CN111563726B

CN111563726B - Method, device, equipment and computer readable storage medium for enterprise rectification supervision

Info

Publication number: CN111563726B
Application number: CN202010369989.9A
Authority: CN
Inventors: 高燕; 黄哲
Original assignee: Ping An International Smart City Technology Co Ltd
Current assignee: Ping An International Smart City Technology Co Ltd
Priority date: 2020-04-30
Filing date: 2020-04-30
Publication date: 2024-07-05
Anticipated expiration: 2040-04-30
Also published as: CN111563726A; WO2021217943A1

Abstract

The invention provides an enterprise rectification supervision method, device, equipment and a computer readable storage medium, wherein the method comprises the following steps: acquiring an initial video before modification and a modified video after modification of an enterprise to be modified, and respectively extracting a first picture and a second picture with the same time point from the initial video and the modified video; judging whether the rectification environment corresponding to the rectification video is effective or not according to the first picture and the second picture; if the modification environment is effective, obtaining a modification task list corresponding to the initial video; analyzing the rectification video according to the rectification task list, generating an analysis result, and monitoring rectification of the enterprise to be rectified according to the analysis result. According to the invention, based on the data processing technology, after the real and effective correction environment is determined, the correction video is analyzed according to the correction task list corresponding to the initial video, so that the correction of an enterprise to be corrected is monitored, the authenticity of the correction environment is ensured, and the monitoring efficiency is improved.

Description

Method, device, equipment and computer readable storage medium for enterprise rectification supervision

Technical Field

The present invention relates to the field of data processing technologies, and in particular, to an enterprise rectification supervision method, an enterprise rectification supervision device, an enterprise rectification supervision apparatus, and a computer readable storage medium.

Background

Along with the improvement of living standard, catering enterprises are vigorously developed, but the kitchen of the current part of catering enterprises has a sanitary problem to influence the health of consumers, so that the supervision and inspection of the kitchen of the catering enterprises are particularly important.

Currently, the supervision department usually performs supervision and inspection on the kitchen of the catering enterprise through a video inspection mode, and once the matters requiring regulation and improvement are found out, a requirement for regulation and improvement is issued to the catering enterprise. The catering unit needs to finish the modification of the kitchen before a specified period and report to the supervision department. Under the traditional supervision mode, after receiving feedback of finishing the modification of the dining enterprise, the supervision person rechecks the modification effect of the dining enterprise. Once the secondary correction is required after the secondary correction is judged to be not in place, the supervisor needs to perform secondary correction again. The whole supervision and inspection process needs to consume a great deal of time of supervision personnel, and the manpower cost is high, and the supervision and inspection efficiency is low.

Disclosure of Invention

The invention mainly aims to provide an enterprise rectification supervision method, device, equipment and computer readable storage medium, and aims to solve the technical problems of high labor cost and low efficiency in supervision and inspection of kitchen after catering enterprises in the prior art.

In order to achieve the above object, an embodiment of the present invention provides an enterprise rectification supervision method, including the following steps:

Acquiring an initial video before modification and a modified video after modification of an enterprise to be modified, and respectively extracting a first picture and a second picture with the same time point from the initial video and the modified video;

Judging whether the rectification environment corresponding to the rectification video is effective or not according to the first picture and the second picture;

If the modification environment is effective, acquiring a modification task list corresponding to the initial video;

Analyzing the rectification video according to the rectification task list, generating an analysis result, and monitoring rectification of the enterprise to be rectified according to the analysis result.

Preferably, the step of determining whether the modification environment corresponding to the modification video is valid according to the first picture and the second picture includes:

extracting first environmental feature points in the first picture, and forming each first environmental feature point into a first feature matrix;

Extracting second environmental feature points in the second picture, and forming each second environmental feature point into a second feature matrix;

And determining a similarity parameter between the first feature matrix and the second feature matrix, and judging whether a rectification environment corresponding to the rectified video is effective or not according to the similarity parameter.

Preferably, the step of extracting first environmental feature points in the first picture and forming each of the first environmental feature points into a first feature matrix includes:

identifying static objects in the first picture, and detecting attribute information of each static object, wherein the attribute information at least comprises names, colors, sizes and coordinates;

forming attribute information of each static object into an attribute sequence, and taking each attribute sequence as a first environment characteristic point in the first picture;

And arranging the first environmental characteristic points to form a first characteristic matrix.

Preferably, the step of obtaining the modification task sheet corresponding to the initial video includes:

analyzing the initial video, and extracting frames of the analyzed initial video according to a preset time interval to obtain a plurality of video frames;

And identifying whether the to-be-modified content exists in the plurality of video frames, and if so, generating the to-be-modified content into a modification task list corresponding to the initial video.

Preferably, the step of parsing the modification video according to the modification task sheet to generate a parsing result includes:

Analyzing and extracting frames from the rectifying video according to the preset time interval to obtain a plurality of rectified pictures;

Judging whether the plurality of rectified pictures contain all to-be-rectified contents in the rectified task list, and if so, generating the plurality of rectified pictures into the analysis result;

and if not, adjusting the preset time interval, and executing the step of analyzing and extracting frames of the rectifying video according to the adjusted preset time interval.

