CN109492602B - Process timing method and system based on human body language - Google Patents

Abstract

The invention discloses a process timing method and system based on human body language. Wherein, the method includes: acquiring video data of a worker in the production process, and then identifying the body language information of the worker in the production process from the video data, and then according to the video data and the body language information. The video frame of the data and the two-dimensional coordinate map of the body language information, and then perform image denoising on the two-dimensional coordinate map, and then label the coordinate data representing the body language information in the two-dimensional coordinate map after image denoising. , mark the peak value, and then according to the marked peak value, take two peak intervals as a process cycle, count the process time in the production process of the worker, and obtain the process timing in the production process of the worker. Through the above method, it is possible to automatically time the processes in the production process of workers without manual participation, so that the process timing is more objective and accurate.

Description

A process timing method and system based on human body language

technical field

The invention relates to the technical field of process timing, in particular to a process timing method and system based on human body language.

Background technique

With the development of computer science and technology, the sewing products industry can use modern computer science and technology to informatize the application of general sewing time in the sewing industry, so that enterprises have a smart, friendly and convenient working environment. Through the computer system software, a powerful static database can be established. The database has the characteristics of both pictures and texts and easy retrieval. It can enable ordinary managers to easily complete the selection, copying and other operations of data, and finally achieve the method analysis and time analysis of the style quickly. Purpose. The computer system software can describe the results of the analysis through graphics, tables, etc., and can perform data operations such as process balance, personnel allocation, machine allocation, production ranking, capacity analysis, etc. through the computer, providing huge management for the sewing industry. support.

In the current sewing industry, the computer management system used is only based on traditional computer software technology. The scope of use and reliability are very limited, especially for timing each process of sewing. At present, the main method is to use the method of workers swiping cards, and swiping cards at the beginning and end of the process to verify the identity and count the sequence. These operations require a large number of manual operations, which is easy to cause waste of personnel, and at the same time, the timing results are not objective and accurate due to the factors of manual participation.

SUMMARY OF THE INVENTION

In view of this, the purpose of the present invention is to propose a process timing method and system based on human body language, which can realize the automatic timing of the processes in the production process of workers without manual participation, and avoid the manual participation factor in the process timing, Make the process timing more objective and accurate.

According to one aspect of the present invention, there is provided a process timing method based on human body language, comprising:

Get the video data of a worker's production process;

According to the acquired video data, identify the body language information of the worker in the production process from the video data;

According to the acquired video data and the identified body language information, establish a two-dimensional coordinate diagram based on the video frame of the video data and the identified body language information;

performing image denoising on the established two-dimensional coordinate map;

Marking the coordinate data representing body language information in the two-dimensional coordinate graph after image denoising, and marking the peak value; wherein, the peak value includes a maximum peak value and a minimum peak value;

According to the marked peaks, two peak intervals are taken as a process cycle, and the process time in the production process of the worker is counted to obtain the process timing in the production process of the worker.

Wherein, the described acquisition of video data in a worker's production process includes:

The video data of a worker's production process is obtained by means of camera.

Wherein, identifying the body language information of the worker in the production process from the video data according to the acquired video data includes:

According to the acquired video data, the body language information in the production process of the worker is identified from the video data by adopting a body language recognition method.

Wherein, according to the marked peaks, two peak intervals are taken as a process cycle, and the process time in the production process of the worker is counted to obtain the process timing in the production process of the worker, including:

According to the marked peak including the maximum peak and minimum peak, take two adjacent peak intervals as a process cycle, perform coordinate transformation on the marked peak interval, and convert the peak interval into time data of the same timing standard , obtain the time data corresponding to the adjacent maximum peak interval or the adjacent minimum peak interval, respectively, according to the obtained time data corresponding to the maximum peak interval and the minimum peak interval, respectively, perform the process time in the production process of the worker. According to statistics, the process timing in the production process of the worker is obtained.

Wherein, according to the marked peaks, two peak intervals are taken as a process cycle, the process time in the production process of the worker is counted, and after the process timing in the production process of the worker is obtained, the process further includes: :

According to the obtained process timing in the production process of the worker, make statistics on the same periodic process time in the obtained process timing in the production process of the worker, and compare the respective identical cycles obtained by the statistics. The process time of the sexual process is compared, and the abnormality of the process time is compared, and according to the abnormality of the process time, the worker is automatically prompted for the abnormality of the process time.

