An engineering construction safety practical training method and system based on big data

Abstract

The present application relates to the field of safety training, especially to a kind of engineering construction safety training method and system based on big data, method includes: the answering number of target personnel is equal to optimization response number, according to frame selection duration representation value and frame selection duration fluctuation value determine the preset analysis condition corresponding to target personnel;Under the first preset analysis condition, according to the point position aggregation similarity value corresponding to the point position distribution diagram of the test question of target personnel and point position aggregation representation value determine that the compensation mode is point position array compensation or frame parameter compensation;Frame parameter compensation, according to the number ratio of first anchor frame and second anchor frame determine that frame parameter compensation strategy is compensated based on frame neighborhood parameter or compensated based on frame area;Under the second preset analysis condition, the optimization response number is adjusted to increase;The present application dynamically adjusts test question to improve safety training efficiency.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of safety training, in particular to an engineering construction safety training method and system based on big data. BACKGROUND

[0002] In the field of engineering construction, safety production is the core guarantee for the development of the industry, and safety training is an important means to improve the safety awareness and operation standards of employees. With the development of big data and artificial intelligence technology, unsafe behavior recognition training based on site images has become one of the important directions of safety training. By analyzing unsafe behaviors (such as not wearing a safety helmet, operating machinery in violation of regulations, etc.) in construction site images, students can master safety standards. However, existing safety training usually generates test questions based on static question banks or fixed scenes, and cannot dynamically adjust the difficulty or content of test questions according to the actual training performance of students, resulting in low training efficiency.

[0003] Chinese patent publication No. CN113807990A discloses a construction site safety training method, system, device and storage medium. The method includes: obtaining a question bank for safety training and obtaining current construction information and construction personnel information to be safety trained, generating a personal tag for the construction personnel to be safety trained, matching the personal tag with the question tag in the question bank, and generating a safety training answer sheet matched with the current construction personnel to be safety trained. In the above technical solution, if the number of wrong questions of the current construction personnel to be safety trained the previous day is less than or equal to the preset threshold, all wrong questions of the previous day are inserted into the safety training answer sheet of the current day; if the number of wrong questions of the current construction personnel to be safety trained the previous day is greater than the preset threshold, the number of wrong questions within the preset threshold range is inserted into the current safety training answer sheet, and for the number of wrong questions exceeding the preset threshold range, a predetermined number of wrong questions are selected and inserted into the safety training answer sheet of the current day, and the predetermined number is less than or equal to the preset threshold. It can be seen that the above technical solution can only optimize the historical wrong questions to the new answer sheet, and the optimization method is single and difficult to meet the training needs of the actual scene. SUMMARY

[0004] Therefore, the present application provides an engineering construction safety training method and system based on big data to overcome the problem that the prior art can only use repeated training of wrong questions as an optimization method, the optimization method is single, the test question content cannot be dynamically adjusted, and the training efficiency is low.

[0005] To achieve the above purpose, the present application provides an engineering construction safety training method based on big data, comprising:

[0006] When the number of answers of the target personnel is equal to the number of optimized responses, the frame selection duration value and the frame selection duration fluctuation value are used to determine the preset analysis condition corresponding to the target personnel;

[0007] Under the first preset analysis condition, a point array compensation or a frame parameter compensation is determined according to the point aggregation similarity value and the point aggregation characteristic value corresponding to the test question point distribution map of the target person.

[0008] Under the frame parameter compensation, a frame parameter compensation strategy is determined according to the number ratio of the first anchor frame and the second anchor frame.

[0009] Under the second preset analysis condition, the optimization response times are adjusted to increase.

[0010] Further, under the first preset analysis condition, a test question point distribution map corresponding to each point touch delay record is obtained.

[0011] The point aggregation similarity value and the point aggregation characteristic value corresponding to the test question point distribution map are obtained.

[0012] If the point aggregation similarity value is greater than a preset point aggregation similarity value and the point aggregation characteristic value is less than or equal to a preset point aggregation characteristic value, the compensation mode is point array compensation.