Preferably, the step of monitoring the modification of the enterprise to be modified according to the analysis result includes:

Acquiring a modification requirement in the modification task list, and judging whether the analysis result is matched with the modification requirement;

if the modification requirement is matched with the modification requirement, finishing modification supervision of the enterprise to be modified;

And if the prompt message is not matched with the modification requirement, outputting a prompt message for continuing modification.

Preferably, before the step of extracting the first picture and the second picture with the same time point from the initial video and the rectification video, the method includes:

extracting a first generation time of the initial video and a second generation time of the correction video, and judging whether the correction video is generated within a preset correction period according to the first generation time and the second generation time;

If the first picture and the second picture are generated within the preset rectifying period, executing the step of respectively extracting the first picture and the second picture with the same time point from the initial video and the rectifying video;

if the correction time is not within the preset correction time limit, outputting prompt information of invalid correction.

In order to achieve the above object, the present invention further provides an enterprise rectification supervision apparatus, the enterprise rectification supervision apparatus including:

The extraction module is used for acquiring an initial video before the rectification and a rectification video after the rectification of the enterprise to be rectified, and respectively extracting a first picture and a second picture with the same time point from the initial video and the rectification video;

The judging module is used for judging whether the rectifying environment corresponding to the rectifying video is effective or not according to the first picture and the second picture;

the acquisition module is used for acquiring the modification task list corresponding to the initial video if the modification environment is effective;

And the monitoring module is used for analyzing the modification video according to the modification task list, generating an analysis result and monitoring modification of the enterprise to be modified according to the analysis result.

Further, in order to achieve the above object, the present invention also provides an enterprise modification supervisory apparatus, where the enterprise modification supervisory apparatus includes a memory, a processor, and an enterprise modification supervisory program stored in the memory and capable of running on the processor, where the enterprise modification supervisory program implements the steps of the enterprise modification supervisory method described above when executed by the processor.

In addition, in order to achieve the above object, the present invention further provides a computer readable storage medium, on which an enterprise modification supervisory program is stored, the enterprise modification supervisory program implementing the steps of the enterprise modification supervisory method described above when being executed by a processor.

The invention provides an enterprise rectification supervision method, device, equipment and a computer readable storage medium, wherein after an initial video before rectification and a rectification video after rectification of an enterprise to be rectified are obtained, a first picture and a second picture with the same time point are respectively extracted from the initial video and the rectification video; judging whether the rectification environment corresponding to the rectification video is effective or not according to the first picture and the second picture; if the modification environment is effective, obtaining a modification task list corresponding to the initial video; and then analyzing the rectification video according to the rectification task list, and generating an analysis result so as to monitor the rectification of the enterprise to be rectified according to the analysis result. Because the first picture and the second picture are derived from the same time point in the initial video and the rectification video, if the rectification video is a real video, supervision environments represented by the first picture and the second picture are the same, so that the effectiveness of the rectification environment judged according to the first picture and the second picture is more real and accurate. On the basis, the rectification video is analyzed according to the rectification task list corresponding to the initial video, so that rectification of an enterprise to be rectified is supervised, frequent review of supervision personnel is avoided, labor cost is saved, and supervision efficiency is improved while authenticity of a rectification environment is ensured.

Drawings

FIG. 1 is a schematic diagram of an enterprise improvement monitoring device in a hardware operating environment according to an embodiment of the present invention;

FIG. 2 is a flowchart of a first embodiment of the enterprise rectification supervision method according to the present invention;

Fig. 3 is a schematic diagram of functional modules of a preferred embodiment of the enterprise rectification and supervision apparatus according to the present invention.

The achievement of the objects, functional features and advantages of the present invention will be further described with reference to the accompanying drawings, in conjunction with the embodiments.

Detailed Description

It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.

As shown in fig. 1, fig. 1 is a schematic structural diagram of an enterprise rectification supervision apparatus of a hardware running environment according to an embodiment of the present invention.

In the following description, suffixes such as "module", "component", or "unit" for representing elements are used only for facilitating the description of the present invention, and have no specific meaning per se. Thus, "module," "component," or "unit" may be used in combination.

The enterprise rectifying and supervising device can be a PC, or can be movable terminal devices such as a tablet computer and a portable computer.

As shown in fig. 1, the enterprise rectification supervision apparatus may include: a processor 1001, such as a CPU, a network interface 1004, a user interface 1003, a memory 1005, a communication bus 1002. Wherein the communication bus 1002 is used to enable connected communication between these components. The user interface 1003 may include a Display, an input unit such as a Keyboard (Keyboard), and the optional user interface 1003 may further include a standard wired interface, a wireless interface. The network interface 1004 may optionally include a standard wired interface, a wireless interface (e.g., WI-FI interface). The memory 1005 may be a high-speed RAM memory or a stable memory (non-volatile memory), such as a disk memory. The memory 1005 may also optionally be a storage device separate from the processor 1001 described above.

Those skilled in the art will appreciate that the enterprise retrofit supervisory device structure shown in fig. 1 does not constitute a limitation of the enterprise retrofit supervisory device, and may include more or fewer components than shown, or may combine certain components, or may have a different arrangement of components.

As shown in fig. 1, an operating system, a network communication module, a user interface module, and a detection program may be included in the memory 1005, which is a computer-readable storage medium.