According to one aspect of the present invention, a process timing system based on human body language is provided, comprising:

Video acquisition unit, body language recognition unit, coordinate establishment unit, image denoising unit, peak value unit and time statistics unit;

The video acquisition unit is used to acquire video data in a worker's production process;

The body language recognition unit is used for identifying the body language information in the production process of the worker from the video data according to the acquired video data;

The coordinate establishment unit is used to establish a two-dimensional coordinate diagram based on the video frame of the video data and the identified body language information according to the obtained video data and the identified body language information;

the image denoising unit, configured to perform image denoising on the established two-dimensional coordinate map;

The peak taking unit is used to mark the coordinate data representing body language information in the two-dimensional coordinate graph after image denoising, and mark the peak value; wherein, the peak value includes a maximum peak value and a minimum peak value;

The time statistics unit is configured to, according to the marked peaks, take two peak intervals as a process cycle, perform statistics on the process time in the worker's production process, and obtain the process timing in the worker's production process.

Wherein, the video acquisition unit is specifically used for:

The video data of a worker's production process is obtained by means of camera.

Wherein, the body language recognition unit is specifically used for:

According to the acquired video data, the body language information in the production process of the worker is identified from the video data by adopting a body language recognition method.

Wherein, the time statistics unit is specifically used for:

According to the marked peak including the maximum peak and minimum peak, take two adjacent peak intervals as a process cycle, perform coordinate transformation on the marked peak interval, and convert the peak interval into time data of the same timing standard , obtain the time data corresponding to the adjacent maximum peak interval or the adjacent minimum peak interval, respectively, according to the obtained time data corresponding to the maximum peak interval and the minimum peak interval, respectively, perform the process time in the production process of the worker. According to statistics, the process timing in the production process of the worker is obtained.

Wherein, the process timing system based on human body language also includes:

A process timing analysis unit, configured to perform statistics on the same periodic process time in the obtained process timing in the worker's production process according to the obtained process timing in the worker's production process, and compare the statistics The process time of each of the same periodic processes is obtained, and the abnormality of the process time is compared, and according to the abnormality of the process time, the worker is automatically prompted for the abnormality of the process time.

It can be found that the above scheme can obtain the video data of a worker in the production process, and then according to the obtained video data, identify the body language information of the worker in the production process from the video data, and then according to the obtained video data and The identified body language information, establishes a two-dimensional coordinate map based on the video frame of the video data and the identified body language information, and then performs image denoising on the established two-dimensional coordinate map, and then denoises the processed image. The coordinate data representing body language information in the two-dimensional coordinate graph after noise is marked, and the peak value is marked, and then according to the marked peak value, two peak intervals are taken as a process cycle, and the process time in the production process of the worker is carried out. According to statistics, the process timing in the production process of the worker can be obtained, which can realize the automatic timing of the process in the production process of the worker without manual participation, avoid the manual participation factor in the process timing, and make the process timing more objective and accurate.

Further, the above scheme, according to the obtained process timing in the production process of the worker, the same periodic process time in the obtained process timing in the production process of the worker can be counted, and the statistics obtained are compared. The process time of the periodic process is compared, and the abnormality of the process time is compared. According to the abnormality of the process time, the workers are automatically prompted for the abnormality of the process time, which can prompt the workers with abnormal work efficiency in time during the production process and improve production. efficiency in the process.

Description of drawings

In order to explain the embodiments of the present invention or the technical solutions in the prior art more clearly, the following briefly introduces the accompanying drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only These are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can also be obtained according to these drawings without creative efforts.

1 is a schematic flowchart of an embodiment of a process timing method based on human body language of the present invention;

Fig. 2 is a process timing method based on human body language of the present invention to establish a two-dimensional coordinate diagram based on a video frame of video data in a worker's production process and a recognized two-dimensional coordinate diagram based on body language information in a worker's production process an example schematic diagram;

Fig. 3 is the video frame of the video data of a worker's production process after image denoising in an embodiment of the process timing method based on human body language of the present invention and the identified two images based on the body language information of a worker's production process. A schematic diagram of an example of a two-dimensional waveform diagram of dimensional coordinates;

4 is a schematic flowchart of another embodiment of a process timing method based on human body language of the present invention;

5 is a schematic structural diagram of an embodiment of a process timing system based on human body language of the present invention;

6 is a schematic structural diagram of another embodiment of the process timing system based on human body language of the present invention;

7 is a schematic structural diagram of another embodiment of the process timing system based on human body language of the present invention.