[0013] If the point aggregation similarity value is less than or equal to a preset point aggregation similarity value or the point aggregation characteristic value is greater than a preset point aggregation characteristic value, the compensation mode is frame parameter compensation.

[0014] Further, when the compensation mode is point array compensation, the cloud storage test questions are extracted according to the point array deviation degree in descending order to compensate for the required number of cloud storage test questions.

[0015] The point array deviation degree is determined according to the array area difference value and the array distance compensation value.

[0016] Further, when the compensation mode is frame parameter compensation, a frame parameter compensation strategy is determined according to the number ratio of the first anchor frame and the second anchor frame.

[0017] If the number ratio is less than a preset number ratio, the frame parameter compensation strategy is based on the frame neighborhood parameter compensation.

[0018] If the number ratio is greater than or equal to a preset number ratio, the frame parameter compensation strategy is based on the frame area compensation.

[0019] Further, the anchor frame category corresponding to each anchor frame is determined according to the frame selection deviation value of the anchor frame, and the anchor frame category includes a first anchor frame with a frame selection deviation value greater than a permitted frame selection deviation value and a second anchor frame with a frame selection deviation value less than or equal to a permitted frame selection deviation value.

[0020] Further, the storage module corresponding to the extraction of the cloud storage test question is determined according to the optimization update times and the engineering diversity of the storage module.

[0021] Further, under the second preset analysis condition, the optimization response number is increased according to the frame selection duration representation difference;

[0022] The increase amount of the optimization response number is positively correlated with the frame selection duration representation difference.

[0023] Further, the preset analysis condition is determined by the frame selection duration representation value and the frame selection duration fluctuation value;

[0024] The first preset analysis condition is that the frame selection duration representation value is less than or equal to the preset frame selection duration representation value and the frame selection duration fluctuation value is greater than the preset frame selection duration fluctuation value;

[0025] The second preset analysis condition is that the frame selection duration representation value is greater than the preset frame selection duration representation value or the frame selection duration fluctuation value is less than or equal to the preset frame selection duration fluctuation value.

[0026] In addition, the present application also provides a system applying the engineering construction safety practical training method based on big data, characterized by comprising:

[0027] The condition analysis unit is used to determine the preset analysis condition corresponding to the target personnel according to the frame selection duration representation value and the frame selection duration fluctuation value when the number of answers of the target personnel is equal to the optimization response number;

[0028] The optimization compensation unit is connected with the condition analysis unit and is used to determine that the compensation mode is point array compensation or frame parameter compensation according to the point position aggregation similarity value and the point position aggregation representation value corresponding to the point touch delay record of the target personnel under the first preset analysis condition, and when the frame parameter compensation is performed, the frame parameter compensation strategy is determined to be compensation based on frame neighborhood parameters or compensation based on frame area according to the number ratio of the first anchor frame and the second anchor frame;

[0029] The category analysis unit is connected with the optimization compensation unit and is used to determine the anchor frame category corresponding to the anchor frame according to the frame selection deviation value of the anchor frame when the frame parameter compensation is performed by the optimization compensation unit;

[0030] The simplification processing unit is connected with the condition analysis unit and is used to increase the optimization response number according to the frame selection duration representation difference under the second preset analysis condition;

[0031] The cloud data storage unit is connected with the condition analysis unit, the optimization compensation unit, the category analysis unit and the simplification processing unit, and the cloud data storage unit comprises a plurality of storage modules used to store test questions in the cloud.

[0032] Compared with the prior art, the beneficial effects of the present application are that, in the technical scheme of the present application, the preset analysis condition corresponding to the target person is determined according to the box selection duration characteristic value and the box selection duration fluctuation value, the box selection duration characteristic value and the box selection duration fluctuation value reflect the duration of the target person answering the question, and according to the different preset analysis conditions, the point aggregation similarity value and the point aggregation characteristic value corresponding to the test question point position distribution map of the target person are selected to determine the compensation mode, or the optimization response number is increased, so that the optimization analysis is more accurate, and the problem of poor question optimization efficiency caused by a single optimization mode and difficult to meet the training requirements is avoided.