In the device shown in fig. 1, the network interface 1004 is mainly used for connecting to a background server, and performing data communication with the background server; the user interface 1003 is mainly used for connecting a client (user side) and performing data communication with the client; and the processor 1001 may be configured to call a detection program stored in the memory 1005 and perform the following operations:

Further, the step of determining whether the modification environment corresponding to the modification video is valid according to the first picture and the second picture includes:

Further, the step of extracting the first environmental feature points in the first picture and forming each of the first environmental feature points into a first feature matrix includes:

Further, the step of obtaining the modification task sheet corresponding to the initial video includes:

Further, the step of analyzing the modification video according to the modification task sheet and generating an analysis result includes:

Further, the step of monitoring the modification of the enterprise to be modified according to the analysis result includes:

Further, before the step of extracting the first picture and the second picture with the same time point from the initial video and the modified video, respectively, the processor 1001 may be configured to call a detection program stored in the memory 1005, and perform the following operations:

The specific implementation manner of the enterprise modification monitoring device is basically the same as the following embodiments of the enterprise modification monitoring method, and is not described herein.

In order that the above-described aspects may be better understood, exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

In order to better understand the above technical solutions, the following detailed description will refer to the accompanying drawings and specific embodiments.

Referring to fig. 2, a first embodiment of the present invention provides a flowchart of an enterprise rectification supervision method. In this embodiment, the enterprise rectification supervision method includes the following steps:

Step S10, an initial video before modification and a modified video after modification of an enterprise to be modified are obtained, and a first picture and a second picture with the same time point are respectively extracted from the initial video and the modified video;

The enterprise rectification supervision method in the embodiment is applied to the server and is suitable for supervising rectification of the enterprise through the server. The enterprises can be various units needing to carry out environment modification and behavior modification, and the embodiment is preferably described by a kitchen of the catering enterprises. The kitchen environment monitoring system monitors dirty, messy and bad environments in the kitchen of the catering enterprises and the correction of the behavior that catering service staff does not wear working caps, working gloves and the like.

Further, the server is in communication connection with the camera capable of shooting video, so that the camera can transmit the shot video to the server for processing. The camera can be installed at a fixed position and can rotate at the fixed position, the rotating angle comprises a kitchen range of a catering enterprise, the kitchen is shot, and the consistency of the shot environments of each rotation period is ensured. In addition, the camera can also exist in the form of video inspection, and the inspection is performed at a fixed speed and a fixed route so as to ensure the consistency of the photographed environment.

Further, the camera transmits the shot video information to the server according to a preset checking period, and for the video information uploaded by a certain catering enterprise for the first time, the server judges whether the content needing to be modified exists in the video information, if the content needing to be modified exists, the enterprise is taken as the enterprise to be modified, and the video information is taken as the initial video before the modification of the enterprise to be modified. Different modification periods are preset in the server according to the difficulty level of the needed modification content, the corresponding modification period can be determined according to the needed modification content in the initial video, and the needed modification content and the modification period are sent to the enterprise to be modified in a notification mode. After receiving the video information uploaded by the enterprise to be rectified again, the server acquires the video information as rectified video, and acquires the initial video stored in the storage unit from the storage unit so as to monitor whether the rectification of the enterprise to be rectified is in place or not through the initial video and the rectification video.

Further, in order to ensure that the initial video and the rectification video are shot aiming at the same scene, namely, ensure that the rectification environment corresponding to the rectification video is an environment which is really needed to be rectified in the enterprise to be rectified, after the initial video and the rectification video are acquired, a first picture and a second picture with the same time point are respectively extracted from the initial video and the rectification video. Because the initial picture and the rectification picture are obtained through the camera, and the camera rotates at a fixed position to shoot or performs inspection shooting at a fixed speed and a fixed route, the scenes shot by the initial video and the rectification video at the same time point are the same. If the modification environment corresponding to the modification video is an environment which is actually required to be modified in the enterprise to be modified, the first picture and the second picture extracted from the same time point should have the same scene. Whether the initial video and the rectification video aim at the same scene or not can be detected according to the scenes displayed by the first picture and the second picture, wherein a plurality of same time points can be set for more accurate judgment of the initial video and the rectification video, and the first picture and the second picture are lifted; such as a first picture and a second picture at 1 minute in duration of video, a first picture and a second picture at 2 minutes, a first picture and a second picture at 3 minutes, etc.

It is understood that the modification of the enterprise to be modified needs to be completed within the modification period, and in order to ensure that the modification of the enterprise to be modified is completed within the modification period, the modification period needs to be determined by the modification video after the modification video is acquired. Specifically, before the step of extracting the first picture and the second picture with the same time point from the initial video and the rectification video respectively, the method comprises the following steps:

step a1, extracting a first generation time of the initial video and a second generation time of the correction video, and judging whether the correction video is generated within a preset correction period according to the first generation time and the second generation time;

Understandably, the initial video carries a time characterizing its recording date, which is extracted as the first generation time of the initial video. Likewise, the modified video also carries a time characterizing the date of its recording, which is extracted as the second generation time of the modified video. And judging whether the rectification video is generated within a preset rectification period according to the time difference value represented by the first generation time and the second time, namely judging whether rectification of the enterprise to be rectified is completed within the required rectification period. The preset modification period is determined according to the content of the modification required in the initial video. If the preset modification period is one month, the first generation time is 1 month 1 of a certain year, and the second generation time is 25 months 1 of the same year, the generation of the modification video in the preset modification period can be judged because the time difference between the first generation time and the second generation time is within the preset modification period, namely the modification of the enterprise to be modified is completed within the required modification period.