Detailed ways

The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It is particularly pointed out that the following examples are only used to illustrate the present invention, but do not limit the scope of the present invention. Likewise, the following embodiments are only some rather than all embodiments of the present invention, and all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of the present invention.

The invention provides a process timing method based on human body language, which can automatically time the processes in the production process of workers without manual participation, avoids the manual participation factor in the process timing, and makes the process timing more objective and accurate.

Please refer to FIG. 1 . FIG. 1 is a schematic flowchart of an embodiment of a process timing method based on human body language of the present invention. It should be noted that, if there is substantially the same result, the method of the present invention is not limited to the sequence of the processes shown in FIG. 1 . As shown in Figure 1, the method includes the following steps:

S101: Acquire video data of a worker in the production process.

Among them, the video data of a worker's production process should be obtained, which may include:

The video data of a worker's production process is obtained by means of camera.

S102: According to the acquired video data, identify the body language information of the worker in the production process from the video data.

Wherein, identifying the body language information of the worker in the production process from the video data according to the acquired video data may include:

According to the acquired video data, the OpenPose (body language recognition) method is used to identify the body language information of the worker in the production process from the video data.

S103: According to the acquired video data and the identified body language information, establish a two-dimensional coordinate map of the video frame based on the video data and the identified body language information.

Please refer to Fig. 2, Fig. 2 is a process timing method based on human body language in an embodiment of the present invention to establish a video frame based on video data in a worker's production process and the identified image based on body language information in a worker's production process. An example schematic diagram of a two-dimensional coordinate graph. As shown in Figure 2, taking the sewing process as an example, in actual operation, most of the sewing processes have to go through a series of operations such as worker picking, sewing, and unwinding, and it is easy to find that in normal processes, workers always Use the left hand to pick up the material, and then sew it under the sewing machine. After the sewing is completed, the finished product is placed in the finished product area.

As shown in Figure 2, when the worker takes the material with his left hand, in the picture of the video frame, the X direction of the worker's left wrist is the farthest from the origin, so in the video frame, the coordinate of the worker's left wrist in the X direction is the maximum value, Draw a two-dimensional diagram of the X-direction value of the frequency frame and the number of video frames as shown in Figure 2. As shown in Figure 2, the X axis represents the number of frames of the frequency frame, and the Y axis represents the coordinates of the worker's left wrist in the X direction of the video frame, then the maximum value in the X direction corresponds to its peak point, that is, the peak point indicates that the worker is in the video frame. The time to take the material. Subsequently, according to the characteristics of the production process and the two-dimensional map, the process in the production process of a worker is automatically timed.

S104: Perform image denoising on the established two-dimensional coordinate map.

S105: Mark the coordinate data representing body language information in the two-dimensional coordinate graph after denoising the image, and mark a peak value; wherein, the peak value includes a maximum peak value and a minimum peak value.

Please refer to FIG. 3, FIG. 3 is a video frame of video data in a worker's production process after image denoising in an embodiment of the process timing method based on human body language of the present invention and the identified video frame based on a worker's production process An example schematic diagram of a two-dimensional waveform of two-dimensional coordinates of body language information. As shown in Figure 3, taking the sewing process as an example, the first curve from top to bottom represents the body language of a worker during the production process, such as the X coordinate of the human wrist joint point and the video data of the worker during the production process The two-dimensional waveform diagram of the number of frames, the second curve from top to bottom represents the body language of the worker during the production process, such as the Y coordinate of the joint point of the human wrist and the second of the frame number of the video data of the worker during the production process. Dimensional waveform diagram, the dots mark the peak value, which can indicate the moment when the worker takes material with his left hand. The peak point marked in the form of dots has two scatter-point connection graphs. It is a binary classifier that can be constructed by GoogleNet (convolutional neural network) to classify whether the current workbench is empty. When the workbench If it is not empty, it is a high threshold. At this time, a scatter plot with a process threshold value of 600 can be selected. On the contrary, it is a low threshold value below. At this time, a scatter plot with a process threshold value of 580 can be selected. As shown in Figure 3, it can be clearly seen that at the peak point, the scatter points with a high threshold value, such as the above threshold value of 600, have large discontinuities, indicating that the workbench is empty for a period of time at this moment.

