difference value in the difference value set that is greater than the set value or the sum of the difference values in the difference value set is greater than the set value, and sends out the alarm information or when the difference value is greater than the set value, after sending out the alarm information, further includes:
recording an alarm real-time scene video;
and sending the alarm real-time scene video to a preset mailbox.
Further, the method for monitoring the motion state of the object, after the selection of the first analysis mode and the second analysis mode, further includes:
setting a set value;
setting the size and position parameters of a monitoring area in a monitoring picture;
setting recording duration, recording image quality and video format of the recorded alarm real-time scene video;
and setting a sending mailbox for the alarm real-time scene video.
The present invention also provides an apparatus for monitoring the state of motion of an object, comprising:
the input module is used for selecting a first analysis mode and a second analysis mode;
the processing module is used for dividing the monitoring process into a plurality of period segments according to the monitoring period when the first analysis mode is selected to analyze the recorded on-site real-time video;
the processing module is used for analyzing and comparing each period section according to the feature set to obtain a difference value and sorting the difference value into a difference value set; the characteristic set is obtained through reference video analysis;
and the alarm module is used for sending alarm information when one difference value in the difference value set is greater than a set value or the sum of the difference values in the difference value set is greater than the set value.
Further, the apparatus for monitoring the motion state of the object further comprises:
the monitoring module is used for recording a reference video;
the processing module is used for setting the duration of the reference video as a monitoring period and calculating and analyzing a first frame of picture of the reference video to obtain picture characteristic points;
and the processing module is used for selecting pictures with specific frame numbers in the reference video to form a reference period picture group, calculating the characteristic points of all the pictures in the reference period picture group and forming a period characteristic set.
Further, the apparatus for monitoring the motion state of the object further comprises:
the monitoring module is used for shooting a site reference picture;
the processing module is used for analyzing the reference picture when a second analysis mode is selected to analyze the recorded live real-time video, and the processed picture is used as a background picture;
the processing module is used for carrying out difference analysis on the recorded live real-time video picture and a background picture, extracting a difference picture and analyzing a difference value;
and the alarm module is used for sending alarm information when the difference value is greater than the set value.
Further, the apparatus for monitoring the motion state of the object further comprises:
the monitoring module is used for recording an alarm real-time scene video;
and the sending module is used for sending the alarm real-time scene video to a preset mailbox.
Further, the apparatus for monitoring the motion state of the object further comprises:
the input module is used for setting a set value;
the input module is used for setting the size and the position parameters of a monitoring area in a monitoring picture;
and the input module is used for setting the recording duration, the recording image quality and the video format of the recorded alarm real-time scene video.
And the input module is used for setting a sending mailbox of the alarm real-time scene video.
According to the method and the device for monitoring the motion state of the object, the reference monitoring picture is expanded into the periodic reference set, so that the judgment is more accurate when the moving object is monitored, the situations of false alarm, missing report and the like are reduced, the monitoring accuracy is improved, the number of artificial secondary judgment is reduced, and the labor cost is reduced.
Drawings
FIG. 1 is a flow chart of analysis in a first analysis mode according to an embodiment of the present invention;
FIG. 2 is a flow chart of a first analysis mode according to an embodiment of the present invention;
FIG. 3 is a flow chart of a first analysis mode of another embodiment of the present invention;
FIG. 4 is a flow chart of a second analysis mode of an embodiment of the present invention;
FIG. 5 is a flow chart of a video recording an alarm real-time scene in accordance with an embodiment of the present invention;
FIG. 6 is a flow chart of the default operation of one embodiment of the present invention;
fig. 7 is a schematic diagram of a module connection relationship according to an embodiment of the invention.
The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Referring to fig. 1, an embodiment of the present invention provides a method for monitoring a motion state of an object, including:
s100, selecting a first analysis mode and a second analysis mode;
s111, when a first analysis mode is selected to analyze the recorded live real-time video, dividing a monitoring process into a plurality of period segments according to a monitoring period;
s112, analyzing and comparing each period section according to the feature set to obtain a difference value, and sorting the difference value into a difference value set; the characteristic set is obtained through reference video analysis;
s113, when one difference value in the difference value set is larger than a set value or the sum of the difference values in the difference value set is larger than the set value, alarm information is sent out.