Step a2, if the first picture and the second picture with the same time point are generated in the preset rectifying period, executing the step of respectively extracting the first picture and the second picture with the same time point from the initial video and the rectifying video;

Further, if it is determined that the modification video is generated within the preset modification period, it is indicated that the generated modification video is effective, and a first picture and a second picture with the same time point can be extracted from the initial video and the modification video respectively, so as to determine whether the initial video and the modification video are shot for the same scene or not through the first picture and the second picture, and whether the modification environment corresponding to the modification video is an environment actually needing modification in the enterprise to be modified, namely, whether the modification environment corresponding to the modification video is effective or not is determined.

And a step a3, outputting prompt information of ineffective rectification if the preset rectification period is not met.

Further, if the video is determined not to be generated within the preset modification period, it is indicated that the modification of the enterprise to be modified is not completed within the required modification period, and the modification of the enterprise to be modified is invalid, so that prompt information of invalid modification is output.

Step S20, judging whether the rectification environment corresponding to the rectification video is effective or not according to the first picture and the second picture;

Furthermore, after the first picture is extracted from the initial video and the second picture with the same time point is extracted from the rectifying video, whether the rectifying environment corresponding to the rectifying video is effective or not can be judged according to the consistency of the scene environment represented in the first picture and the scene environment represented in the second picture, namely, whether the environment shot by the rectifying video is the real environment of the rectifying enterprise after rectifying. It should be noted that, in the video shooting process, a dynamic person may be shot, where the dynamic person has a moving characteristic, so that scenes in a first picture and a second picture with the same time point are not identical, if the first picture includes a certain staff, the second picture does not include the staff, and so on. Therefore, in order to avoid the influence of the dynamic character on the consistency of the scene environments represented by the first picture and the second picture, the present embodiment may set an identification mechanism of the dynamic character. Training an initial model through a large number of dynamic character samples in advance to obtain a dynamic character recognition model, transmitting the extracted first picture and second picture into the dynamic character recognition model, recognizing whether the first picture and the second picture contain dynamic characters or not by the dynamic character recognition model, eliminating the dynamic characters if the dynamic characters are contained, judging whether scene environments represented by the first picture and the second picture after eliminating the dynamic characters are consistent or not, and accordingly judging the consistency of the scene environments by the dynamic characters is avoided. If the environment is consistent with the environment, the environment shot by the rectification video is the real environment of the rectification enterprise after rectification. If the environments are inconsistent, the correction environments corresponding to the correction videos are invalid, the environments shot by the correction videos are not the real environments of the enterprises to be corrected, and false correction conditions can exist.

In addition, consistency judgment can be performed by the similarity between the first picture and the second picture extracted from the same time points, the similarity between the first picture and the second picture extracted from the same time points is detected, and average value operation is performed between the similarities, so that the overall similarity between the first pictures and the second pictures is obtained. If the overall similarity is greater than a preset threshold, the similarity between the first pictures and the second pictures is higher, the scene environments represented by the initial video and the rectification video can be judged to be consistent, the rectification environment corresponding to the rectification video is effective, and the environment shot by the rectification video is the real environment of the rectification enterprise to be rectified. Otherwise, the situation that the scene environments represented by the initial video and the rectifying video are inconsistent is indicated, and the real environment of the enterprise after rectifying is not subjected to rectifying in the environment shot by the rectifying video, and false rectifying conditions possibly exist.

Step S30, if the modification environment is effective, obtaining a modification task list corresponding to the initial video;

Further, if the first picture and the second picture determine that the modification environment is effective, the modification video is shot for the to-be-modified enterprise after modification, and then the modification task list corresponding to the initial video is obtained to represent the content required to be modified by the to-be-modified enterprise. The modification task list can be pre-generated and stored in a storage unit of the server and carries an identifier for representing the initial video, and the server reads the modification task list from the storage unit through the identifier to obtain the modification task list corresponding to the initial video. In addition, the modification task list can also be generated by detecting whether the modification items are contained in the initial video in real time through preset modification items, namely, identifying that the modification items contained in the initial video are generated into the modification task list in real time.

And step S40, analyzing the modification video according to the modification task list, generating an analysis result, and monitoring modification of the enterprise to be modified according to the analysis result.

Furthermore, the content to be rectified contained in the rectification task list is used as the content to be rectified, the rectified video is analyzed according to the content to be rectified, a plurality of rectified pictures containing all the content to be rectified are obtained, and the plurality of rectified pictures are used as analysis results. And then, monitoring the rectification of the enterprise to be rectified by judging whether all the contents to be rectified in the plurality of rectified pictures serving as analysis results are rectified in place. Specifically, according to the analysis result, the step of monitoring the rectification of the enterprise to be rectified includes:

Step S41, obtaining the modification requirement in the modification task list, and judging whether the analysis result is matched with the modification requirement;

Further, setting the corresponding modification requirements for all preset modification items, and setting the modification requirements corresponding to the modification content to be modified in the modification task list to the modification task list no matter whether the modification task list is generated in advance or in real time. The server acquires the modification requirements corresponding to the contents to be modified from the modification task list, searches modified pictures corresponding to the contents to be modified in the plurality of modified pictures of the analysis result, further identifies the modified pictures corresponding to the contents to be modified according to the modification requirements corresponding to the contents to be modified, and judges whether the modified pictures corresponding to the contents to be modified meet the modification requirements, thereby realizing the judgment of whether the analysis result is matched with the modification requirements.

