CN109492602A - A kind of process clocking method and system based on human body limb language - Google Patents
A kind of process clocking method and system based on human body limb language Download PDFInfo
- Publication number
- CN109492602A CN109492602A CN201811393385.7A CN201811393385A CN109492602A CN 109492602 A CN109492602 A CN 109492602A CN 201811393385 A CN201811393385 A CN 201811393385A CN 109492602 A CN109492602 A CN 109492602A
- Authority
- CN
- China
- Prior art keywords
- worker
- peak
- production process
- video data
- peak value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/20—Movements or behaviour, e.g. gesture recognition
-
- 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
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C1/00—Registering, indicating or recording the time of events or elapsed time, e.g. time-recorders for work people
- G07C1/10—Registering, indicating or recording the time of events or elapsed time, e.g. time-recorders for work people together with the recording, indicating or registering of other data, e.g. of signs of identity
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
- G06F2218/10—Feature extraction by analysing the shape of a waveform, e.g. extracting parameters relating to peaks
Abstract
The invention discloses a kind of process clocking methods and system based on human body limb language.Wherein, the described method includes: obtaining the video data in worker's production process, in turn from the body language information identified in the video data in worker's production process, and then according to the video data and the body language information, establish the two-dimensional coordinate figure of video frame and the body language information based on the video data, and then image denoising is carried out to the two-dimensional coordinate figure, and then the coordinate data of expression body language information in the two-dimensional coordinate figure after image denoising is labeled, mark out peak value, and then the peak value marked out according to this, taking two peak value sections is a process period, activity time in worker's production process is counted, obtain the process timing in worker's production process.It participates in carrying out timing to the process in worker's production process automatically without artificial by the above-mentioned means, can be realized, so that process timing is more objective and accurate.
Description
Technical field
The present invention relates to process clocking technique field more particularly to a kind of process clocking methods based on human body limb language
And system.
Background technique
With the development of computer science and technology, sewed product industry can use modern computer science and technology, will sew
The application message of the general sewing time of industry makes enterprise side possess an intelligence, friendly, convenient and fast working environment.Pass through
Computer system software can establish powerful static data library, and data bank has that both pictures and texts are excellent, retrieve convenient feature, can be with
So that common management personnel is easily accomplished other operations such as the selection of data, duplication, is finally reached the method point for fast implementing style
The purpose of analysis and time analysis.Computer system software can will analyze the result come by figure, and the modes such as table are depicted
Come, and technological balance can be carried out by computer, personnel assignment, machine assignment, production ranking, the data operations such as productivity analysis,
Huge support is provided for sewing trade management.
In current sewed product industry, used computer management system is also based only on traditional computer software
Technology.Use scope and reliability are all very limited, and especially to the every procedure timing of sewing, current main method uses worker
The method swiped the card swipes the card respectively at the beginning and end of process and then carries out the verifying of identity and the counting of timing.These operations
A large amount of artificial Attended Operations are needed, be easy to cause the waste of personnel, while to easily lead to timing result inadequate because manually participating in factor
It is objective and not accurate enough.
Summary of the invention
In view of this, it is an object of the invention to propose a kind of process clocking method based on human body limb language and be
System can be realized and participate in carrying out timing to the process in worker's production process automatically without artificial, avoids process timing
In artificial participation factor so that process timing is more objective and accurate.
According to an aspect of the present invention, a kind of process clocking method based on human body limb language is provided, comprising:
Obtain the video data in worker's production process;
According to the video data of the acquisition, from the limbs identified in the video data in worker's production process
Language message;
According to the video data of the acquisition and the body language information identified, establishes and be based on the video data
Video frame and the body language information identified two-dimensional coordinate figure;
Image denoising is carried out to the two-dimensional coordinate figure of the foundation;
To indicating that the coordinate data of body language information is labeled in the two-dimensional coordinate figure after image denoising, mark
Outpour peak value;Wherein, the peak value includes peak-peak and minimum peak;
According to the peak value marked out, taking two peak value sections is a process period, to worker's production process
In activity time counted, obtain the process timing in worker's production process.
Wherein, the video data obtained in worker's production process, comprising:
Using camera shooting mode, the video data in worker's production process is obtained.
Wherein, the video data according to the acquisition identifies that the worker produced from the video data
Body language information in journey, comprising:
According to the video data of the acquisition, using body language identification method, identified from the video data described
Body language information in worker's production process.
Wherein, the peak value marked out according to, taking two peak value sections is a process period, to the worker
Activity time in production process is counted, and the process timing in worker's production process is obtained, comprising:
Include peak-peak minimum peak according to the peak value marked out, takes two neighboring peak value section as a work
The sequence period carries out coordinate conversion to the peak value section marked out, the peak value section is converted to same timing criteria
Time data obtain the corresponding time data in the adjacent peak-peak section or adjacent minimum peak section, according to institute
The corresponding time data in peak-peak section and minimum peak section stated, to the work in worker's production process
The sequence time is counted, and the process timing in worker's production process is obtained.