As described in step S100, a first analysis method and a second analysis method are selected, and in the present embodiment, the two analysis methods are the first analysis method in steps S111 to S113 and the second analysis method in steps S120 to S122;
when the first analysis mode is selected to analyze the recorded live real-time video, the monitoring process is divided into a plurality of period segments according to the monitoring period, the monitoring period segments are automatically generated, the monitoring period segments are at least one period, the monitoring can be considered to be stopped before the period is ended, the monitoring start signal is the last characteristic picture in the reference period picture group, when the monitored object is in a moving state all the time, the monitoring start point is usually default to be the start point of a moving period, when the monitoring equipment monitors that the live picture is the same as the last characteristic picture in the period picture group, the monitoring function is started to operate, and similarly, in each monitoring period, the monitoring period of the last characteristic picture of each period starts to be a new period;
as shown in step S112, each period segment is analyzed and compared according to the feature set to obtain a difference value, the difference value is sorted into a difference value set, a feature description in a picture of each specified frame number in each period segment is obtained and compared with the period feature set, the plane coordinate value of each feature point is mainly compared to change, so as to calculate a feature difference value, and the difference value is sorted into a difference value set, and comparing the difference value generally includes two comparisons: comparing the difference value of each appointed frame number picture with a set value during the analysis period, sending alarm information when the difference value of one appointed frame number picture is larger than the set value, after the analysis is finished, superposing and summing the difference values of all appointed frame number pictures in the same period to obtain a difference set, and sending the alarm information when the difference set is larger than the set value, wherein the set value compared with the difference value of a single picture and the set value compared with the difference set value can be different in size; the characteristic set is obtained through reference video analysis;
as the step S113, when there is a difference value greater than the set value in the difference value set or the sum of the difference values in the difference value set is greater than the set value, an alarm message is sent, where the alarm message is typically a warning sound or a short message, and a forced termination signal for the moving object may also be added according to the requirement.
Referring to fig. 2, in the present embodiment, the method for monitoring the motion state of the object, where the feature description is obtained by reference video analysis, includes the steps of:
s101, recording a reference video;
s102, setting the duration of the reference video as a monitoring period, and calculating and analyzing a first frame of picture of the reference video to obtain picture characteristic points;
s103, selecting pictures with specific frame number in the reference video to form a reference period picture group, and calculating the feature points of each picture in the reference period picture group to form a period feature set.
As the step S101, a complete motion cycle video of the monitored target in the normal motion state is recorded as a reference video, the reference video of each second length is default to include twenty-four frames, the number of frames of the frames is adjustable, the upper limit of the number of frames contained in each second video is constant according to the specification of the monitoring device, and generally does not exceed one hundred frames;
as the step S102, the duration of the reference video is set as a monitoring period, and the first frame of the reference video is calculated and analyzed to obtain a frame feature point; setting the time length of the reference video as a monitoring period, for example, if the time length of the reference video is two and eighteen seconds, then every two and eighteen seconds is a complete analysis and determination period in the monitoring process, in each period, under the condition that no abnormality occurs, the target motion track in the monitoring area and the time length required for generation or continuation of each motion track are very similar or identical, the degree of similarity is not lower than a set value, and performing calculation and analysis on a first frame of picture of the reference video to obtain picture feature points, wherein the picture feature points generally comprise color salient points in the picture, and contour edge random points of the monitored target in the picture are regarded as specially selected target points, and the number of the picture feature points generally comprises at least thirty; combining the gray values of the characteristic points into a binary data string, taking the binary data string as the characteristic description of the characteristic points, wherein the characteristic description also comprises plane coordinate values of the characteristic points, and the same coordinate value can only correspond to one unique characteristic point in the same picture, but a plurality of characteristic points with the same gray value can appear in the same picture;