Step S42, if the modification requirement is matched with the modification requirement, finishing modification supervision of the enterprise to be modified;

Further, if it is determined that the modified pictures corresponding to the contents to be modified meet the respective modification requirements, the analysis result is matched with the modification requirements, and the contents to be modified in the enterprise to be modified are modified in place, so that modification supervision of the enterprise to be modified is completed. It should be noted that, until the next supervision period, the original video of the enterprise is obtained again to determine whether it needs to be rectified, if so, the rectification supervision is performed again according to the above manner until the rectification is in place.

And step S43, if the correction request is not matched with the correction request, outputting a prompt message for continuing the correction.

Further, if it is determined that any item of the modified picture corresponding to each piece of the content to be modified does not meet the corresponding modification requirement, at least one piece of content to be modified in the enterprise to be modified is indicated to be not modified in place, and at the moment, prompt information for continuing modification is output. The prompt information comprises content needing to be continuously modified and modification period. And when the modification video is received again, judging whether the received modification video is effective again, if so, acquiring the content which is not modified in place to generate a new modification task list, analyzing the received modification video again according to the new modification task list, and monitoring modification of the enterprise to be modified according to the analysis result until all the content to be modified in place in the enterprise to be modified.

According to the enterprise rectification supervision method, after an initial video before rectification and a rectification video after rectification of an enterprise to be rectified are obtained, a first picture and a second picture with the same time point are respectively extracted from the initial video and the rectification video; judging whether the rectification environment corresponding to the rectification video is effective or not according to the first picture and the second picture; if the modification environment is effective, obtaining a modification task list corresponding to the initial video; and then analyzing the rectification video according to the rectification task list, and generating an analysis result so as to monitor the rectification of the enterprise to be rectified according to the analysis result. Because the first picture and the second picture are derived from the same time point in the initial video and the rectification video, if the rectification video is a real video, supervision environments represented by the first picture and the second picture are the same, so that the effectiveness of the rectification environment judged according to the first picture and the second picture is more real and accurate. On the basis, the rectification video is analyzed according to the rectification task list corresponding to the initial video, so that rectification of an enterprise to be rectified is supervised, frequent review of supervision personnel is avoided, labor cost is saved, and supervision efficiency is improved while authenticity of a rectification environment is ensured.

Further, based on the first embodiment of the enterprise rectification supervision method, a second embodiment of the enterprise rectification supervision method is provided, in the second embodiment, the step of judging whether the rectification environment corresponding to the rectification video is valid according to the first picture and the second picture includes:

step S21, extracting first environment feature points in the first picture, and forming each first environment feature point into a first feature matrix;

According to the embodiment, whether the rectification environment aimed at by the rectification video is effective or not is judged according to the first picture and the second picture, and the higher the similarity of the scene represented between the first picture and the second picture is, the greater the possibility that the initial video and the rectification video aim at the same scene is, namely, the rectification video is effective, otherwise, the rectification video is ineffective. Specifically, a first environmental feature point is extracted from the first picture, wherein the first environmental feature point is feature information of a static object contained in the first picture. Thereafter, feature information of each of the static objects is formed into a first feature matrix to characterize all of the static object features contained in the first picture by the first feature matrix. The step of extracting the first environmental feature points in the first picture and forming each first environmental feature point into a first feature matrix includes:

step S211, identifying static objects in the first picture, and detecting attribute information of each static object, where the attribute information at least includes a name, a color, a size, and coordinates;

Further, training is performed on the initial model by taking pictures of solid objects such as dining tables, chairs, pans, stoves and the like which are commonly used in catering enterprises as training samples in advance to obtain a recognition model for recognizing the solid objects in the environment. The server calls the identification model to identify the static objects in the first picture, and then detects the number of the static objects and attribute information of each static object, wherein the attribute information at least comprises the name, color, size, coordinates and the like of the static objects. Specifically, the name of each static object can be identified by labeling the name of each training sample in the initial model training process; meanwhile, the color of the static object is identified through an OpenCV image processing technology, a preset coordinate system is set, and the size and position coordinates are determined by detecting the position of the static object in the preset coordinate system.

Step S212, attribute information of each static object is formed into attribute sequences, and each attribute sequence is used as a first environment characteristic point in the first picture;

further, attribute information of each static object is formed into attribute sequences of the static objects, and the attribute sequences are used as first environmental feature points in the first picture. If the static object a in the first picture is identified, the name, color, size and coordinates thereof are respectively: table, black, 1 x 1, (20, 30), then the attribute sequences formed for it are [ table, black, 1 x 1, (20, 30) ]. And the attribute sequence is used as a first environmental characteristic point in the first picture to represent various characteristics of a static object table in the first picture.

Step S213, arranging the first environmental feature points to form a first feature matrix.