Wherein, in the peak value marked out according to, taking two peak value sections is a process period, to the work
Activity time during life produces is counted, after obtaining the process timing in worker's production process, further includes:
According to the process timing in obtained worker's production process, the obtained worker was produced
The same period activity time in process timing in journey is counted, each same period counted described in comparison
The activity time of property process, compares activity time abnormal conditions, according to the activity time abnormal conditions, automatically to worker into
The prompt of row activity time exception.
According to an aspect of the present invention, a kind of process timekeeping system based on human body limb language is provided, comprising:
Video acquisition unit, body language recognition unit, coordinate establish unit, image denoising unit, take peak cell and
Time statistic unit;
The video acquisition unit, for obtaining the video data in worker's production process;
The body language recognition unit is identified from the video data for the video data according to the acquisition
Body language information in worker's production process out;
The coordinate establishes unit, for being believed according to the video data and the body language identified of the acquisition
Breath establishes the two-dimensional coordinate figure of video frame and the body language information identified based on the video data;
Described image denoises unit, for carrying out image denoising to the two-dimensional coordinate figure of the foundation;
It is described to take peak cell, for expression body language information in the two-dimensional coordinate figure after image denoising
Coordinate data is labeled, and marks out peak value;Wherein, the peak value includes peak-peak and minimum peak;
The time statistic unit, the peak value for marking out according to, taking two peak value sections is a process week
Phase counts the activity time in worker's production process, obtains the process timing in worker's production process.
Wherein, the video acquisition unit, is specifically used for:
Using camera shooting mode, the video data in worker's production process is obtained.
Wherein, the body language recognition unit, is specifically used for:
According to the video data of the acquisition, using body language identification method, identified from the video data described
Body language information in worker's production process.
Wherein, the time statistic unit, is specifically used for:
Include peak-peak minimum peak according to the peak value marked out, takes two neighboring peak value section as a work
The sequence period carries out coordinate conversion to the peak value section marked out, the peak value section is converted to same timing criteria
Time data obtain the corresponding time data in the adjacent peak-peak section or adjacent minimum peak section, according to institute
The corresponding time data in peak-peak section and minimum peak section stated, to the work in worker's production process
The sequence time is counted, and the process timing in worker's production process is obtained.
Wherein, the process timekeeping system based on human body limb language, further includes:
Process chronometric analysis unit, the process timing in worker's production process for being obtained according to, to institute
The same period activity time in the process timing in worker's production process stated counts, the system
The activity time for each same period process counted out, compares activity time abnormal conditions, when according to the process
Between abnormal conditions, automatically to worker carry out activity time exception prompt.
It can be found that above scheme, video data in available worker's production process, and then according to the acquisition
Video data, from the body language information identified in the video data in worker's production process, and then according to the acquisition
Video data and the body language information identified, establish video frame and the limbs identified based on the video data
The two-dimensional coordinate figure of language message, and then image denoising is carried out to the two-dimensional coordinate figure of the foundation, and then to this through image denoising
Indicate that the coordinate data of body language information is labeled, and marks out peak value, and then according to the mark in two-dimensional coordinate figure afterwards
Peak value out, taking two peak value sections is a process period, counts, obtains to the activity time in worker's production process
To the process timing in worker's production process, can be realized can be automatically to the work in worker's production process without artificial participation
Sequence carries out timing, the artificial participation factor in process timing is avoided, so that process timing is more objective and accurate.
Further, above scheme, the process timing in worker's production process that can be obtained according to this, obtains this
Worker's production process in process timing in same period activity time count, this for comparing that this counts is each
The activity time of a same period process, compares activity time abnormal conditions, according to the activity time abnormal conditions, automatically
To worker carry out activity time exception prompt, can be realized in process of production can in time the worker to working efficiency exception into
Row prompt, improves the working efficiency in production process.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Fig. 1 is the flow diagram of one embodiment of process clocking method the present invention is based on human body limb language;
Fig. 2 is established in one embodiment of process clocking method the present invention is based on human body limb language based on worker's life
The video frame of video data during production and two based on the body language information in worker's production process identified
Tie up an example schematic of coordinate diagram;
Fig. 3 is one after image denoising in one embodiment of process clocking method the present invention is based on human body limb language
The video frame of video data in worker's production process is believed with what is identified based on the body language in worker's production process
One example schematic of the dimensional waveform figure of the two-dimensional coordinate of breath;
Fig. 4 is the flow diagram of another embodiment of process clocking method the present invention is based on human body limb language;
Fig. 5 is the structural schematic diagram of one embodiment of process timekeeping system the present invention is based on human body limb language;
Fig. 6 is the structural schematic diagram of another embodiment of process timekeeping system the present invention is based on human body limb language;
Fig. 7 is the structural schematic diagram of the another embodiment of process timekeeping system the present invention is based on human body limb language.