[0033] Further, in the technical scheme of the present application, under the first preset analysis condition, the compensation mode is determined according to the point aggregation similarity value and the point aggregation characteristic value corresponding to the test question point position distribution map, and the compensation mode is point array compensation or frame parameter compensation. Compared with the prior art of single question optimization only according to historical mistakes, the present application analyzes the question image itself and selects different compensation modes, so that the compensation mode is more in line with the actual application scenario, thereby improving the question optimization efficiency and further improving the subsequent training effect.

[0034] Further, in the technical scheme of the present application, the storage module corresponding to the process of extracting the cloud storage test question is determined according to the optimal update number of the storage module and the engineering diversity, so that the selection of the test question is more in line with the demand of the actual scene, and the problem of poor test question selection effect caused by manual test question selection or random test question selection is avoided. BRIEF DESCRIPTION OF DRAWINGS

[0035] Figure 1 It is a schematic diagram of the engineering construction safety practical training method based on big data of the present application;

[0036] Figure 2 It is a flowchart of determining the compensation mode of the present application;

[0037] Figure 3 It is a flowchart of determining the frame parameter compensation strategy of the present application;

[0038] Figure 4 It is a unit connection diagram of the engineering construction safety practical training system based on big data of the present application. DETAILED DESCRIPTION

[0039] In order to make the purpose and advantages of the present application more clear and obvious, the present application will be further described below in combination with examples; it should be understood that the specific examples described herein are only used to explain the present application, and do not limit the present application.

[0040] The preferred embodiments of the present application will be described below with reference to the accompanying drawings. Those skilled in the art will understand that the embodiments are only used to explain the technical principles of the present application and are not intended to limit the protection scope of the present application.

[0041] It should be noted that, in the description of the present application, the terms indicating the direction or positional relationship of "up", "down", "left", "right", "inner", "outer" and the like are based on the direction or positional relationship shown in the drawings, which is only for the convenience of description, and does not indicate or imply that the device or element must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the present application.

[0042] In addition, it should also be noted that, in the description of the present application, unless otherwise explicitly specified and limited, the terms "mounting", "connecting", "connecting" should be understood broadly, for example, it can be fixedly connected, or it can be detachably connected, or integrally connected; it can be mechanically connected, or it can be electrically connected; it can be directly connected, or it can be indirectly connected through an intermediate medium; it can be the communication inside two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0043] Please refer to Figures 1 to 3 The present application provides an engineering construction safety training method based on big data, which comprises:

[0044] When the number of answers of the target person is equal to the optimized response number, the preset analysis condition corresponding to the target person is determined according to the frame selection duration value and the frame selection duration fluctuation value;

[0045] Under the first preset analysis condition, the compensation mode is point array compensation or frame parameter compensation according to the point position aggregation similarity value and the point position aggregation value corresponding to the point position distribution map of the target person;

[0046] When the frame parameter compensation is performed, the frame parameter compensation strategy is determined according to the number ratio of the first anchor frame and the second anchor frame, which is compensation based on frame neighborhood parameters or compensation based on frame area;

[0047] Under the second preset analysis condition, the optimized response number is increased.