Further, after the attribute information of each static object is formed into an attribute sequence, that is, each first environmental feature point representing the feature of each static object in the first picture is obtained, each first environmental feature point is formed into a first feature matrix, each row in the first feature matrix represents each feature of one static object, and each column represents the performance of each static object on a feature, for example

Step S22, extracting second environmental characteristic points in the second picture, and forming each second environmental characteristic point into a second characteristic matrix;

Similarly, a second environmental feature point is extracted from the second picture, and the second environmental feature point is feature information of the static object contained in the second picture. Thereafter, feature information of each of the static objects is formed into a second feature matrix to characterize all of the static object features contained in the second picture by the second feature matrix. The method for extracting the second environmental characteristic points to form the second characteristic matrix is the same as the method for extracting the first environmental characteristic points to form the first characteristic matrix. Namely, identifying static objects in the second picture, and detecting attribute information of each static object, wherein the attribute information at least comprises names, colors, sizes and coordinates; forming attribute information of each static object into attribute sequences, and taking each attribute sequence as a second environment characteristic point in a second picture; and arranging the second environmental characteristic points to form a second characteristic matrix. The specific implementation manner of the method is the same as that of forming the first feature matrix, and is not described herein.

Step S23, determining a similarity parameter between the first feature matrix and the second feature matrix, and judging whether a rectifying environment corresponding to the rectified video is effective according to the similarity parameter.

Further, matrix operation is performed between the first feature matrix and the second feature matrix, so as to obtain a similarity parameter for representing the similarity degree between the first feature matrix and the second feature matrix. In addition, a similarity threshold value representing the similarity degree is preset in the server, the obtained similarity parameter is compared with the similarity threshold value, and whether the similarity parameter is larger than the similarity threshold value is judged. If the similarity parameter is larger than the similarity threshold, the more similar the characteristics of the static object in the first picture and the characteristics of the static object in the second picture are represented, the environment aimed at by the rectification is the environment of the enterprise to be rectified, and therefore the rectification environment corresponding to the whole video is judged to be effective. Otherwise, if the similarity parameter is not greater than the similarity threshold, the similarity degree of the features of the static object in the first picture and the features of the static object in the second picture is low, the environment to which the correction is aimed is not the environment of the enterprise to be corrected, and the situation of false correction is possible, so that the correction environment corresponding to the whole video is determined to be invalid.

According to the embodiment, the characteristic points of the respective static objects are extracted from the first picture and the second picture, the first characteristic matrix and the second characteristic matrix are formed to represent the characteristics of all the static objects contained in the first picture and the second picture, and further whether the rectifying environment aimed at by the rectifying video is effective or not is judged through the similarity parameters between the first characteristic matrix and the second characteristic matrix. The first characteristic matrix contains the characteristics of all static objects in the first picture, and the second characteristic matrix contains the characteristics of all static objects in the second picture, so that the effectiveness of the correction environment determined by the similarity parameters of the first characteristic matrix and the second characteristic matrix is more real and accurate.

Further, based on the first embodiment or the second embodiment of the enterprise modification monitoring method, a third embodiment of the enterprise modification monitoring method is provided, and in the third embodiment, the step of obtaining the modification task list corresponding to the initial video includes:

step S31, analyzing the initial video, and extracting frames of the analyzed initial video according to a preset time interval to obtain a plurality of video frames;

In this embodiment, the modification task list corresponding to the initial video is preferably generated by detecting modification items included in the initial video in real time according to preset modification items. Specifically, the initial video is parsed through ffmpeng, and frames of the parsed initial video are extracted according to a preset time interval, so that a plurality of video frames are obtained. The preset time interval represents the time span of frame extraction of the analyzed initial video, if the preset time interval is 5 seconds, frame extraction is carried out again from the 0 th second to the 5 th second of the initial video until the ending time of the initial video. The smaller the preset time interval, the more the number of frames is extracted, and the more accurate the rectification task list is formed.

Step S32, identifying whether the to-be-modified content exists in a plurality of video frames, and if so, generating the to-be-modified content into a modification task list corresponding to the initial video.

Further, according to the preset modification items, the plurality of video stations are identified, and whether the content needing modification exists in each video frame or not, namely whether the content to be modified exists or not is identified. The contents to be modified include, but are not limited to, that the worker does not wear the work cap, that the kitchen has cockroaches and mice, that the kitchen is messy, and the like. If the to-be-modified content exists, adding all to-be-modified content into a template of the modification task list to form a modification task list corresponding to the initial video, and representing the to-be-modified content required by the to-be-modified enterprise through all to-be-modified content in the modification task list.

Further, the step of analyzing the modification video according to the modification task sheet to generate an analysis result includes:

Step S44, analyzing and extracting frames of the rectifying video according to the preset time interval to obtain a plurality of rectified pictures;

Further, for the rectifying video after rectifying of the enterprise to be rectified, analyzing and extracting frames through ffmpeng according to a preset time interval to obtain a plurality of rectified pictures. And reflecting the modification condition of the enterprise to be modified through a plurality of modified pictures, whether all contents needing modification are modified in place or not, and the like.

Step S45, judging whether the plurality of rectified pictures contain all the contents to be rectified in the rectified task list, and if so, generating the plurality of rectified pictures into the analysis result;

It is understandable that, in order to make the rectified picture completely reflect the rectified situation of the enterprise to be rectified, it is necessary to ensure that the plurality of rectified pictures include all the contents to be rectified in the rectified task list. If the to-be-modified content in the modification task list comprises that the worker does not wear the working cap and the kitchen has cockroaches, the plurality of modified pictures of the frame drawing at least comprise one worker and at least comprise the kitchen environment. Therefore, after the plurality of rectified pictures are obtained, the plurality of rectified pictures are judged according to the rectified task list, and whether the plurality of rectified pictures contain all the contents to be rectified in the rectified task list is judged. If the content to be modified is included, the frame extraction of the modified video is described, and the obtained multiple modified pictures meet the requirements and are generated into analysis results.