Specific embodiment
With reference to the accompanying drawings and examples, the present invention is described in further detail.It is emphasized that following implement
Example is merely to illustrate the present invention, but is not defined to the scope of the present invention.Likewise, following embodiment is only portion of the invention
Point embodiment and not all embodiments, institute obtained by those of ordinary skill in the art without making creative efforts
There are other embodiments, shall fall within the protection scope of the present invention.
The present invention provides a kind of process clocking method based on human body limb language, can be realized without artificial that participate in can be with
Automatically timing is carried out to the process in worker's production process, the artificial participation factor in process timing is avoided, so that process meter
Shi Gengjia is objective and accurate.
Referring to Figure 1, Fig. 1 is the process signal of one embodiment of process clocking method the present invention is based on human body limb language
Figure.It is noted that if having substantially the same as a result, method of the invention is not limited with process sequence shown in FIG. 1.Such as
Shown in Fig. 1, this method comprises the following steps:
S101: the video data in worker's production process is obtained.
Wherein, the video data in one worker's production process of the acquisition may include:
Using camera shooting mode, the video data in worker's production process is obtained.
S102: according to the video data of the acquisition, from the limbs identified in the video data in worker's production process
Language message.
Wherein, video data according to the acquisition, from the limb identified in the video data in worker's production process
Body language message may include:
It is identified from the video data according to the video data of the acquisition using OpenPose (body language identification) mode
Body language information in worker's production process out.
S103: it according to the video data of the acquisition and the body language information identified, establishes and is based on the video data
The two-dimensional coordinate figure of body language information that is identified with this of video frame.
Fig. 2 is referred to, Fig. 2 is to establish to be based in one embodiment of process clocking method the present invention is based on human body limb language
The video frame of video data in one worker's production process with identify based on the limbs language in worker's production process
Say an example schematic of the two-dimensional coordinate figure of information.As shown in Fig. 2, be by taking SEWING SEQUENCE as an example, in real operation, big portion
When dividing SEWING SEQUENCE to pass through the sequence of operations such as worker's feeding, sewing, blowing, and being easy discovery normal procedure, worker is always
With left hand come feeding, sewing is then carried out under sewing machine, finished product is put into the area Wan Pin again after finishing by sewing.
As shown in Fig. 2, when worker takes material with left hand, in the picture of video frame, the left hand wrist X-direction of worker this moment
It is farthest apart from origin, so coordinate is maximum value in worker's left hand wrist X-direction in the video frame, draw out as shown in Figure 2
Frequency frame X-direction value and video frame frame number X-Y scheme.As shown in Fig. 2, X-axis indicates the frame number of frequency frame, Y-axis indicates worker's left hand
For wrist in the coordinate of video frame X-direction, then it is its peak point that the maximum value in X-direction is corresponding, i.e., peak point indicates worker
At the time of feeding.The subsequent feature according in production process and the X-Y scheme, automatically to the work in worker's production process
Sequence carries out timing.
S104: image denoising is carried out to the two-dimensional coordinate figure of the foundation.
S105: to indicating that the coordinate data of body language information is marked in the two-dimensional coordinate figure after image denoising
Note, marks out peak value;Wherein, which includes peak-peak and minimum peak.
Fig. 3 is referred to, Fig. 3 is to go in one embodiment of process clocking method the present invention is based on human body limb language through image
The video frame of the video data in worker's production process after making an uproar with identify based in worker's production process
One example schematic of the dimensional waveform figure of the two-dimensional coordinate of body language information.As shown in figure 3, be by taking SEWING SEQUENCE as an example,
Basipetal first segment curve indicates the body language such as human body wrist artis X-coordinate of a worker in process of production
With the dimensional waveform figure of the frame number of the video data of the worker in process of production, basipetal second segment curve indicates the work
The body language of people in process of production such as human body wrist artis Y-coordinate and the video data of the worker in process of production
Frame number dimensional waveform figure, dot marks out peak value, at the time of can indicating worker's left hand feeding.This uses dot form
There are two scatterplot line graphs under the peak point of mark, is one two for being constructed by GoogleNet (convolutional neural networks)
Classify classifier, whether is that sky is classified to work at present platform, when workbench non-empty is then high threshold, can use process threshold at this time
It is worth the scatter plot that value is 600, conversely, being following Low threshold, can use the scatter plot that process threshold value value is 580 at this time.Such as Fig. 3
It is shown, it is clear that it can be seen that the scatterplot that for example above-mentioned threshold value of high threshold is 600 has larger interruption at peak point, to indicate this moment
Workbench is empty for some time.
S106: the peak value marked out according to this, taking two peak value sections is a process period, to worker's production process
In activity time counted, obtain the process timing in worker's production process.
Wherein, the peak value marked out according to this, taking two peak value sections is a process period, is produced to the worker
Activity time in journey is counted, and is obtained the process timing in worker's production process, be may include:
Include peak-peak minimum peak according to the peak value that this is marked out, takes two neighboring peak value section as a process
Period carries out coordinate conversion to the peak value section marked out, which is converted to the time number of same timing criteria
According to obtaining the corresponding time data in the adjacent peak-peak section or adjacent minimum peak section, obtained most according to this
Big peak value section and the corresponding time data in minimum peak section, unite to the activity time in worker's production process
Meter, obtains the process timing in worker's production process.
Wherein, in the peak value marked out according to this, taking two peak value sections is a process period, is produced to the worker
Activity time in the process is counted, and after obtaining the process timing in worker's production process, can also include:
According to the process timing in obtained worker's production process, to the work in obtained worker's production process
Same period activity time in sequence timing is counted, and the process of each same period process counted is compared
Time compares activity time abnormal conditions, and according to the activity time abnormal conditions, it is abnormal to carry out activity time to worker automatically
Prompt.
In the present embodiment, being somebody's turn to do the process clocking method based on human body limb language can be applied in sewed product industry,
Vision camera shooting mode can use to carry out process timing to the process in production process, it can be to avoid artificial in process timing
Factor.
In the present embodiment, being somebody's turn to do the process clocking method based on human body limb language can use computer vision technique,
Especially human body attitude estimation technique is analyzed working time, that is, activity time of the worker in production process, is mentioned automatically
The time cycle of every product of production is taken, provides the data of core for subsequent management and working efficiency analysis etc..
In the present embodiment, should be not necessarily to artificial hand based on the process clocking method of human body limb language takes stopwatch to carry out timing
Or worker oneself timing, video capture can be carried out for example, by the mode of camera shooting, the video content can be automatically analyzed and obtained
Activity time can fill up such as view-based access control model in the market as a result, it is possible to realize the process timing course of active prosthetic factor
Sewed product process timekeeping system blank.
It can be found that above scheme, video data in available worker's production process, and then according to the acquisition
Video data, from the body language information identified in the video data in worker's production process, and then according to the acquisition
Video data and the body language information identified, establish video frame and the limbs identified based on the video data
The two-dimensional coordinate figure of language message, and then image denoising is carried out to the two-dimensional coordinate figure of the foundation, and then to this through image denoising
Indicate that the coordinate data of body language information is labeled, and marks out peak value, and then according to the mark in two-dimensional coordinate figure afterwards
Peak value out, taking two peak value sections is a process period, counts, obtains to the activity time in worker's production process
To the process timing in worker's production process, can be realized can be automatically to the work in worker's production process without artificial participation
Sequence carries out timing, the artificial participation factor in process timing is avoided, so that process timing is more objective and accurate.
Fig. 4 is referred to, Fig. 4 is that the process of another embodiment of process clocking method the present invention is based on human body limb language is shown
It is intended to.In the present embodiment, method includes the following steps:
S401: the video data in worker's production process is obtained.
Can be as above described in S101, therefore not to repeat here.
S402: according to the video data of the acquisition, from the limbs identified in the video data in worker's production process
Language message.
Can be as above described in S102, therefore not to repeat here.
S403: it according to the video data of the acquisition and the body language information identified, establishes and is based on the video data
The two-dimensional coordinate figure of body language information that is identified with this of video frame.
Can be as above described in S103, therefore not to repeat here.
S404: image denoising is carried out to the two-dimensional coordinate figure of the foundation.
S405: to indicating that the coordinate data of body language information is marked in the two-dimensional coordinate figure after image denoising
Note, marks out peak value;Wherein, which includes peak-peak and minimum peak.
Can be as above described in S105, therefore not to repeat here.
S406: the peak value marked out according to this, taking two peak value sections is a process period, to worker's production process
In activity time counted, obtain the process timing in worker's production process.
Can be as above described in S106, therefore not to repeat here.
S407: according to the process timing in obtained worker's production process, to obtained worker's production process
In process timing in same period activity time counted, compare each same period process counted
Activity time, compare activity time abnormal conditions, according to the activity time abnormal conditions, automatically to worker carry out process when
Between abnormal prompt.
It can be found that in the present embodiment, the process timing in worker's production process that can be obtained according to this, to this
The obtained same period activity time in the process timing in worker's production process counts, and compares what this was counted
The activity time of each same period process, compares activity time abnormal conditions, according to the activity time abnormal conditions,
Automatically the prompt that activity time exception is carried out to worker, can be realized work that in process of production can in time to working efficiency exception
People prompts, and improves the working efficiency in production process.
The present invention also provides a kind of process timekeeping system based on human body limb language, can be realized can without artificial participation
To carry out timing to the process in worker's production process automatically, the artificial participation factor in process timing is avoided, so that process
Timing is more objective and accurate.
Fig. 5 is referred to, Fig. 5 is the structural representation of one embodiment of process timekeeping system the present invention is based on human body limb language
Figure.It should include video acquisition unit 51, body language knowledge based on the process timekeeping system 50 of human body limb language in the present embodiment
Other unit 52, coordinate establish unit 53, image denoising unit 54, take peak cell 55 and time statistic unit 56.
The video acquisition unit 51, for obtaining the video data in worker's production process.
The body language recognition unit 52 identifies this from the video data for the video data according to the acquisition
Body language information in worker's production process.
The coordinate establishes unit 53, for the video data and the body language information identified according to the acquisition, builds
The two-dimensional coordinate figure of body language information that the video frame of the video data that is based on is identified with this.
The image denoising unit 54, for carrying out image denoising to the two-dimensional coordinate figure of the foundation.
This takes peak cell 55, for the seat for indicating body language information in the two-dimensional coordinate figure after image denoising
Mark data are labeled, and mark out peak value;Wherein, which includes peak-peak and minimum peak.
The time statistic unit 56, the peak value for being marked out according to this, taking two peak value sections is a process period,
Activity time in worker's production process is counted, the process timing in worker's production process is obtained.
Optionally, the video acquisition unit 51, can be specifically used for:
Using camera shooting mode, the video data in worker's production process is obtained.
Optionally, the body language recognition unit 52, can be specifically used for:
According to the video data of the acquisition, OpenPose mode is identified using body language, is identified from the video data
Body language information in worker's production process.
Optionally, the time statistic unit 56, can be specifically used for:
Include peak-peak minimum peak according to the peak value that this is marked out, takes two neighboring peak value section as a process
Period carries out coordinate conversion to the peak value section marked out, which is converted to the time number of same timing criteria
According to obtaining the corresponding time data in the adjacent peak-peak section or adjacent minimum peak section, obtained most according to this
Big peak value section and the corresponding time data in minimum peak section, unite to the activity time in worker's production process
Meter, obtains the process timing in worker's production process.
In the present embodiment, which can be any camera shooting mode that can be used and obtains worker's production
Terminal of video data in the process, such as video camera, camera, mobile phone etc., the present invention is not limited.
Fig. 6 is referred to, Fig. 6 is that the structure of another embodiment of process timekeeping system the present invention is based on human body limb language is shown
It is intended to.It is different from an embodiment, the process timekeeping system 60 based on human body limb language described in the present embodiment further include: process
Chronometric analysis unit 61.
The process chronometric analysis unit 61, for according to the process timing in obtained worker's production process, to this
The obtained same period activity time in the process timing in worker's production process counts, and compares what this was counted
The activity time of each same period process, compares activity time abnormal conditions, according to the activity time abnormal conditions,
Automatically the prompt of activity time exception is carried out to worker.
Each unit module of the process timekeeping system 50/60 based on human body limb language can execute the above method respectively
Step is corresponded in embodiment, therefore each unit module is not repeated herein, and the explanation of the above corresponding step is referred to.
Fig. 7 is referred to, Fig. 7 is that the structure of the another embodiment of process timekeeping system the present invention is based on human body limb language is shown
It is intended to.Each unit module of the process timekeeping system based on human body limb language can execute above method embodiment respectively
Middle corresponding step.Related content refers to the detailed description in the above method, no longer superfluous herein to chat.
In the present embodiment, being somebody's turn to do the process timekeeping system based on human body limb language includes: processor 71 and 71 coupling of processor
Memory 72, calculator 73 and the analyzer 74 of conjunction.
The processor 71, for obtaining the video data in worker's production process, and the video counts according to the acquisition
According to from the body language information identified in the video data in worker's production process, and the video data according to the acquisition
The body language information identified with this establishes the body language information that the video frame based on the video data is identified with this
Two-dimensional coordinate figure, and image denoising is carried out to the two-dimensional coordinate figure of the foundation, and to the two-dimensional coordinate after image denoising
Indicate that the coordinate data of body language information is labeled, and marks out peak value in figure;Wherein, which includes peak-peak and most
Small leak.
The memory 72, the instruction etc. executed for storage program area, processor 71.
The calculator 73, the peak value for being marked out according to this, taking two peak value sections is a process period, to the work
Activity time during life produces is counted, and the process timing in worker's production process is obtained.
The analyzer 74, for according to the process timing in obtained worker's production process, to the obtained work
Life produce during process timing in same period activity time counted, this for comparing that this counts is each identical
The activity time of periodical process compares activity time abnormal conditions, according to the activity time abnormal conditions, automatically to worker
Carry out the prompt of activity time exception.
Optionally, the processor 71, can be specifically used for:
Using camera shooting mode, the video data in worker's production process is obtained.
Optionally, the processor 71, can be specifically used for:
According to the video data of the acquisition, OpenPose mode is identified using body language, is identified from the video data
Body language information in worker's production process.
Optionally, the calculator 73, can be specifically used for:
Include peak-peak minimum peak according to the peak value that this is marked out, takes two neighboring peak value section as a process
Period carries out coordinate conversion to the peak value section marked out, which is converted to the time number of same timing criteria
According to obtaining the corresponding time data in the adjacent peak-peak section or adjacent minimum peak section, obtained most according to this
Big peak value section and the corresponding time data in minimum peak section, unite to the activity time in worker's production process
Meter, obtains the process timing in worker's production process.
It can be found that above scheme, video data in available worker's production process, and then according to the acquisition
Video data, from the body language information identified in the video data in worker's production process, and then according to the acquisition
Video data and the body language information identified, establish video frame and the limbs identified based on the video data
The two-dimensional coordinate figure of language message, and then image denoising is carried out to the two-dimensional coordinate figure of the foundation, and then to this through image denoising
Indicate that the coordinate data of body language information is labeled, and marks out peak value, and then according to the mark in two-dimensional coordinate figure afterwards
Peak value out, taking two peak value sections is a process period, counts, obtains to the activity time in worker's production process
To the process timing in worker's production process, can be realized can be automatically to the work in worker's production process without artificial participation
Sequence carries out timing, the artificial participation factor in process timing is avoided, so that process timing is more objective and accurate.
Further, above scheme, the process timing in worker's production process that can be obtained according to this, obtains this
Worker's production process in process timing in same period activity time count, this for comparing that this counts is each
The activity time of a same period process, compares activity time abnormal conditions, according to the activity time abnormal conditions, automatically
To worker carry out activity time exception prompt, can be realized in process of production can in time the worker to working efficiency exception into
Row prompt, improves the working efficiency in production process.
In several embodiments provided by the present invention, it should be understood that disclosed system, device and method can
To realize by another way.For example, device embodiments described above are only schematical, for example, module or
The division of unit, only a kind of logical function partition, there may be another division manner in actual implementation, such as multiple units
Or component can be combined or can be integrated into another system, or some features can be ignored or not executed.Another point, institute
Display or the mutual coupling, direct-coupling or communication connection discussed can be through some interfaces, device or unit
Indirect coupling or communication connection can be electrical property, mechanical or other forms.
Unit may or may not be physically separated as illustrated by the separation member, shown as a unit
Component may or may not be physical unit, it can and it is in one place, or may be distributed over multiple networks
On unit.It can select some or all of unit therein according to the actual needs to realize the mesh of present embodiment scheme
's.
In addition, each functional unit in each embodiment of the present invention can integrate in one processing unit, it can also
To be that each unit physically exists alone, can also be integrated in one unit with two or more units.It is above-mentioned integrated
Unit both can take the form of hardware realization, can also realize in the form of software functional units.
It, can if integrated unit is realized in the form of SFU software functional unit and when sold or used as an independent product
To be stored in a computer readable storage medium.Based on this understanding, technical solution of the present invention substantially or
Say that all or part of the part that contributes to existing technology or the technical solution can embody in the form of software products
Out, which is stored in a storage medium, including some instructions are used so that a computer equipment
(can be personal computer, server or the network equipment etc.) or processor (processor) execute each implementation of the present invention
The all or part of the steps of methods.And storage medium above-mentioned include: USB flash disk, mobile hard disk, read-only memory (ROM,
Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic or disk etc. it is various
It can store the medium of program code.
The foregoing is merely section Examples of the invention, are not intended to limit protection scope of the present invention, all utilizations
Equivalent device made by description of the invention and accompanying drawing content or equivalent process transformation are applied directly or indirectly in other correlations
Technical field, be included within the scope of the present invention.
Claims (10)
1. a kind of process clocking method based on human body limb language characterized by comprising
Obtain the video data in worker's production process;
According to the video data of the acquisition, from the body language identified in the video data in worker's production process
Information;
According to the video data of the acquisition and the body language information identified, the view based on the video data is established
The two-dimensional coordinate figure of frequency frame and the body language information identified;
Image denoising is carried out to the two-dimensional coordinate figure of the foundation;
To indicating that the coordinate data of body language information is labeled in the two-dimensional coordinate figure after image denoising, mark out
Peak value;Wherein, the peak value includes peak-peak and minimum peak;
According to the peak value marked out, taking two peak value sections is a process period, in worker's production process
Activity time is counted, and the process timing in worker's production process is obtained.
2. as described in claim 1 based on the process clocking method of human body limb language, which is characterized in that described to obtain one
Video data in worker's production process, comprising:
Using camera shooting mode, the video data in worker's production process is obtained.
3. as claimed in claim 1 or 2 based on the process clocking method of human body limb language, which is characterized in that the basis
The video data of the acquisition, from the body language information identified in the video data in worker's production process, packet
It includes:
The worker is identified from the video data using body language identification method according to the video data of the acquisition
Body language information in production process.
4. as claimed in claim 3 based on the process clocking method of human body limb language, which is characterized in that described according to
The peak value marked out, taking two peak value sections is a process period, is carried out to the activity time in worker's production process
Statistics, obtains the process timing in worker's production process, comprising:
Include peak-peak minimum peak according to the peak value marked out, takes two neighboring peak value section as a process week
Phase carries out coordinate conversion to the peak value section marked out, the peak value section is converted to the time of same timing criteria
Data obtain the corresponding time data in the adjacent peak-peak section or adjacent minimum peak section, are obtained according to described
The corresponding time data in peak-peak section and minimum peak section arrived, when to process in worker's production process
Between counted, obtain the process timing in worker's production process.
5. as claimed in claim 4 based on the process clocking method of human body limb language, which is characterized in that described according to institute
State the peak value marked out, take two peak value sections be a process period, to the activity time in worker's production process into
Row statistics, after obtaining the process timing in worker's production process, further includes:
According to the process timing in obtained worker's production process, in obtained worker's production process
Process timing in same period activity time counted, relatively described in each same period work for counting
The activity time of sequence compares activity time abnormal conditions, according to the activity time abnormal conditions, carries out work to worker automatically
The prompt of sequence time anomaly.
6. a kind of process timekeeping system based on human body limb language characterized by comprising
Video acquisition unit, body language recognition unit, coordinate establish unit, image denoising unit, take peak cell and time
Statistic unit;
The video acquisition unit, for obtaining the video data in worker's production process;
The body language recognition unit identifies institute for the video data according to the acquisition from the video data
State the body language information in worker's production process;
The coordinate establishes unit, for according to the acquisition video data and the body language information identified, build
The video frame of the video data that is based on and the two-dimensional coordinate figure of the body language information identified;
Described image denoises unit, for carrying out image denoising to the two-dimensional coordinate figure of the foundation;
It is described to take peak cell, for the coordinate for indicating body language information in the two-dimensional coordinate figure after image denoising
Data are labeled, and mark out peak value;Wherein, the peak value includes peak-peak and minimum peak;
The time statistic unit, the peak value for marking out according to, taking two peak value sections is a process period, right
Activity time in worker's production process counts, and obtains the process timing in worker's production process.
7. as claimed in claim 6 based on the process timekeeping system of human body limb language, which is characterized in that the video acquisition
Unit is specifically used for:
Using camera shooting mode, the video data in worker's production process is obtained.
8. the process timekeeping system based on human body limb language as claimed in claims 6 or 7, which is characterized in that the limbs
Language identification unit, is specifically used for:
The worker is identified from the video data using body language identification method according to the video data of the acquisition
Body language information in production process.
9. as claimed in claim 8 based on the process timekeeping system of human body limb language, which is characterized in that the time statistics
Unit is specifically used for:
Include peak-peak minimum peak according to the peak value marked out, takes two neighboring peak value section as a process week
Phase carries out coordinate conversion to the peak value section marked out, the peak value section is converted to the time of same timing criteria
Data obtain the corresponding time data in the adjacent peak-peak section or adjacent minimum peak section, are obtained according to described
The corresponding time data in peak-peak section and minimum peak section arrived, when to process in worker's production process
Between counted, obtain the process timing in worker's production process.
10. as claimed in claim 9 based on the process timekeeping system of human body limb language, which is characterized in that described to be based on people
The process timekeeping system of body body language, further includes:
Process chronometric analysis unit, the process timing in worker's production process for obtaining according to are obtained to described
To worker's production process in process timing in same period activity time count, relatively described in count
Each same period process activity time, compare activity time abnormal conditions, it is different according to the activity time
Reason condition carries out the prompt of activity time exception to worker automatically.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811393385.7A CN109492602B (en) | 2018-11-21 | 2018-11-21 | Process timing method and system based on human body language |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811393385.7A CN109492602B (en) | 2018-11-21 | 2018-11-21 | Process timing method and system based on human body language |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109492602A true CN109492602A (en) | 2019-03-19 |
CN109492602B CN109492602B (en) | 2020-11-03 |
Family
ID=65697282
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811393385.7A Active CN109492602B (en) | 2018-11-21 | 2018-11-21 | Process timing method and system based on human body language |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109492602B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111507231A (en) * | 2020-04-10 | 2020-08-07 | 三一重工股份有限公司 | Automatic detection method and system for correctness of process steps |
US11348355B1 (en) | 2020-12-11 | 2022-05-31 | Ford Global Technologies, Llc | Method and system for monitoring manufacturing operations using computer vision for human performed tasks |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8718337B1 (en) * | 2010-06-30 | 2014-05-06 | Imdb.Com, Inc. | Identifying an individual for a role |
CN105469073A (en) * | 2015-12-16 | 2016-04-06 | 安徽创世科技有限公司 | Kinect-based call making and answering monitoring method of driver |
CN108062533A (en) * | 2017-12-28 | 2018-05-22 | 北京达佳互联信息技术有限公司 | Analytic method, system and the mobile terminal of user's limb action |
-
2018
- 2018-11-21 CN CN201811393385.7A patent/CN109492602B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8718337B1 (en) * | 2010-06-30 | 2014-05-06 | Imdb.Com, Inc. | Identifying an individual for a role |
CN105469073A (en) * | 2015-12-16 | 2016-04-06 | 安徽创世科技有限公司 | Kinect-based call making and answering monitoring method of driver |
CN108062533A (en) * | 2017-12-28 | 2018-05-22 | 北京达佳互联信息技术有限公司 | Analytic method, system and the mobile terminal of user's limb action |
Non-Patent Citations (1)
Title |
---|
张建,彭佳林,杜吉祥: "采用共享空间稀疏表示的单幅图像超分辨率方法", 《华侨大学学报(自然科学版)》 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111507231A (en) * | 2020-04-10 | 2020-08-07 | 三一重工股份有限公司 | Automatic detection method and system for correctness of process steps |
CN111507231B (en) * | 2020-04-10 | 2023-06-23 | 盛景智能科技(嘉兴)有限公司 | Automatic detection method and system for correctness of process steps |
US11348355B1 (en) | 2020-12-11 | 2022-05-31 | Ford Global Technologies, Llc | Method and system for monitoring manufacturing operations using computer vision for human performed tasks |
Also Published As
Publication number | Publication date |
---|---|
CN109492602B (en) | 2020-11-03 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102662473B (en) | The device and method of man-machine information interaction is realized based on eye motion recognition | |
CN107483834B (en) | Image processing method, continuous shooting method and device and related medium product | |
CN110111246B (en) | Virtual head portrait generation method and device and storage medium | |
CN103795931B (en) | A kind of information processing method and electronic equipment | |
CN108200334B (en) | Image shooting method and device, storage medium and electronic equipment | |
CN106713811B (en) | Video call method and device | |
CN107194316A (en) | A kind of evaluation method of mood satisfaction, apparatus and system | |
CN109800731A (en) | Fingerprint input method and relevant apparatus | |
CN109492602A (en) | A kind of process clocking method and system based on human body limb language | |
CN104883505A (en) | Electronic equipment and photographing control method therefor | |
CN109977906A (en) | Gesture identification method and system, computer equipment and storage medium | |
CN104915009B (en) | The method and system of gesture anticipation | |
CN110084885A (en) | A kind of cloud and image optimization method, device, equipment and storage medium | |
CN111209811B (en) | Method and system for detecting eyeball attention position in real time | |
CN109660936A (en) | A kind of face and mobile device identification number mapping method, system and readable storage medium storing program for executing | |
CN115601811A (en) | Facial acne detection method and device | |
WO2012143939A2 (en) | System and method for demographic analytics based on multimodal information | |
AU2021240278A1 (en) | Face identification methods and apparatuses | |
CN113052127A (en) | Behavior detection method, behavior detection system, computer equipment and machine readable medium | |
CN114385012A (en) | Motion recognition method and device, electronic equipment and readable storage medium | |
CN113592427A (en) | Method and apparatus for counting man-hours and computer readable storage medium | |
CN107403146A (en) | detection method and related product | |
CN106227608A (en) | A kind of data processing method, terminal and system | |
CN109118163A (en) | Automatically enter the method, apparatus, computer equipment and storage medium of suggestions made after examination | |
CN110503150A (en) | Sample data acquisition method, device 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 |