as mentioned above in step S103, selecting a frame of a specific number of frames in the reference video to form a reference periodic frame group, for example, under the condition that the reference video has a duration of ten seconds and each second includes thirty frames, selecting fifty frames of frames 1, 6, 12, 18 … 288, 294, 300 as the reference periodic frame group, where the selected frame number can be changed according to the accuracy requirement of the monitoring target and the length of the monitoring period, generally, the interval of each frame is not less than three frames, calculating the plane coordinate value of the feature description corresponding to each frame in the reference periodic frame group, when the feature description points with the same gray value are more than the feature description points with the same gray value of the reference frame obtained in steps S101-S102, calling several frames (at most 3 frames) before the frame in the reference periodic frame group to perform path analysis on the feature points with the same gray value in the region where the newly added feature points appear, obtaining a characteristic point moving curve formed by a plurality of pictures, selecting a characteristic point closest to the curve as an effective characteristic point, and summarizing the characteristic description of the characteristic point of each picture in a reference periodic picture group to form a periodic characteristic set;
referring to fig. 3, in another specific embodiment, the analysis method may further monitor a still object, and when monitoring the still object, the step S103 is stopped, the used part of the step S102 becomes step S141, and the steps S111, S112, and S113 are replaced with the following steps S140, S142, and S143, and only the first frame picture of the reference video needs to be used as a reference picture, and the feature description in the feature point of the reference picture is used as an analysis reference of the monitoring process, which specifically includes the following steps: s101, recording a reference video; s140, calculating and analyzing a first frame of picture of the reference video to obtain picture characteristic points; s141, combining the gray values of the feature points into a binary string, and using the binary string as feature description of the feature points; s142, comparing the video picture with the feature description to calculate a difference value; 143. and when the difference value is larger than the set value, sending alarm information.
Referring to fig. 4, in this embodiment, the method for monitoring the motion state of the object further includes the following steps:
s120, shooting a field reference picture;
s121, when a second analysis mode is selected to analyze the recorded live real-time video, analyzing and processing a reference picture, wherein the processed picture is used as a background picture;
s122, subtracting the recorded live real-time video picture from the background picture, performing difference analysis on the picture, extracting a difference picture, and analyzing a difference value;
and S123, when the difference value is larger than the set value, sending alarm information.
As the above step S120, a picture of the monitoring target in the normal still state is taken as a reference picture;
in step S121, the reference picture is grayed and filtered to remove the influence monitoring factors such as mottle; carrying out edge detection processing on the grayed picture to obtain an edge picture only with a monitoring target and a background, wherein the processed picture is used as a background picture;
as the step S122, the real-time video picture and the background picture are subtracted to obtain a difference point between the real-time video picture and the background picture, the picture is binarized to obtain a difference picture, the difference picture is extracted, and a difference value is analyzed according to a difference degree of the difference picture;
as the above step S123, when the difference value is greater than the set value, an alarm message is sent, where the alarm message is typically a warning sound, a short message, or a forced termination signal for the moving object may be added according to the requirement.
Referring to fig. 5, in this embodiment, after issuing an alarm message when there is a difference value in the difference value set that is greater than the set value or the sum of difference values in the difference value set is greater than the set value, or issuing an alarm message when the difference value is greater than the set value, the method for monitoring the motion state of the object further includes:
s130, recording an alarm real-time scene video;
s131, sending the alarm real-time scene video to a preset mailbox.
As the step S130, a real-time scene video of the alarm is recorded, when the alarm condition occurs, the device records the condition within a period of time since the alarm condition occurs through the video while the alarm information is sent, and the length of the video is generally at most 10 minutes;
as mentioned in step S131, after the real-time scene video of the alarm is recorded, the real-time scene video of the alarm is sent to a preset mailbox through a wireless or wired network, where the number of the mailboxes is generally one, and a plurality of receiving mailboxes can be set.
Referring to fig. 6, in this embodiment, after the step S100 of selecting the first analysis manner and the second analysis manner, the method for monitoring the motion state of the object further includes:
s10, setting a set value;
s20, setting the size and position parameters of the monitoring area in the monitoring picture;
s30, setting the recording duration, the recording image quality and the video format of the recorded alarm real-time scene video;
and S40, setting a sending mailbox of the alarm real-time scene video.