Step S46, if not all the contents to be rectified are included, adjusting the preset time interval, and executing the step of analyzing and extracting frames for the rectified video according to the adjusted preset time interval.

Further, if it is determined that the plurality of rectified pictures do not include all to-be-rectified contents in the rectification task list, that is, the plurality of rectified pictures lack any to-be-rectified contents in the to-be-rectified task list, the preset time interval is adjusted, and the preset time interval is adjusted to be smaller in time span. And then, based on the adjusted preset time span, re-framing the rectification video to extract more rectification pictures until all the contents to be rectified in the rectification task list are contained, so as to generate an analysis result. It should be noted that, if the plurality of rectified pictures include a plurality of repeated contents to be rectified, the extracted rectified pictures are too many, and the preset time interval can be adjusted to a larger time span to extract frames, so as to avoid generating the rectified pictures with too many repetitions into an analysis result, which results in increasing the processing data volume of the analysis result by the server, and reducing the processing efficiency.

According to the embodiment, an initial video is analyzed and frame-extracted to form a rectification task list according to a preset time interval, the rectification video is analyzed and frame-extracted to form rectified pictures according to the rectification task list, and analysis results are formed by the rectified pictures to reflect rectification conditions of enterprises to be rectified. The extracted rectified pictures comprise all to-be-rectified contents in the rectified task list through setting the preset time interval to be adjustable, so that the to-be-rectified enterprise rectification is accurately monitored according to the analysis result formed by the rectified pictures.

Further, the invention also provides an enterprise rectification supervision device.

Referring to fig. 3, fig. 3 is a schematic functional block diagram of a first embodiment of the enterprise rectification and supervision apparatus according to the present invention. The enterprise rectification supervision device comprises:

the extraction module 10 is used for obtaining an initial video before modification and a modified video after modification of the enterprise to be modified, and extracting a first picture and a second picture with the same time point from the initial video and the modified video respectively;

The judging module 20 is configured to judge whether a modification environment corresponding to the modification video is valid according to the first picture and the second picture;

The obtaining module 30 is configured to obtain a modification task list corresponding to the initial video if the modification environment is valid;

and the supervision module 40 is configured to parse the modification video according to the modification task sheet, generate a parsing result, and supervise modification of the enterprise to be modified according to the parsing result.

In the enterprise modification supervision apparatus of the present embodiment, after an initial video before modification and a modified video after modification of an enterprise to be modified are obtained, the extraction module 10 extracts a first picture and a second picture with the same time point from the initial video and the modified video, respectively; judging whether the rectification environment corresponding to the rectification video is effective or not by the judging module 20 according to the first picture and the second picture; if the modification environment is valid, the obtaining module 30 obtains a modification task list corresponding to the initial video; and then, the supervision module 40 analyzes the rectification video according to the rectification task list to generate an analysis result so as to supervise rectification of the enterprise to be rectified according to the analysis result. Because the first picture and the second picture are derived from the same time point in the initial video and the rectification video, if the rectification video is a real video, supervision environments represented by the first picture and the second picture are the same, so that the effectiveness of the rectification environment judged according to the first picture and the second picture is more real and accurate. On the basis, the rectification video is analyzed according to the rectification task list corresponding to the initial video, so that rectification of an enterprise to be rectified is supervised, frequent review of supervision personnel is avoided, labor cost is saved, and supervision efficiency is improved while authenticity of a rectification environment is ensured.

Further, the judging module 20 includes:

a first extraction unit, configured to extract first environmental feature points in the first picture, and form each of the first environmental feature points as a first feature matrix;

a second extraction unit, configured to extract second environmental feature points in the second picture, and form each of the second environmental feature points as a second feature matrix;

And the judging unit is used for determining the similarity parameter between the first feature matrix and the second feature matrix and judging whether the rectifying environment corresponding to the rectified video is effective or not according to the similarity parameter.

Further, the first extraction unit is further configured to include:

the step of extracting the first environmental feature points in the first picture and forming each first environmental feature point into a first feature matrix includes:

Further, the obtaining module 30 further includes:

The analysis unit is used for analyzing the initial video, and extracting frames of the analyzed initial video according to a preset time interval to obtain a plurality of video frames;

the identification unit is used for identifying whether the to-be-modified content exists in the plurality of video frames, and if the to-be-modified content exists, the to-be-modified content is generated into a modification task list corresponding to the initial video.

Further, the supervision module 40 further includes:

the frame extraction unit is used for analyzing and extracting frames of the rectification video according to the preset time interval to obtain a plurality of rectified pictures;

The generation unit is used for judging whether the plurality of rectified pictures contain all the contents to be rectified in the rectified task list, and if so, generating the plurality of rectified pictures into the analysis result;

And the adjusting unit is used for adjusting the preset time interval if all the contents to be rectified are not contained, and executing the step of analyzing and extracting frames for the rectified video according to the adjusted preset time interval.