S106: According to the marked peak value, two peak intervals are taken as a process cycle, and the process time in the production process of the worker is counted to obtain the process timing in the production process of the worker.

Wherein, according to the marked peak value, two peak intervals are taken as a process cycle, the process time in the production process of the worker is counted, and the process timing in the production process of the worker is obtained, which may include:

According to the marked peak including the maximum peak and the minimum peak, take two adjacent peak intervals as a process cycle, perform coordinate transformation on the marked peak interval, and convert the peak interval into the time data of the same timing standard to obtain the The time data corresponding to the adjacent maximum peak interval or the adjacent minimum peak interval respectively, according to the obtained time data corresponding to the maximum peak interval and the minimum peak interval respectively, the process time in the production process of the worker is counted, and the production of the worker is obtained. The timing of the steps in the process.

Wherein, according to the marked peak value, two peak intervals are taken as a process cycle, the process time in the production process of the worker is counted, and after the process timing in the production process of the worker is obtained, it can also include:

According to the obtained process timing in the production process of the worker, make statistics on the same periodic process time in the obtained process timing in the production process of the worker, and compare the calculated process time of each of the same periodic processes, The abnormality of the process time is compared, and according to the abnormality of the process time, the worker is automatically prompted for the abnormality of the process time.

In this embodiment, the process timing method based on human body language can be applied to the sewing product industry, and the visual camera method can be used to time the process in the production process, which can avoid artificial factors in the process timing.

In this embodiment, the process timing method based on human body language can use computer vision technology, especially human body posture estimation technology, to analyze the working time of the workers in the production process, that is, the process time, and automatically extract the production time of each product. The time period provides core data for subsequent management and work efficiency analysis.

In this embodiment, the process timing method based on human body language does not require manual timekeeping with a stopwatch or workers themselves, and video recording can be performed by, for example, videography, and the content of the video can be automatically analyzed to obtain process time results. Achieving an active process timing process without artificial factors can fill the gap in the market such as vision-based sewing process timing systems.

It can be found that the above scheme can obtain the video data of a worker in the production process, and then according to the obtained video data, identify the body language information of the worker in the production process from the video data, and then according to the obtained video data and The identified body language information, establishes a two-dimensional coordinate map based on the video frame of the video data and the identified body language information, and then performs image denoising on the established two-dimensional coordinate map, and then denoises the processed image. The coordinate data representing body language information in the two-dimensional coordinate graph after noise is marked, and the peak value is marked, and then according to the marked peak value, two peak intervals are taken as a process cycle, and the process time in the production process of the worker is carried out. According to statistics, the process timing in the production process of the worker can be obtained, which can realize the automatic timing of the process in the production process of the worker without manual participation, avoid the manual participation factor in the process timing, and make the process timing more objective and accurate.

Please refer to FIG. 4 . FIG. 4 is a schematic flowchart of another embodiment of the process timing method based on human body language of the present invention. In this embodiment, the method includes the following steps:

S401: Acquire video data of a worker in the production process.

It can be described in S101 above, and details are not repeated here.

S402: Identify the body language information of the worker in the production process from the video data according to the acquired video data.

It can be described in the above S102, which is not repeated here.

S403: According to the acquired video data and the identified body language information, establish a two-dimensional coordinate map based on the video frame of the video data and the identified body language information.

It can be described in the above S103, which is not repeated here.

S404: Perform image denoising on the established two-dimensional coordinate map.

S405: Mark the coordinate data representing body language information in the two-dimensional coordinate graph after denoising the image, and mark a peak value; wherein, the peak value includes a maximum peak value and a minimum peak value.

It can be described in the above S105, which is not repeated here.

S406: According to the marked peak value, two peak intervals are taken as a process cycle, and the process time in the production process of the worker is counted to obtain the process timing in the production process of the worker.

It can be described in the above S106, which is not repeated here.

S407: According to the obtained process timing in the production process of the worker, perform statistics on the same periodic process time in the obtained process timing in the production process of the worker, and compare the counted processes of the same periodic process Time, compare the abnormality of the process time, and automatically prompt the worker of the abnormality of the process time according to the abnormality of the process time.