Setting a set value according to the step S10, wherein the difference value is a percentage value between 1 and 100, and the smaller the alarm difference value is, the more accurate the analysis precision is;
as the step S20, setting the size and position parameters of the monitoring area in the monitoring picture, and setting the monitoring area in the video picture, where the monitoring area may be composed of multiple areas or a single area, and the area size and plane coordinate parameters of each area may be set according to the situation;
as mentioned in the step S30, the length of the recording time for recording the warning real-time scene video, the recording quality and the video format are set, the length of the recording time for recording the video is usually at most 10 minutes, the recording quality generally includes three alternative recording qualities of standard definition, high definition and super definition, which are generally set as standard definition quality by default, and the video format includes MP4, MOV, RM and WMV, which are generally set as MP4 format by default.
As the step S40, a sending mailbox for the real-time scene video of the alarm is set, and the mailbox is generally one mailbox, and a plurality of receiving mailboxes can be set.
Referring to fig. 7, an embodiment of the present invention provides an apparatus for monitoring a motion state of an object, including: the input module 100 is used for selecting a first analysis mode and a second analysis mode; the processing module 120 is configured to, when the first analysis mode is selected to analyze the recorded live real-time video, divide the monitoring process into a plurality of period segments according to the monitoring period; the processing module 120 is configured to analyze and compare each period segment according to the feature set to obtain a difference value, and sort the difference value into a difference value set; the characteristic set is obtained through reference video analysis; the alarm module 130 is configured to send an alarm message when one difference value in the difference value set is greater than the set value or the sum of the difference values in the difference value set is greater than the set value.
The input module 100 is configured to select an analysis manner, in this embodiment, the analysis manner includes two analysis manners, which are a first analysis manner in steps S100 to S113 and a second analysis manner in steps S120 to S123;
the processing module 120 is configured to divide the monitoring process into a plurality of period segments according to the monitoring period, where the monitoring period segment is automatically generated, where the monitoring period segment is at least one period, and the monitoring can be considered to be stopped before the period is ended, and the monitoring start signal is the last feature picture in the reference period picture group, and when the monitored object is always in a motion state and the monitoring start point is usually defaulted to be the start point of a motion period, when the monitoring device monitors that the field picture is the same as the last feature picture in the period picture group, the monitoring function starts to operate, and similarly, in each monitoring period, during the monitoring period of the last feature picture in each period, a new period of monitoring is started;
the processing module 120 is configured to analyze each period segment according to the period feature set, obtain feature descriptions in the pictures of each specified frame number in each period segment, compare the feature descriptions with the period feature set, mainly compare changes of plane coordinate values of each feature point, thereby calculate a feature difference value, and sort the feature difference value into a difference value set, where comparing the difference values generally includes two comparisons: comparing the difference value of each appointed frame number picture with a set value during the analysis period, sending alarm information when the difference value of one appointed frame number picture is larger than the set value, after the analysis is finished, superposing and summing the difference values of all appointed frame number pictures in the same period to obtain a difference set, and sending the alarm information when the difference set is larger than the set value, wherein the set value compared with the difference value of a single picture and the set value compared with the difference set value can be different in size;
the alarm module 130 is configured to send an alarm message when one difference value in the difference value set is greater than the set value or the sum of the difference values in the difference value set is greater than the set value, where the alarm message is generally a warning sound or a short message, and a forced termination signal for the moving object may be added according to a requirement.
In this embodiment, the apparatus for monitoring the motion state of an object further includes: a monitoring module 110, configured to record a reference video; the processing module 120 is configured to set the duration of the reference video as a monitoring period, and perform calculation analysis on a first frame of picture of the reference video to obtain a picture feature point; the processing module 120 is configured to select a frame of a specific frame number in the reference video to form a reference period frame group, and calculate feature points of each frame in the reference period frame group to form a period feature set.