Further, the supervision module 40 further includes:

The acquisition unit is used for acquiring the modification requirement in the modification task list and judging whether the analysis result is matched with the modification requirement or not;

The completion unit is used for completing the rectification supervision of the enterprise to be rectified if the completion unit is matched with the rectification requirement;

And the first output unit is used for outputting prompt information for continuing the rectification if the first output unit is not matched with the rectification requirement.

Further, the extraction module 10 includes:

The third extraction unit is used for extracting the first generation time of the initial video and the second generation time of the rectifying video, and judging whether the rectifying video is generated within a preset rectifying period according to the first generation time and the second generation time;

the execution unit is used for executing the step of respectively extracting the first picture and the second picture with the same time point from the initial video and the rectifying video if the first picture and the second picture are generated within the preset rectifying period;

and the second output unit is used for outputting prompt information that the rectification is invalid if the rectification is not within the preset rectification period.

The specific implementation of the enterprise modification supervision device is basically the same as the above embodiments of the enterprise modification supervision method, and will not be described in detail here.

In addition, the embodiment of the invention also provides a computer readable storage medium.

The computer readable storage medium has stored thereon an enterprise modification supervisory program which when executed by the processor implements the steps of the enterprise modification supervisory method as described above.

The specific implementation of the computer readable storage medium of the present invention is substantially the same as the above embodiments of the enterprise rectification and supervision method, and will not be described herein.

It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or system. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or system that comprises the element.

The foregoing embodiment numbers of the present invention are merely for the purpose of description, and do not represent the advantages or disadvantages of the embodiments.

From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a computer readable storage medium (e.g. ROM/RAM, magnetic disk, optical disk) as described above, comprising instructions for causing a terminal device (which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to perform the method according to the embodiments of the present invention.

The foregoing description is only of the preferred embodiments of the present invention, and is not intended to limit the scope of the invention, but rather is intended to cover any equivalents of the structures or equivalent processes disclosed herein or in the alternative, which may be employed directly or indirectly in other related arts.

Claims

1. The enterprise rectification supervision method is characterized by comprising the following steps of:

Analyzing the rectification video according to the rectification task list, generating an analysis result, and supervising rectification of the enterprise to be rectified according to the analysis result;

The step of judging whether the rectification environment corresponding to the rectification video is valid according to the first picture and the second picture comprises the following steps:

determining a similarity parameter between the first feature matrix and the second feature matrix, and judging whether a rectification environment corresponding to the rectified video is effective or not according to the similarity parameter;

training the initial model according to the dynamic character sample to obtain a dynamic character recognition model, extracting a first picture and a second picture, transmitting the first picture and the second picture to the dynamic character recognition model, and recognizing whether the first picture and the second picture contain dynamic characters according to the dynamic character recognition model;

If the dynamic figures are included, eliminating the dynamic figures, and judging whether scene environments represented by the first picture and the second picture after eliminating the dynamic figures are consistent;

the step of obtaining the rectification task list corresponding to the initial video comprises the following steps:

Identifying whether the to-be-modified content exists in the plurality of video frames, and if the to-be-modified content exists, generating the to-be-modified content into a modification task list corresponding to the initial video;

the step of analyzing the rectification video according to the rectification task list and generating an analysis result comprises the following steps:

2. The method of enterprise rectification supervision as claimed in claim 1, wherein the step of extracting the first environmental feature points in the first picture and forming each of the first environmental feature points into a first feature matrix comprises:

3. The method for monitoring and managing the modification of an enterprise according to any one of claims 1 to 2, wherein the step of monitoring and managing the modification of the enterprise to be modified according to the analysis result includes:

4. The method for managing and managing rectification of an enterprise as set forth in any one of claims 1 to 2, wherein the step of extracting the first picture and the second picture having the same time point from the initial video and the rectification video, respectively, is preceded by the steps of:

5. An enterprise rectification supervision apparatus, characterized in that the enterprise rectification supervision apparatus comprises:

The monitoring module is used for analyzing the modification video according to the modification task list, generating an analysis result and monitoring modification of the enterprise to be modified according to the analysis result;

The judging module comprises:

The judging unit is used for determining similarity parameters between the first feature matrix and the second feature matrix and judging whether the rectifying environment corresponding to the rectified video is effective or not according to the similarity parameters; the method is also used for training the initial model according to the dynamic character sample to obtain a dynamic character recognition model, extracting a first picture and a second picture, transmitting the first picture and the second picture to the dynamic character recognition model, and recognizing whether the first picture and the second picture contain dynamic characters or not according to the dynamic character recognition model; if the dynamic figures are included, eliminating the dynamic figures, and judging whether scene environments represented by the first picture and the second picture after eliminating the dynamic figures are consistent;

the acquisition module comprises:

The identification unit is used for identifying whether the to-be-modified content exists in the plurality of video frames, and if so, generating the to-be-modified content into a modification task list corresponding to the initial video;

the supervision module comprises:

6. An enterprise modification supervisory device comprising a memory, a processor, and an enterprise modification supervisory program stored on the memory and executable on the processor, the enterprise modification supervisory program when executed by the processor implementing the steps of the enterprise modification supervisory method according to any of claims 1-4.

7. A computer readable storage medium, wherein an enterprise modification supervisory program is stored on the computer readable storage medium, which when executed by a processor implements the steps of the enterprise modification supervisory method according to any one of claims 1-4.