[0048] The application is applied to safety training in engineering construction, the target personnel is the personnel for practical training, the target personnel obtains the test paper image through a display device, and when the target personnel replies to the test question image, the area in the test question image that is considered to have dangerous behavior is framed on the display device, the framed area is recorded as an anchor frame, the cloud storage test question is the construction site image used in the safety training uploaded by each engineering unit, and each cloud storage test question corresponds to the shape of the preset area, the position of the area center point of the preset area and the area of the preset area, the preset area is the area with dangerous behavior, the preset area and the corresponding preset area shape, the area center point of the preset area and the area of the preset area are all set by the user, the user can collect a large number of construction site images in advance and select the construction site images with dangerous behavior, and set the preset area correspondingly, which is easily understood by those skilled in the art and will not be described here; the test question image and the cloud storage test question both have the preset area, the test question point distribution map is the test question image after labeling (framing labeling) of each preset area, the number of answers of the target personnel is the number of test question image processing of the target personnel after the determination of the last preset analysis condition is completed, processing a test question image is recorded as an answer number, wherein, when the engineering unit uploads the construction site image, the construction site image is preferentially distributed to the storage module with the least cloud storage test question, if the number of the storage module with the least cloud storage test question is greater than 1, one of the storage modules with the least cloud storage test question is randomly selected for uploading.

[0049] The application has a history record, the history record at least includes point aggregation similarity value, point aggregation characteristic value, array area difference value, array distance compensation value, frame deviation value, frame duration characteristic value and frame duration fluctuation value in a historical process, and the history record also corresponds to a qualified mark, the qualified mark records whether the history record meets the user's demand, wherein, whether the history record meets the user's demand is determined according to the self-set index (for example, the correct rate of answering) of the training effect, which is mastered by those skilled in the art and will not be described here.

[0050] Specifically, under the first preset analysis condition, the test question point distribution map corresponding to each point touch delay record is obtained;

[0051] The point aggregation similarity value and the point aggregation characteristic value corresponding to the test question point distribution map are obtained;

[0052] If the point aggregation similarity value is greater than the preset point aggregation similarity value and the point aggregation characteristic value is less than or equal to the preset point aggregation characteristic value, the compensation mode is point array compensation;

[0053] If the point aggregation similarity value is less than or equal to the preset point aggregation similarity value or the point aggregation characteristic value is greater than the preset point aggregation characteristic value, the compensation mode is frame parameter compensation.

[0054] The point touch delay record is a record of an answer in which a time length value of a lower level of a frame selection in an optimized response frequency of a target person in a recent record is less than a frame selection time length value, and the single answer record includes a test image processed by the target person, a frame selection time length corresponding to the test image, a test point distribution map, and a frame selection deviation value of each anchor frame.

[0055] The confirmation method of the point aggregation value is to detect an aggregation range area of each test point distribution map corresponding to the point touch delay record, and the average value of the aggregation range area is recorded as the point aggregation value. For a single test point distribution map, the aggregation range is the smallest rectangle that can include all the preset areas in the test point distribution map. The point aggregation similarity value is the absolute value of the difference between the maximum and minimum values of the aggregation range area corresponding to each test point distribution map.

[0056] The values of the preset point aggregation similarity value and the preset point aggregation value can be determined by the user according to the actual application scenario. It can be understood that when the point aggregation similarity value is greater than the preset point aggregation similarity value and the point aggregation value is less than or equal to the preset point aggregation value, it reflects that the distribution of the preset areas corresponding to each test image in the recent reply of the target person is dense and the similarity degree of the distribution area of the preset areas between the test images is large. Therefore, the compensation method is point array compensation to improve the distribution diversity of the preset areas of the subsequent test images, thereby improving the training effect and avoiding the problem of poor training effect caused by the approach of the questions. Therefore, the greater the user's demand for the distribution diversity of the preset areas, the greater the value of the preset point aggregation similarity value and the smaller the value of the preset point aggregation value. A value determination method is provided to extract the point aggregation similarity value and the point aggregation value corresponding to the historical records that meet the user's demand, remove the outliers in the point aggregation similarity value and the point aggregation value, respectively, and record the average values of the point aggregation similarity value and the point aggregation value after removing the outliers as the preset point aggregation similarity value and the preset point aggregation value, respectively. The method of removing outliers includes but is not limited to the 3σ criterion method or the IQR method.