The monitoring module 110 is configured to record a complete motion cycle video of the monitored target in a normal motion state, as a reference video, where the reference video of each second length default to include twenty-four frames, the frame number of the frame is adjustable, and the upper limit of the frame number contained in each second video is constant according to the specification of the monitoring module 110, and generally does not exceed one hundred frames;
the processing module 120 is configured to set the duration of the reference video as a monitoring period, for example, if the duration of the reference video is two minutes and eighteenth seconds, then every two minutes and eighteenth seconds is a complete analysis and determination period in the monitoring process, and in each period, under the condition that no abnormality occurs, the target motion trajectory in the monitoring area and the duration required for generating or continuing each motion trajectory are very similar or identical, and the degree of similarity is not lower than a set value, and perform calculation and analysis on a first frame of the picture of the reference video to obtain picture feature points, where the picture feature points generally include color salient points in the picture, and contour edge random points of the monitored target in the picture, and the number of the picture feature points is considered to be a specifically selected target point, and the number of the picture feature points generally includes at least thirty, combine gray values of the feature points into a binary data string, and use the binary data string as a feature description of the feature point, the feature description also comprises a plane coordinate value of the feature point, and the same coordinate value can only correspond to one unique feature point in the same picture, but a plurality of feature points with the same gray value can appear in the same picture;
the processing module 120 is configured to select a frame of a specific number of frames in the reference video to form a reference periodic frame group, for example, fifty frames of 1, 6, 12, 18 … 288, 294, 300 frames are selected as the reference periodic frame group under the condition that the reference video has a duration of ten seconds and each second includes thirty frames, the selected number of frames can be changed according to the accuracy requirement of the monitoring target and the length of the monitoring period, generally, the interval of each frame is not less than three frames, a plane coordinate value of the feature description corresponding to each frame in the reference periodic frame group is calculated, when feature description points having the same gray value are more than the feature description points having the same gray value of the reference frame obtained in step S102, a number of frames (at most 3 frames) before the frame in the reference periodic frame group are called to perform path analysis on the feature points having the same gray value in the region where the newly added feature points occur, and obtaining a characteristic point moving curve consisting of a plurality of pictures, selecting the characteristic point closest to the curve as an effective characteristic point, and summarizing the characteristic description of the characteristic point of each picture in the reference periodic picture group to form a periodic characteristic set.
In another specific embodiment, the analysis method may further monitor a static target, and when monitoring the static target, the monitoring module 110 only needs to use the first frame picture of the reference video as a reference picture, and the processing module 120 uses the feature description in the feature point of the reference picture as an analysis reference in the monitoring process, which is specifically as follows: a monitoring module 110, configured to record a reference video; the monitoring module 110 is used for recording a live real-time video; the processing module 120 is configured to perform calculation analysis on a first frame of picture of the reference video to obtain picture feature points; the processing module 120 is configured to combine the grayscale values of the feature points into a binary string, and use the binary string as a feature description of the feature point; the processing module 120 is configured to compare the video picture with the feature description and calculate a difference value; and an alarm module 130, configured to send an alarm message when the difference value is greater than the set value.
In this embodiment, the apparatus for monitoring the motion state of an object further includes: the monitoring module 110 is used for shooting a site reference picture; the processing module 120 is configured to, when a second analysis mode is selected to analyze the recorded live real-time video, analyze the reference picture, and use the processed picture as a background picture; the processing module 120 is configured to perform difference between the recorded live real-time video picture and the background picture, perform difference analysis on the picture, extract a difference picture, and analyze a difference value; and an alarm module 130, configured to send an alarm message when the difference value is greater than the set value.
The monitoring module 110 is configured to shoot a picture of the monitored target in a normal static state as a reference picture;
the processing module 120 is configured to perform graying and filtering processing on the reference picture, remove monitoring factors such as mottle and the like, perform edge detection processing on the grayed picture, and obtain an edge picture with only a monitoring target and a background, where the processed picture is used as a background picture;
the processing module 120 is configured to perform a difference between the real-time video picture and the background picture to obtain a difference point between the real-time video picture and the background picture, perform binarization on the picture to obtain a difference picture, extract the difference picture, and analyze a difference value according to a difference degree of the difference picture;
the alarm module 130 is configured to send an alarm message when the difference value is greater than the set value, where the alarm message is generally a warning sound or a short message, and may also add a forced termination signal to the moving object according to a requirement.