It can be found that, in this embodiment, according to the obtained process timing in the production process of the worker, the same periodic process time in the obtained process timing in the production process of the worker can be counted, and the statistics obtained can be compared. The process time of each of the same periodic processes is compared, and the abnormality of the process time is compared. According to the abnormality of the process time, the worker is automatically prompted for the abnormality of the process time, which can realize the prompt of the abnormal work efficiency in the production process. , improve the work efficiency in the production process.

The invention also provides a process timing system based on human body language, which can automatically time the processes in the production process of workers without manual participation, avoids the manual participation factor in the process timing, and makes the process timing more objective and accurate.

Please refer to FIG. 5 , which is a schematic structural diagram of an embodiment of a process timing system based on human body language of the present invention. In this embodiment, the process timing system 50 based on human body language includes a video acquisition unit 51 , a body language recognition unit 52 , a coordinate establishment unit 53 , an image denoising unit 54 , a peak value unit 55 and a time statistics unit 56 .

The video acquisition unit 51 is used to acquire video data of a worker in the production process.

The body language identification unit 52 is configured to identify the body language information of the worker in the production process from the video data according to the acquired video data.

The coordinate establishing unit 53 is configured to, according to the acquired video data and the identified body language information, establish a two-dimensional coordinate map of the video frame based on the video data and the identified body language information.

The image denoising unit 54 is configured to perform image denoising on the established two-dimensional coordinate map.

The peak taking unit 55 is used to mark the coordinate data representing body language information in the two-dimensional coordinate graph after image denoising, and mark the peak value; wherein, the peak value includes a maximum peak value and a minimum peak value.

The time statistic unit 56 is configured to take two peak intervals as a process cycle according to the marked peak value, perform statistics on the process time in the production process of the worker, and obtain the process timing in the production process of the worker.

Optionally, the video acquisition unit 51 can be specifically used for:

The video data of a worker's production process is obtained by means of camera.

Optionally, the body language recognition unit 52 can be specifically used for:

According to the obtained video data, the body language information of the worker in the production process is identified from the video data by using the method of body language recognition OpenPose.

Optionally, the time statistics unit 56 can be specifically used for:

According to the marked peak including the maximum peak and the minimum peak, take two adjacent peak intervals as a process cycle, perform coordinate transformation on the marked peak interval, and convert the peak interval into the time data of the same timing standard to obtain the The time data corresponding to the adjacent maximum peak interval or the adjacent minimum peak interval respectively, according to the obtained time data corresponding to the maximum peak interval and the minimum peak interval respectively, the process time in the production process of the worker is counted, and the production of the worker is obtained. The timing of the steps in the process.

In this embodiment, the video acquisition unit 51 can be any terminal that can acquire video data of a worker in the production process by means of a camera, such as a video camera, a camera, a mobile phone, etc., which is not limited in the present invention.

Please refer to FIG. 6 , which is a schematic structural diagram of another embodiment of the process timing system based on human body language of the present invention. Different from the previous embodiment, the process timing system 60 based on human body language in this embodiment further includes: a process timing analysis unit 61 .

The process timing analysis unit 61 is configured to perform statistics on the same periodic process time in the obtained process timing in the worker's production process according to the obtained process timing in the worker's production process, and compare the calculated The process time of each same periodic process is compared, and the abnormality of the process time is compared. According to the abnormality of the process time, the worker is automatically prompted for the abnormality of the process time.

Each unit module of the process timing system 50/60 based on human body language can respectively execute the corresponding steps in the above method embodiments, so each unit module will not be repeated here. For details, please refer to the description of the corresponding steps above.

Please refer to FIG. 7 , which is a schematic structural diagram of another embodiment of the process timing system based on human body language of the present invention. Each unit module of the process timing system based on human body language can respectively execute the corresponding steps in the above method embodiments. For related content, please refer to the detailed description in the above method, which will not be repeated here.

In this embodiment, the process timing system based on human body language includes: a processor 71 , a memory 72 coupled with the processor 71 , a calculator 73 and an analyzer 74 .

The processor 71 is configured to acquire video data of a worker in the production process, and according to the acquired video data, identify the body language information of the worker in the production process from the video data, and according to the acquired video data and The identified body language information, establishes a two-dimensional coordinate map based on the video frame of the video data and the identified body language information, and performs image denoising on the established two-dimensional coordinate map, and denoises the image The coordinate data representing the body language information in the two-dimensional coordinate graph after the noise is marked, and the peak value is marked; wherein, the peak value includes the maximum peak value and the minimum peak value.

The memory 72 is used to store the operating system, instructions executed by the processor 71, and the like.

The calculator 73 is used for taking two peak intervals as a process cycle according to the marked peak value, and making statistics on the process time in the production process of the worker to obtain the process timing in the worker's production process.

The analyzer 74 is configured to perform statistics on the same periodic process time in the obtained process timing in the worker's production process according to the obtained process timing in the worker's production process, and compare the statistics of the same periodic process time. The process time of the periodic process is compared and the abnormality of the process time is compared. According to the abnormality of the process time, the worker is automatically prompted for the abnormality of the process time.

Optionally, the processor 71 may be specifically used for:

The video data of a worker's production process is obtained by means of camera.

Optionally, the processor 71 may be specifically used for:

According to the obtained video data, the body language information of the worker in the production process is identified from the video data by using the method of body language recognition OpenPose.

Optionally, the calculator 73 can be specifically used for:

According to the marked peak including the maximum peak and the minimum peak, take two adjacent peak intervals as a process cycle, perform coordinate transformation on the marked peak interval, and convert the peak interval into the time data of the same timing standard to obtain the The time data corresponding to the adjacent maximum peak interval or the adjacent minimum peak interval respectively, according to the obtained time data corresponding to the maximum peak interval and the minimum peak interval respectively, the process time in the production process of the worker is counted, and the production of the worker is obtained. The timing of the steps in the process.

It can be found that the above scheme can obtain the video data of a worker in the production process, and then according to the obtained video data, identify the body language information of the worker in the production process from the video data, and then according to the obtained video data and The identified body language information, establishes a two-dimensional coordinate map based on the video frame of the video data and the identified body language information, and then performs image denoising on the established two-dimensional coordinate map, and then denoises the processed image. The coordinate data representing body language information in the two-dimensional coordinate graph after noise is marked, and the peak value is marked, and then according to the marked peak value, two peak intervals are taken as a process cycle, and the process time in the production process of the worker is carried out. According to statistics, the process timing in the production process of the worker can be obtained, which can realize the automatic timing of the process in the production process of the worker without manual participation, avoid the manual participation factor in the process timing, and make the process timing more objective and accurate.

Further, the above scheme, according to the obtained process timing in the production process of the worker, the same periodic process time in the obtained process timing in the production process of the worker can be counted, and the statistics obtained are compared. The process time of the periodic process is compared, and the abnormality of the process time is compared. According to the abnormality of the process time, the workers are automatically prompted for the abnormality of the process time, which can prompt the workers with abnormal work efficiency in time during the production process and improve production. efficiency in the process.

In the several embodiments provided by the present invention, it should be understood that the disclosed system, apparatus and method may be implemented in other manners. For example, the apparatus implementations described above are only illustrative, for example, the division of modules or units is only a logical function division, and there may be other divisions in actual implementation, for example, multiple units or components may be combined or Can be integrated into another system, or some features can be ignored, or not implemented. On the other hand, the shown or discussed mutual coupling or direct coupling or communication connection may be through some interfaces, indirect coupling or communication connection of devices or units, and may be in electrical, mechanical or other forms.

Units described as separate components may or may not be physically separated, and components shown as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution in this implementation manner.

In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit. The above-mentioned integrated units may be implemented in the form of hardware, or may be implemented in the form of software functional units.

The integrated unit, if implemented as a software functional unit and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium , including several instructions to make a computer device (which may be a personal computer, a server, or a network device, etc.) or a processor (processor) to execute all or part of the steps of the methods in the various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, Read-Only Memory (ROM, Read-Only Memory), Random Access Memory (RAM, Random Access Memory), magnetic disk or optical disk and other media that can store program codes .

The above descriptions are only part of the embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Any equivalent device or equivalent process transformation made by using the contents of the description and drawings of the present invention, or directly or indirectly applied to other related All technical fields are similarly included in the scope of patent protection of the present invention.

Claims (8)

1. a process timing method based on human body language, is characterized in that, comprises: Get the video data of a worker's production process; According to the acquired video data, identify the body language information of the worker in the production process from the video data; According to the acquired video data and the identified body language information, establish a two-dimensional coordinate diagram based on the video frame of the video data and the identified body language information; Perform image denoising on the established two-dimensional coordinate map; Mark the coordinate data representing body language information in the two-dimensional coordinate graph after image denoising, and mark the peak value; wherein, the peak value includes a maximum peak value and a minimum peak value; According to the marked peaks, two peak intervals are taken as a process cycle, and the process time in the production process of the worker is counted to obtain the process timing in the production process of the worker; including: According to the marked peaks including the maximum peak and the minimum peak, take two adjacent peak intervals as a process cycle, perform coordinate transformation on the marked peak interval, and convert the peak interval into the time data of the same timing standard to obtain The time data corresponding to the adjacent maximum peak interval or the adjacent minimum peak interval respectively, according to the obtained time data corresponding to the maximum peak interval and the minimum peak interval, respectively, perform statistics on the process time in the production process of the worker, and obtain the result. Describe the process timing in the worker's production process. 2. the process timing method based on human body language as claimed in claim 1, is characterized in that, described obtaining the video data in a worker's production process, comprises: The video data of a worker's production process is obtained by means of camera. 3. The process timing method based on human body language according to claim 1 or 2, wherein, according to the acquired video data, the limbs in the production process of the worker are identified from the video data Linguistic information, including: According to the acquired video data, the body language information in the production process of the worker is identified from the video data by adopting a body language recognition method. 4. The process timing method based on human body language as claimed in claim 1, characterized in that, in the peaks marked according to the marking, two peak intervals are taken as one process cycle, and in the production process of the workers, The process time is counted, and after the process timing in the production process of the worker is obtained, it also includes: According to the obtained process timing in the worker's production process, the same periodic process time in the obtained process timing in the worker's production process is counted, and the counted processes of each same periodic process are compared. time, compares the abnormality of the process time, and automatically prompts the worker of the abnormality of the process time according to the abnormality of the process time. 5. A process timing system based on human body language, characterized in that, comprising: Video acquisition unit, body language recognition unit, coordinate establishment unit, image denoising unit, peak value unit and time statistics unit; The video acquisition unit is used to acquire video data in a worker's production process; The body language recognition unit is used to identify the body language information of the worker in the production process from the video data according to the acquired video data; The coordinate establishment unit is used to establish a two-dimensional coordinate diagram based on the video frame of the video data and the identified body language information according to the acquired video data and the identified body language information; The image denoising unit is used to perform image denoising on the established two-dimensional coordinate map; The peak-taking unit is used to mark the coordinate data representing body language information in the two-dimensional coordinate graph after image denoising, and mark the peak value; wherein, the peak value includes a maximum peak value and a minimum peak value; The time statistics unit is configured to take two peak intervals as a process cycle according to the marked peak value, and perform statistics on the process time in the production process of the worker to obtain the process timing in the production process of the worker; include: According to the marked peak including the maximum peak and minimum peak, take two adjacent peak intervals as a process cycle, perform coordinate transformation on the marked peak interval, and convert the peak interval into time data of the same timing standard , obtain the time data corresponding to the adjacent maximum peak interval or the adjacent minimum peak interval respectively, and according to the obtained time data corresponding to the maximum peak interval and the minimum peak interval respectively, perform statistics on the process time in the production process of the worker, and obtain the Describe the process timing in the worker's production process. 6. the process timing system based on human body language as claimed in claim 5, is characterized in that, described video acquisition unit is specially used for: The video data of a worker's production process is obtained by means of camera. 7. The process timing system based on human body language as claimed in claim 5 or 6, wherein the body language recognition unit is specifically used for: According to the acquired video data, the body language information in the production process of the worker is identified from the video data by adopting a body language recognition method. 8. The process timing system based on human body language as claimed in claim 5, wherein the process timing system based on human body language further comprises: The process timing analysis unit is configured to perform statistics on the same periodic process time in the obtained process timing in the worker's production process according to the obtained process timing in the worker's production process, and compare the calculated The process time of each same periodic process is compared, and the abnormality of the process time is compared, and according to the abnormality of the process time, the worker is automatically prompted for the abnormality of the process time.

Images