[0057] Specifically, when the compensation method is point array compensation, the cloud storage test questions are extracted in descending order of point array deviation degree to compensate for the required number of cloud storage test questions.

[0058] The point array deviation degree is determined according to the array area difference value and the array distance compensation value.

[0059] For a single cloud storage test question, the corresponding point array deviation degree = array area difference value / preset array area difference value + array distance compensation value / preset array distance compensation value, the array area difference value = the aggregation range area of the cloud storage test question - the point aggregation representation value, and the array distance compensation value is determined by obtaining the minimum distance between each preset region of the cloud storage test question, and the minimum value of the minimum distance is recorded as the array distance compensation value.

[0060] The values of the preset array area difference value and the preset array distance compensation value can be set by the user according to the actual application scenario. It can be understood that the higher the user's requirement for the compensation effect of the cloud storage test question, the larger the values of the preset array area difference value and the preset array distance compensation value. A value setting method is provided, the array area difference value and the array distance compensation value corresponding to the historical records meeting the user's demand are extracted, the abnormal values in the array area difference value and the array distance compensation value are removed respectively, and the average values of the array area difference value and the array distance compensation value after removing the abnormal values are recorded as the preset array area difference value and the preset array distance compensation value.

[0061] The value of the demand quantity is set by the user. It can be understood that the greater the user's demand for the compensation amount of the cloud storage test question, the greater the demand quantity, and the more abundant the number of subsequent test question images. A specific implementation value is provided, and the demand quantity = 10.

[0062] Specifically, when the compensation method is frame parameter compensation, the frame parameter compensation strategy is determined according to the number ratio of the first anchor frame and the second anchor frame;

[0063] If the number ratio is less than the preset number ratio, the frame parameter compensation strategy is compensation based on the frame neighborhood parameter;

[0064] If the number ratio is greater than or equal to the preset number ratio, the frame parameter compensation strategy is compensation based on the frame area.

[0065] The number ratio = the number of first anchor frames in all point touch delay records / the number of second anchor frames. As a special case, if the number of first anchor frames = 0, there is no need to analyze the number ratio, and the frame parameter compensation strategy is fixed to be compensation based on the frame neighborhood parameter. If the number of second anchor frames = 0, there is no need to analyze the number ratio, and the frame parameter compensation strategy is compensation based on the frame area. It can be understood that the number of first anchor frames and the number of second anchor frames cannot be 0 at the same time.

[0066] The preset number ratio value can be set by the user according to the actual situation. It can be understood that the greater the training demand of the user for the accuracy of the anchor frame of the target person, the smaller the preset number ratio. A value is provided, and the preset number ratio = 20%.

[0067] Specifically, the anchor frame category corresponding to each anchor frame is determined according to the frame selection deviation value of the anchor frame. The anchor frame category includes a first anchor frame with a frame selection deviation value greater than an allowed frame selection deviation value and a second anchor frame with a frame selection deviation value less than or equal to the allowed frame selection deviation value.

[0068] The frame selection deviation value of the anchor frame is obtained by obtaining the overlapping preset area of the anchor frame, calculating the area of the anchor frame not in the overlapping preset area, and recording it as the frame selection deviation value. If an anchor frame does not have an overlapping preset area, the anchor frame does not have a frame selection deviation value, and the anchor frame is an invalid anchor frame. It can be understood that the overlapping preset area includes partial or complete overlapping, without limitation.

[0069] The value of the allowed frame selection deviation value is set by the user. It can be understood that the greater the user's acceptance of the frame selection deviation value, the greater the value of the allowed frame selection deviation value. A value of the allowed frame selection deviation value is provided, and the average value of the frame selection deviation value after removing outliers is recorded as the allowed frame selection deviation value.

[0070] Specifically, the frame neighborhood parameter = adjacent object distribution value, and the adjacent object distribution value is the total number of buildings in the relevant range of the preset area.