In this embodiment, the apparatus for monitoring the motion state of an object further includes: the monitoring module 110 is used for alarming a real-time scene video; and the sending module 140 is configured to send the alarm real-time scene video to a preset mailbox.
The monitoring module 110 is further configured to record an alarm real-time scene video, when an alarm condition occurs, after the alarm information is sent, the monitoring module 110 records a condition within a period of time since the alarm condition occurs through a video, and the length of the video is generally at most 10 minutes;
the sending module 140 is configured to send the real-time scene video of the alarm to a preset mailbox through a wireless or wired network after the real-time scene video of the alarm is recorded, where the number of mailboxes is generally one, and multiple receiving mailboxes may be set.
In this embodiment, the apparatus for monitoring the motion state of an object further includes: an input module 100 for setting a setting value; an input module 100, configured to set a size and a position parameter of a monitoring area in a monitoring screen; the input module 100 is configured to set a recording duration, a recording quality, and a video format for recording the alarm real-time scene video. And the input module 100 is used for setting a sending mailbox of the alarm real-time scene video.
The input module 100 is configured to; setting a set value, wherein the difference value is a percentage value of 1-100, and the smaller the alarm difference value is, the more accurate the analysis precision is;
the input module 100 is configured to set a size and a position parameter of a monitoring region in a monitoring picture, and set a monitoring region in a video picture, where the monitoring region may be composed of a plurality of regions or a single region, and an area size and a planar coordinate parameter of each region may be set according to a situation;
the input module 100 is configured to set a recording time length, a recording quality and a video format for recording the warning real-time scene video, where the recording time length of the video is generally at most 10 minutes, the recording quality generally includes three selectable recording qualities of standard definition, high definition and super definition, and is generally set as the standard definition quality by default, and the video format includes MP4, MOV, RM and WMV, and is generally set as the MP4 format by default.
The input module 100 is further configured to set a sending mailbox for the real-time scene video of the alarm, where the number of the mailboxes is generally one, and multiple receiving mailboxes can be set.
The above description is only a preferred embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.
Claims (6)
1. A method for monitoring the motion state of an object is applied to mobile monitoring and is characterized by comprising the following steps:
dividing the monitoring process into a plurality of period sections according to the monitoring period;
analyzing and comparing each period section according to the feature set to obtain a difference value, and sorting the difference value into a difference value set; the characteristic set is obtained through reference video analysis, specifically, a reference video is recorded, wherein the reference video is a video of a complete action cycle of a recorded monitoring target in a normal motion state; setting the duration of the reference video as a monitoring period, and calculating and analyzing a first frame of picture of the reference video to obtain picture characteristic points; selecting pictures with specific frame numbers in the reference video to form a reference period picture group, and calculating characteristic points of each picture in the reference period picture group to form a period characteristic set;
and when one difference value in the difference value set is greater than the set value or the sum of the difference values in the difference value set is greater than the set value, sending alarm information.
2. The method according to claim 1, wherein when there is a difference value in the difference value set greater than a set value or the sum of the difference values in the difference value set is greater than a set value, and an alarm message is issued, the method further comprises:
recording an alarm real-time scene video;
and sending the alarm real-time scene video to a preset mailbox.
3. The method of monitoring the state of motion of an object of claim 1, further comprising, prior to the step of dividing the monitoring process into a plurality of cycle segments according to a monitoring cycle:
setting a set value;
setting the size and position parameters of a monitoring area in a monitoring picture;
setting recording duration, recording image quality and video format of the recorded alarm real-time scene video;
and setting a sending mailbox for the alarm real-time scene video.
4. An apparatus for monitoring the motion state of an object, for use in mobile monitoring, comprising:
the processing module is used for dividing the monitoring process into a plurality of period sections according to the monitoring period;
the processing module is used for analyzing and comparing each period section according to the feature set to obtain a difference value and sorting the difference value into a difference value set; the characteristic set is obtained through reference video analysis, specifically, a reference video is recorded, wherein the reference video is a video of a complete action cycle of a recorded monitoring target in a normal motion state; setting the duration of the reference video as a monitoring period, and calculating and analyzing a first frame of picture of the reference video to obtain picture characteristic points; selecting pictures with specific frame numbers in the reference video to form a reference period picture group, and calculating characteristic points of each picture in the reference period picture group to form a period characteristic set;
and the alarm module is used for sending alarm information when one difference value in the difference value set is greater than a set value or the sum of the difference values in the difference value set is greater than the set value.