[0071] For a single preset area, the corresponding relevant range is a circular area, the area center point of the relevant range coincides with the area center point of the preset area, and the entire preset area is within the relevant range. The area of the relevant range is smaller than the area of the test image in which the preset area is located. The specific area of the relevant range is set by the user. The greater the user's demand for the analysis range of the frame neighborhood parameter of the preset area, the greater the value of the relevant range. A value is provided, and the area of the relevant range is equal to twice the area of the smallest circular area that can include the preset area.

[0072] When compensation is based on the frame neighborhood parameter, the cloud storage test questions are extracted in descending order of the frame neighborhood parameter.

[0073] When compensation is based on the frame area, the cloud storage test questions are extracted in descending order of the frame area for compensation. The frame area corresponding to the cloud storage test question is the average value of the area of the preset area corresponding to the cloud storage test question.

[0074] The extracted cloud storage test question which is compensated is denoted as a compensation test question, and is supplemented to a test question image in next target personnel practical training.

[0075] Specifically, in the process of extracting the cloud storage test question, the corresponding storage module is determined according to the preferred update frequency of the storage module and the engineering diversity.

[0076] In the extraction of the cloud storage test question, the extraction of the cloud storage test question is performed only from a single storage module, wherein the preferred update frequency of each storage module is calculated, the first three storage modules corresponding to the preferred update frequency are selected in descending order, and the storage module with the largest engineering diversity among the three storage modules is selected as the storage module for extracting the cloud storage test question.

[0077] The preferred update frequency is the upload frequency of each engineering unit data corresponding to the storage module, and the data upload frequency is the number of times of uploading the construction site image. The number of construction site images uploaded in a single data upload is not limited. The engineering diversity is the total number of engineering units that have performed data upload corresponding to the storage module.

[0078] Specifically, under the second preset analysis condition, the optimization response frequency is increased by adjusting according to the frame selection duration representation difference;

[0079] The increase amount of the optimization response frequency has a positive correlation with the frame selection duration representation difference.

[0080] The frame selection duration representation difference = the frame selection duration representation value corresponding to the target personnel - the preset frame selection duration representation value.

[0081] The increased optimization response frequency after the adjustment = the optimization response frequency before the adjustment + the increase amount of the optimization response frequency, the increase amount of the optimization response frequency = k x the frame selection duration representation difference, k is a conversion coefficient, and the optimization response frequency before the adjustment and the value of k are set by the user. It can be understood that the larger the frame selection duration representation difference is, the worse the test answering effect corresponding to the target personnel is. Therefore, the larger the optimization response frequency is, the less the data calculation resource consumption caused by frequent test question optimization analysis is. In addition, the smaller the optimization response frequency is, the larger the test question optimization frequency is. Therefore, the greater the user's acceptance of data calculation resource consumption is, the smaller the value of k is, the larger the optimization response frequency is, and the increase amount of the optimization response frequency is an integer taken upward. A value of k = 0.4 is provided, and the optimization response frequency before the adjustment = 10.

[0082] Specifically, the preset analysis condition is determined by the frame selection duration representation value and the frame selection duration fluctuation value.

[0083] The first preset analysis condition is that the frame selection duration representation value is less than or equal to the preset frame selection duration representation value and the frame selection duration fluctuation value is greater than the preset frame selection duration fluctuation value.

[0084] The second preset analysis condition is that the frame selection duration characteristic value is greater than a preset frame selection duration characteristic value or the frame selection duration fluctuation value is less than or equal to a preset frame selection duration fluctuation value.

[0085] The frame selection duration characteristic value is an average value of lower-level frame selection duration characteristic values of the answer record of the latest optimized response number of the target person, and for a single answer record, the confirmation manner of the corresponding lower-level frame selection duration characteristic value is to detect the interval duration between adjacent anchor frames in the construction time sequence corresponding to the answer record, the interval duration is the duration between the construction time points corresponding to the two adjacent anchor frames in the construction time sequence, and the average value of the interval duration is recorded as the lower-level frame selection duration characteristic value of the answer record.

[0086] The frame selection duration fluctuation value is recorded as S, and the calculation manner of the frame selection duration fluctuation value is:

[0087]

[0088] Wherein, Lu is the lower-level frame selection duration characteristic value corresponding to the u-th answer record in the answer record of the latest optimized response number of the target person, L0 is the frame selection duration characteristic value, P=optimized response number, u=1, 2, 3, …, P, and the value sequence of u corresponding to the answer record of the latest optimized response number of the target person does not need to be specifically limited, and has no influence on the calculation result.

[0089] The values of the preset frame selection duration characteristic value and the preset frame selection duration fluctuation value are set by the user, and it can be understood that the higher the user's judgment requirement for the fluency of the answer of the target person, the smaller the preset frame selection duration characteristic value and the preset frame selection duration fluctuation value, a value selection method is provided, the frame selection duration characteristic value and the frame selection duration fluctuation value corresponding to the historical record meeting the user's demand are extracted, the abnormal values in the frame selection duration characteristic value and the frame selection duration fluctuation value are removed respectively, and the average values of the frame selection duration characteristic value and the frame selection duration fluctuation value after removing the abnormal values are recorded as the preset frame selection duration characteristic value and the preset frame selection duration fluctuation value respectively.

[0090] Please refer to Figure 4 It is a unit connection diagram of the engineering construction safety training system based on big data, and the application further provides a system applying the engineering construction safety training method based on big data, characterized by comprising:

[0091] The condition analysis unit is used to determine the preset analysis condition corresponding to the target person according to the frame selection duration characteristic value and the frame selection duration fluctuation value when the answer number of the target person is equal to the optimized response number.

[0092] An optimization compensation unit, connected with the condition analysis unit, is configured to determine the compensation mode as point array compensation or frame parameter compensation according to the point aggregation similarity value and the point aggregation representation value corresponding to each point touch delay record of the target person under the first preset analysis condition, and determine the frame parameter compensation strategy as compensation based on frame neighborhood parameters or compensation based on frame area according to the number ratio of the first anchor frame and the second anchor frame when the frame parameter compensation is performed;

[0093] A category analysis unit, connected with the optimization compensation unit, is configured to determine the anchor frame category corresponding to the anchor frame according to the frame selection deviation value of the anchor frame when the optimization compensation unit performs the frame parameter compensation.

[0094] A simplification processing unit, connected with the condition analysis unit, is configured to increase the optimization response times according to the frame selection duration representation difference under the second preset analysis condition.

[0095] A cloud data storage unit, connected with the condition analysis unit, the optimization compensation unit, the category analysis unit and the simplification processing unit, includes a plurality of storage modules configured to store the test questions in the cloud.

[0096] So far, the technical solutions of the present application have been described in combination with the preferred embodiments shown in the drawings, but it is easy for those skilled in the art to understand that the protection scope of the present application is obviously not limited to these specific embodiments. Those skilled in the art can make equivalent changes or replacements to the related technical features without departing from the principles of the present application, and the technical solutions after the changes or replacements will fall within the protection scope of the present application.

[0097] The above description is only the preferred embodiments of the present application and is not used to limit the present application; for those skilled in the art, the present application can have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A big data-based engineering construction safety training method, characterized in that, include: When the number of times the target personnel answer questions equals the number of times the response is optimized, the preset analysis conditions corresponding to the target personnel are determined based on the frame selection duration characterization value and the frame selection duration fluctuation value. Under the first preset analysis conditions, the compensation method is determined to be point array compensation or frame parameter compensation based on the similarity value of point clustering and the characteristic value of point clustering corresponding to the distribution map of the test points of the target personnel. When compensating for box parameters, the box parameter compensation strategy is determined based on the ratio of the number of first anchor boxes and second anchor boxes, either by compensating based on box neighborhood parameters or by compensating based on box area. Under the second preset analysis condition, the number of optimized response times is increased; Under the first preset analysis conditions, obtain the distribution map of the test question locations corresponding to each touch delay record; Obtain the point cluster similarity value and point cluster characterization value corresponding to the point distribution map of the test questions; If the similarity value of point clustering is greater than the preset similarity value of point clustering and the characteristic value of point clustering is less than or equal to the preset characteristic value of point clustering, then the compensation method is point array compensation. If the similarity of point clusters is less than or equal to the preset similarity of point clusters or the value of point clusters is greater than the preset value of point clusters, then the compensation method is box parameter compensation. The preset analysis conditions are determined by the frame selection duration characterization value and the frame selection duration fluctuation value; The first preset analysis condition is that the frame selection duration characterization value is less than or equal to the preset frame selection duration characterization value and the frame selection duration fluctuation value is greater than the preset frame selection duration fluctuation value; The second preset analysis condition is that the frame selection duration characterization value is greater than the preset frame selection duration characterization value, or the frame selection duration fluctuation value is less than or equal to the preset frame selection duration fluctuation value.

2. The engineering construction safety training method based on big data according to claim 1, characterized in that, When the compensation method is point array compensation, the cloud storage test questions are extracted in descending order of the point array deviation degree to obtain the required number of cloud storage test questions for compensation. The deviation of the point array is determined based on the difference in array area and the array distance compensation value.

3. The engineering construction safety training method based on big data according to claim 1, characterized in that, When the compensation method is frame parameter compensation, the frame parameter compensation strategy is determined according to the ratio of the number of the first anchor frame and the second anchor frame. If the quantity ratio is less than the preset quantity ratio, the box parameter compensation strategy is to compensate based on the box neighborhood parameters; If the quantity ratio is greater than or equal to the preset quantity ratio, the frame parameter compensation strategy is to compensate based on the frame area.

4. The engineering construction safety training method based on big data according to claim 3, characterized in that, The anchor frame category corresponding to each anchor frame is determined based on the selection deviation value of the anchor frame. The anchor frame category includes the first anchor frame with a selection deviation value greater than the allowable selection deviation value and the second anchor frame with a selection deviation value less than or equal to the allowable selection deviation value.

5. The engineering construction safety training method based on big data according to claim 4, characterized in that, During the process of extracting test questions from cloud storage, the storage module corresponding to the extracted test questions is determined based on the optimal update frequency of each storage module and the engineering diversity.

6. The engineering construction safety training method based on big data according to claim 1, characterized in that, Under the second preset analysis condition, the number of times the optimized response is increased based on the difference in the frame selection time. The increase in the number of optimized responses is positively correlated with the difference in the frame selection time.

7. A system applying the big data-based engineering construction safety training method according to any one of claims 1 to 6, characterized in that, include: The condition analysis unit is used to determine the preset analysis conditions corresponding to the target personnel based on the frame selection duration characterization value and the frame selection duration fluctuation value when the number of times the target personnel answer questions equals the number of times the response is optimized. An optimization compensation unit, which is connected to the condition analysis unit, is used to determine the compensation method as point array compensation or frame parameter compensation based on the point cluster similarity value and point cluster characterization value corresponding to the point touch delay record of each point of the target person under the first preset analysis condition. When performing frame parameter compensation, the frame parameter compensation strategy is determined as compensation based on frame neighborhood parameters or compensation based on frame area based on the ratio of the number of the first anchor frame and the second anchor frame. A category analysis unit, which is connected to the optimization compensation unit, is used to determine the anchor box category corresponding to the anchor box based on the selection deviation value of the anchor box when the optimization compensation unit performs box parameter compensation. A simplified processing unit, which is connected to the conditional analysis unit, is used to increase the number of optimized responses based on the difference in the frame selection duration under the second preset analysis conditions. The cloud data storage unit is connected to the condition analysis unit, optimization compensation unit, category analysis unit and simplification processing unit. The cloud data storage unit includes several storage modules for storing test questions in the cloud.

Images