5. The apparatus for monitoring the motion state of an object according to claim 4, further comprising:
the monitoring module is used for recording an alarm real-time scene video;
and the sending module is used for sending the alarm real-time scene video to a preset mailbox.
6. The apparatus for monitoring the motion state of an object according to claim 4, further comprising:
the input module is used for setting a set value;
the input module is used for setting the size and the position parameters of a monitoring area in a monitoring picture;
the input module is used for setting the recording duration, the recording image quality and the video format of the recorded alarm real-time scene video;
and the input module is used for setting a sending mailbox of the 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 CN106781167A (en) | 2017-05-31 |
| CN106781167B true 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) |
Families Citing this family (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 (2)
| 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 |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE10050083A1 (en) * | 2000-10-10 | 2002-04-18 | Sick Ag | Device and method for detecting 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 |
| CN103286782B (en) * | 2013-06-07 | 2016-06-29 | 上海发那科机器人有限公司 | The flexible tracing-positioning system of a kind of robot and method for tracking and positioning |
| US20150092040A1 (en) * | 2013-10-01 | 2015-04-02 | Broadcom Corporation | Gesture-Based Industrial Monitoring |
| CN105282521B (en) * | 2015-11-21 | 2018-09-14 | 浙江宇视科技有限公司 | Method for testing motion and device in a kind of cruise of web camera |
| CN106131375B (en) * | 2016-06-29 | 2020-02-21 | 联想(北京)有限公司 | Device and motion state detection method |
-
2016
- 2016-12-29 CN CN201611247277.XA patent/CN106781167B/en active Active
Patent Citations (2)
| 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 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106781167A (en) | 2017-05-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106781167B (en) | Method and device for monitoring motion state of object | |
| US20200193619A1 (en) | Method and device for tracking an object | |
| CN110689054B (en) | Worker violation monitoring method | |
| CN108319926A (en) | A kind of the safety cap wearing detecting system and detection method of building-site | |
| US7778445B2 (en) | Method and system for the detection of removed objects in video images | |
| US7751647B2 (en) | System and method for detecting an invalid camera in video surveillance | |
| KR102195706B1 (en) | Method and Apparatus for Detecting Intruder | |
| CN109766779B (en) | Loitering person identification method and related product | |
| CN102348128A (en) | Surveillance camera system having camera malfunction detection function | |
| WO2021139049A1 (en) | Detection method, detection apparatus, monitoring device, and computer readable storage medium | |
| CN111161206A (en) | Image capturing method, monitoring camera and monitoring system | |
| CN111010547A (en) | Target object tracking method and device, storage medium and electronic device | |
| CN112329691A (en) | Monitoring video analysis method and device, electronic equipment and storage medium | |
| KR20200052418A (en) | Automated Violence Detecting System based on Deep Learning | |
| CN109815839B (en) | Loitering person identification method under micro-service architecture and related product | |
| CN111353436A (en) | Super store operation analysis method and device based on image deep learning algorithm | |
| CN112381054A (en) | Method for detecting working state of camera and related equipment and system | |
| CN111860187A (en) | High-precision worn mask identification method and system | |
| KR101581162B1 (en) | Automatic detection method, apparatus and system of flame, smoke and object movement based on real time images | |
| CN110505438B (en) | Queuing data acquisition method and camera | |
| CN113947744A (en) | Fire image detection method, system, equipment and storage medium based on video | |
| CN114584758A (en) | City-level monitoring video quality assessment method and system | |
| JP2612662B2 (en) | Intruder identification system | |
| KR101848367B1 (en) | metadata-based video surveillance method using suspective video classification based on motion vector and DCT coefficients | |
| CN115394025A (en) | Monitoring method, monitoring device, electronic equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |