AI-based special homework examination and training management method and system

Abstract

The invention relates to the technical field of examination management, and discloses an AI-based special homework examination management method and system, and the method comprises the steps: configuring examination tool equipment in an examination site, deploying a plurality of camera devices in the examination site to carry out the shooting of an examination region at different angles, the action posture, the position and the related key articles of the examinee during the examination are collected; and constructing a data set which comprises the annotations of the key target object and the operation attitude of each step in the examination process, and training a target detection task by using an attitude estimation model based on AI. The operation postures, actions and key articles of the examinee are monitored and recorded in real time by using the camera equipment and the AI technology, so that the scoring process is more objective and accurate, the data-driven evaluation mode can better reflect the actual operation ability of the examinee, and compared with a traditional manual evaluation mode, the evaluation efficiency can be improved, and the evaluation accuracy is improved. And the influence of nonstandard scoring caused by human factors is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of examination and training management, and specifically to an AI-based special operation examination and training management method and system. Background Art

[0002] The assessment method of the special operation practical examination and training system is that the trainees actually operate and the examiners manually score. The examinees actually operate on the physical equipment, and the examiners give a comprehensive score based on the examinees' wearing of labor protection supplies, operation procedures, and step operation results.

[0003] In the examination for high-altitude special operations, relying on the traditional examination and training management method, each candidate needs a professional examiner to conduct one-on-one real-time evaluation of the candidate's operations throughout the whole process. This requires a large number of examiners and invigilators at the test sites and training institutions. In addition, it takes a lot of time from the assessment to the calculation of statistical scores, resulting in low examination efficiency and not conducive to large-scale batch examinations. In addition, when manual scoring is used for real operations, it is easy to cause unfairness in the examination due to the personal subjective behavior of the examiners.

[0004] To this end, the present invention provides an AI-based special operations training management method and system. Summary of the invention

[0005] In view of the problems existing in the above-mentioned prior art, the purpose of the present invention is to provide an AI-based special operation examination and training management method and system, so that in the assessment of special operations, the operating conditions of the examinees can be judged and scored through artificial intelligence video recognition and analysis, so as to be more suitable for large-scale assessment and improve the fairness of the assessment.

[0006] In order to achieve the above-mentioned purpose, the present invention provides the following technical solution: a special operation test and training management method based on AI, the method comprising the steps of:

[0007] Test tools and equipment are installed at the test site, and multiple cameras are deployed at the test site to shoot the test area from different angles to capture the test takers' movements, postures, positions, and key items involved during the test;

[0008] Construct a dataset, including the annotation of key target objects and operation postures in each step of the test process, and use an AI-based posture estimation model to train the target detection task. Use the posture estimation model to train and verify the details of the examinee's operation to identify the target objects and operations in the test steps.

[0009] An intermediate node is set in the total number of steps of the test, so that the examinee executes the pass estimation strategy when he / she completes half of the steps of the test. The pass estimation strategy is used to estimate whether the examinee can pass the test when he / she completes all the steps of the test, and to adjust the simulated high-altitude operation dynamic background accordingly according to the estimation result;

[0010] The completion status of the test steps is determined based on the target detection results. Specifically, when a step is detected to be completed, the current frame image is automatically captured and saved to the specified directory, and the timestamp, step information and final score are recorded.

[0011] In some embodiments, the test tools and equipment configured at the test site include: tools and equipment for candidates to operate and equipment that simulates the dynamic background of high-altitude operations, so that candidates can be suspended by safety belts during the high-altitude special operations examination to create a distance between their bodies and the ground, and wind equipment to simulate wind swaying at high altitudes.

[0012] In some embodiments, the pass estimation strategy includes obtaining the passing score ratio. When the examinee's test steps reach the intermediate node, the total score of the steps that the examinee has taken and the actual score are obtained, and the half-way safety score is obtained by the total score of the steps that have been taken, and the actual score is compared with the half-way safety score, and a corresponding response is made according to the comparison result.

[0013] In some embodiments, the specific method of obtaining the half-time safety score is: the total score of the special operation test and passing score Get the passing score ratio and pass the qualifying score ratio and the total score of the steps tested Get half-way safety score ,in, is an adjustment factor greater than 1.

[0014] In some embodiments, the specific response to the result of comparing the actual score with the half-way safety score is: if the actual score is greater than or equal to the half-way safety score, it will be estimated that the candidate has a high probability of passing the exam based on the performance in the first half of the exam, and no adjustment will be made; if the actual score is less than the half-way safety score, it will be estimated that the candidate will find it difficult to complete the exam with a passing score based on the performance in the first half of the exam, and the wind intensity of the wind equipment will be reduced.

[0015] In some embodiments, an environmental adaptability score is added to the test score as an additional score. After the test reaches an intermediate node, a corresponding score is given based on whether the wind strength of the wind power equipment is reduced in the second half of the test.

[0016] When the candidate reaches the middle node in the test steps, if the candidate's actual score is greater than or equal to the half-way safety score, the wind equipment will not be adjusted, and the candidate will be given an environmental adaptability score; if the candidate's actual score is less than the half-way safety score, resulting in the wind equipment reducing the wind speed in the second half of the test, the candidate will not be given an environmental adaptability score.

[0017] In some embodiments, when the examinee has completed half of the steps of the test, if the actual score obtained by the examinee is seriously unsatisfactory, the environmental factor bonus strategy will be implemented to increase the upper limit of the environmental adaptation score that the examinee can obtain;

[0018] The environmental factor bonus strategy includes setting a half-way unsatisfactory score. When the examinee's examination steps reach the middle node, the examinee's actual score at this time is compared with the half-way unsatisfactory score, and a corresponding decision is made based on the comparison result.

[0019] In some embodiments, the specific decision made based on the comparison between the candidate's actual score and the unsatisfactory score of the first half is: if the candidate's actual score is greater than or equal to the unsatisfactory score of the first half, the environmental factor bonus strategy will not be implemented; if the candidate's actual score is less than the unsatisfactory score of the first half, it will be estimated that the probability that the candidate can pass the test based on the performance in the first half of the test is extremely low, then the environmental factor bonus strategy will be implemented, the wind intensity of the wind power equipment will be increased, and after the candidate completes this high-altitude special operations test, the candidate will be given an additional environmental adaptability score and further additional points will be added.

[0020] The present invention also provides the following technical solution: an AI-based special operation test and training management system, comprising:

[0021] The motion capture module includes configuring test tools and equipment at the test site, deploying multiple cameras at the test site to shoot the test area at different angles to capture the test takers' movements, postures, positions, and key items involved during the test;

[0022] The process monitoring module includes building a data set, including the annotation of key target objects and operation postures in each step of the test process, and using an AI-based posture estimation model to train the target detection task. The posture estimation model is used to train and verify the details of the examinee's operation, and identify the target objects and operations in the test steps.

[0023] The estimation adjustment module includes setting an intermediate node in the total number of steps of the test, so that the examinee executes the qualified estimation strategy when completing half of the steps of the test. The qualified estimation strategy is to estimate whether the examinee can pass the test when completing all the steps of the test, and adjust the simulated high-altitude operation dynamic background accordingly according to the estimation result;

[0024] The score recording module includes judging the completion status of the test steps according to the target detection results. Specifically, when a step is detected to be completed, the current frame image is automatically captured and saved to the specified directory, and the timestamp, step information and final score are recorded at the same time.

[0025] The present invention further provides a computer-readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the above-mentioned AI-based special operation training and management method.

[0026] Compared with the prior art, the technical solution provided by the present invention has the following beneficial effects:

[0027] Firstly, the present invention utilizes video equipment and AI technology to monitor and record the examinee's operating posture, movements and key items in real time, making the scoring process more objective and accurate. This data-driven evaluation method can better reflect the examinee's actual operating ability. Compared with the traditional manual assessment method, it can improve the efficiency of the assessment and reduce the impact of non-standard scoring caused by human factors.

[0028] Secondly, by setting intermediate nodes, the present invention can estimate the final scores of candidates during the assessment process, and can implement qualified estimation strategies for candidates near the qualified scores based on the estimated results, so that the performance and scores of candidates in the second half depend more on their operational capabilities and are not overly affected by external environmental factors.

[0029] Thirdly, the present invention incorporates the adjustment of wind simulation into the scoring considerations, which can provide students whose final score estimates are extremely unsatisfactory with an opportunity to gain extra points by increasing the interference of environmental factors, so that the background environmental factors of the examination venue can be dynamically adjusted according to the estimated results of the examinee's final score. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 It is a flow chart of a special operation examination and training management method based on AI of the present invention;

[0031] Figure 2 This is a module schematic diagram of an AI-based special operations training and management system of the present invention. DETAILED DESCRIPTION

[0032] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0033] It is to be understood that the term "one" should be understood as "at least one" or "one or more", that is, in one embodiment, the number of an element may be one, while in another embodiment, the number of the element may be multiple, and the term "one" should not be understood as a limitation on the quantity.

[0034] See also Figure 1 The present invention provides a special operation test and training management method based on AI, the method comprising the following steps:

[0035] Test tools and equipment are installed at the test site, including tools and equipment for candidates to operate and equipment simulating dynamic backgrounds of high-altitude operations, so that candidates can be suspended by safety belts during the high-altitude special operations test to keep their bodies away from the ground, and wind equipment is used to simulate wind swaying at high altitudes. At the same time, multiple cameras are deployed at the test site to shoot the test area from different angles to capture the candidates' operating postures, positions, and key items involved during the test;

[0036] Construct a data set, including the annotation of key target objects (such as tools and equipment) and operation postures in each step of the test process, and use an AI-based posture estimation model to train the target detection task. The posture estimation model is used to train and verify the details of the examinee's operation (such as the posture of grabbing tools and installing equipment). The posture estimation model will be trained and optimized with a data set customized for the test scenario, and can accurately identify the target objects and operations in the test steps;

[0037] The completion of the test steps is determined based on the target detection results. Specifically, when a step is detected to be completed, the current frame image is automatically captured and saved to the specified directory, and the timestamp, step information, and final score are recorded. At the end of the test, a score is generated based on the completion of each step. For example, if there are 2 steps in the test, 10 points are awarded for completing step 1 and 20 points are awarded for completing step 2. The total score is the sum of all step scores, that is, 30 points. At the same time, a test report is generated, including: the total score of the test, the completion status of each step, and the picture record of each step.

[0038] In addition, during the examination process, the total number of steps in the high-altitude special operations examination is obtained, and intermediate nodes are set so that the candidate can execute the pass estimation strategy when he completes half of the examination steps. The role of the pass estimation strategy is to estimate whether the candidate can pass when completing all the steps of the examination, and make corresponding adjustments to the simulated high-altitude operation dynamic background based on the estimation results to increase the candidate's pass probability.

[0039] The pass estimation strategy includes obtaining the pass score ratio. When the examinee's test steps reach the middle node, the total score of the examinee's passed steps and the actual score are obtained, and the half-way safety score is obtained by the total score of the passed steps. The actual score is compared with the half-way safety score, and a corresponding response is made according to the comparison result. Specifically, the total score of the special operation test is and passing score Get the passing score ratio and pass the qualifying score ratio and the total score of the steps tested Get half-way safety score , where k is an adjustment factor greater than 1. For example, the total number of steps in the high-altitude special operation test is set to 4 steps, the total score of the operation test is 100 points, the passing score is 60 points, the score of step 1 is set to 20, the score of step 2 is set to 20, the score of step 3 is set to 30, and the score of step 4 is set to 30. It can be obtained that the passing score ratio is 60÷100=0.6. When the examinee completes step 1 and step 2, the test proceeds to the middle node, and the total score of the steps that have been tested (step 1 and step 2) is 40 points. The adjustment factor k is set to 1.2, and the half-way safety score after rounding is 40×0.6×1.2=29;

[0040] The specific response to the result of comparing the actual score with the half-way safety score is: if the actual score is greater than or equal to the half-way safety score, it means that the score obtained by the examinee when half of the test steps of the high-altitude special operation test are completed is ideal enough, and it will be estimated that the examinee has a high probability of passing the test based on the performance of the first half of the test, and no adjustment will be made; if the actual score is less than the half-way safety score, it means that the score obtained by the examinee when half of the test steps of the high-altitude special operation test are completed is not ideal, and it will be estimated that it will be difficult for the examinee to complete the test with a qualified score based on the performance of the first half of the test, and the wind strength of the wind equipment will be reduced. For example, the half-way safety score is set to 29 points, and the examinee's actual score is 26 points when the test steps reach the middle node. Since the actual score is less than the half-way safety score, the wind strength of the wind equipment will be reduced, so that the examinee will be less likely to be shaken by the wind in the second half of the test, thereby increasing the examinee's passing probability.

[0041] The reason for adjusting the wind force is that during the test, although wind equipment is needed to allow candidates to simulate the high-altitude working environment more realistically, the purpose of using wind equipment in the test is only to simulate the shaking environment at high altitudes, and it will not be used as an indicator for assessment and scoring, because under the same standard wind force and wind direction settings, the shaking degree caused to each candidate will be different due to the different body shapes, different positions of the center of gravity of the body, and different weights of each candidate. When there is a situation where a candidate may fail, reducing the wind force can reduce the possibility of the candidate's posture and movements being non-standard due to wind shaking to a certain extent, and make the candidate's performance and results in the second half of the test more dependent on their operating ability, rather than being overly affected by external environmental factors. In addition, the test for high-altitude special operations is not to eliminate as many candidates as possible, but to verify whether the candidate has the ability to actually operate and safety awareness. Through this reasonable design of the test and personalized adjustments, it is the goal of the test design to ensure that the candidate can demonstrate his or her ability under controllable risks.

[0042] In addition, considering the examination and training for different high-altitude work scenarios, in addition to testing the candidates' operating accuracy, posture standards and time control indicators, when necessary, the adjustment of wind simulation can also be included in the scoring considerations. The performance of candidates under different wind intensities can be given an "environmental adaptability" score. In this way, not only can the wind factor be taken into consideration, but also it can ensure that candidates can get reasonable challenges and scores under different physical conditions. Specifically, the environmental adaptability score is added to the score of the exam as an additional score. After the exam reaches the middle node, the corresponding score is made according to whether the wind intensity of the wind equipment is reduced in the second half of the exam. When the candidate reaches the middle node in the exam steps, if the candidate's actual score is greater than or equal to the half-way safety score, the wind equipment will not be adjusted, and the candidate will be given an environmental adaptability score; if the candidate's actual score is less than the half-way safety score, resulting in the wind equipment reducing the wind in the second half of the exam, the candidate will not be given an environmental adaptability score.

[0043] At the same time, the present invention provides another preferred embodiment. During the examination, if the actual score obtained by the examinee is seriously unsatisfactory when the examinee completes half of the examination steps, the environmental factor bonus strategy will be implemented to increase the upper limit of the environmental adaptation score that the examinee can obtain. The environmental factor bonus strategy includes setting a half-way unsatisfactory score. When the examinee's examination steps reach the middle node, the examinee's actual score at this time is compared with the half-way unsatisfactory score, and a corresponding decision is made based on the comparison result. Furthermore, through the qualified score ratio and the total score of the steps tested Get unsatisfactory half-course score , where w is an adjustment factor less than 1. Taking the above embodiment as an example, the total score of the passed steps (step 1 and step 2) is set to 40 points, and the adjustment factor w is set to 0.5, so that the half-way unsatisfactory score is 40×0.6×0.5=12.

[0044] The specific decision made based on the comparison between the candidate's actual score and the unsatisfactory score of the first half is: if the candidate's actual score is greater than or equal to the unsatisfactory score of the first half, it will be estimated that the candidate has a normal probability of passing the exam based on his performance in the first half of the exam, and the environmental factor bonus strategy will not be implemented; if the candidate's actual score is less than the unsatisfactory score of the first half, it means that the candidate's score in the first half of the exam is extremely unsatisfactory, and it will be estimated that the probability of the candidate passing the exam based on his performance in the first half of the exam is extremely low, then the environmental factor bonus strategy will be implemented, the wind intensity of the wind equipment will be increased, and after the candidate completes this high-altitude special operations test, the candidate will be given an additional environmental adaptability score and further additional points will be given.

[0045] For example, the total score of the homework test can be set to 100 points, and a 5-point environmental adaptability score can be set as an additional score. When the candidate's test steps reach the middle node (completion of steps 1 and 2), the actual score is set to 11 points. This score is less than the unsatisfactory score of 12 points in the first half. At this time, the wind strength of the wind power equipment will be increased, so that the candidate will receive a stronger environmental factor challenge in the second half of the test steps. At the same time, the environmental adaptability score as an additional score will be doubled to the candidate, that is, the total score of the candidate's test is 100 points, and the additional score is 10 points. This design constructs a dynamic reverse incentive mechanism. For candidates who have reached the middle node of the test but actually perform poorly, the worse the performance, the greater the challenge given to the candidate in the second half of the test. It also enables the candidate to increase the probability of passing the second half of the test by additionally scoring the environmental adaptability score, giving the candidate a chance to compensate in the second half of the test steps. In addition, when further adding points to the additional environmental adaptability score, not only can the double addition method be used, but the half-way passing score can also be obtained, that is, through the passing score ratio and the total score of the steps tested Get half-way passing score , calculate the difference between the half-way passing score and the actual score of the examinee when the test steps reach the middle node, and mark this difference as the adjustment value. The size of the adjustment value represents the degree to which the examinee is able to pass the test. The larger the difference, the more it means that the examinee's score result in the first half is not ideal and it is difficult to pass. When the wind power of the wind power equipment is strengthened, the degree of wind power strengthening can be adaptively adjusted according to the size of the adjustment value. The larger the adjustment value, the greater the degree to which the wind power equipment strengthens the wind power. Correspondingly, the further additional score value given to the examinee's environmental adaptability score is also greater.

[0046] In general, the present invention aims to design a special operation test and training management method based on AI, which is aimed at the problem that in the existing high-altitude special operation examination, each examinee needs to be equipped with an examiner, resulting in low examination efficiency and non-standard scoring. The present invention uses camera equipment and AI technology to monitor and record the examinee's operating posture, movements and key items in real time, making the scoring process more objective and accurate. This data-driven evaluation method can better reflect the examinee's actual operating ability, and can improve the efficiency of the assessment compared to the traditional manual assessment method, and reduce the impact of non-standard scoring caused by human factors. After the examination, a detailed examination report can be generated, including the completion status and image records of each step, providing rich feedback information for the examinee's subsequent learning and improvement. By setting the intermediate node, the examinee's final score can be roughly evaluated during the assessment process, and the examinee near the qualified score can be executed according to the estimated results. The qualified estimation strategy makes the examinee's performance and score in the second half more dependent on their operating ability, and not overly affected by external environmental factors. By incorporating the adjustment of wind simulation into the scoring considerations, students whose final score estimates are extremely unsatisfactory can gain extra points by increasing the interference of environmental factors, so that the background environmental factors of the examination venue can be dynamically adjusted according to the estimated results of the candidate's final score.

[0047] See also Figure 2 The present invention provides an AI-based special operation test and training management system, comprising:

[0048] The motion capture module includes configuring test tools and equipment at the test site, deploying multiple cameras at the test site to shoot the test area at different angles to capture the test takers' movements, postures, positions, and key items involved during the test;

[0049] The process monitoring module includes building a data set, including the annotation of key target objects and operation postures in each step of the test process, and using an AI-based posture estimation model to train the target detection task. The posture estimation model is used to train and verify the details of the examinee's operation, and identify the target objects and operations in the test steps.

[0050] The estimation adjustment module includes setting an intermediate node in the total number of steps of the test, so that the examinee executes the qualified estimation strategy when completing half of the steps of the test. The qualified estimation strategy is to estimate whether the examinee can pass the test when completing all the steps of the test, and adjust the simulated high-altitude operation dynamic background accordingly according to the estimation result;

[0051] The score recording module includes judging the completion status of the test steps according to the target detection results. Specifically, when a step is detected to be completed, the current frame image is automatically captured and saved to the specified directory, and the timestamp, step information and final score are recorded at the same time.

[0052] The embodiments disclosed in the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. The embodiments disclosed in the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part, and / or installed from a removable medium. When the computer program is executed by the central processing unit, the above functions defined in the method of the present application are executed. It should be noted that the computer-readable medium mentioned above in the present application can be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium can be, for example, but not limited to, a system, device or device of an electrical, magnetic, optical, electromagnetic, infrared segment, or semiconductor, or any combination of the above. More specific examples of computer-readable storage media can include, but are not limited to: an electrical connection with one or more wire segments, a portable computer disk, a hard disk, a random access memory, a read-only memory, an erasable programmable read-only memory, an optical fiber, a portable compact disk read-only memory, an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, apparatus, or device. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in combination with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium may be transmitted using any suitable medium, including but not limited to: wireless segments, wire segments, optical cables, RF, etc., or any suitable combination of the above.

[0053] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0054] Those skilled in the art should understand that the above description is only a specific implementation mode of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be covered by the protection scope of the present application.

Claims

1. A special operation training management method based on AI, characterized in that: The method comprises the steps of: Test tools and equipment are installed at the test site, and multiple cameras are deployed at the test site to shoot the test area from different angles to capture the test takers' movements, postures, positions, and key items involved during the test; Construct a dataset, including the annotation of key target objects and operation postures in each step of the test process, and use an AI-based posture estimation model to train the target detection task. Use the posture estimation model to train and verify the details of the examinee's operation to identify the target objects and operations in the test steps. An intermediate node is set in the total number of steps of the test, so that the examinee executes the pass estimation strategy when he / she completes half of the steps of the test. The pass estimation strategy is used to estimate whether the examinee can pass the test when he / she completes all the steps of the test, and to adjust the simulated high-altitude operation dynamic background accordingly according to the estimation result; The completion status of the test steps is determined based on the target detection results. Specifically, when a step is detected to be completed, the current frame image is automatically captured and saved to the specified directory, and the timestamp, step information and final score are recorded.

2. According to the AI-based special operation training management method of claim 1, it is characterized in that: The examination tools and equipment configured at the examination venue include: tools and equipment for candidates to operate and equipment simulating the dynamic background of high-altitude operations, so that candidates can be suspended by safety belts during the high-altitude special operations examination to create a distance between their bodies and the ground, as well as wind equipment to simulate wind swaying at high altitudes.

3. The AI-based special operation training management method according to claim 2 is characterized in that: The pass estimation strategy includes obtaining the pass score ratio. When the examinee's test steps reach the middle node, the total score of the examinee's passed steps and the actual score are obtained, and the half-way safety score is obtained by the total score of the passed steps, and the actual score is compared with the half-way safety score, and a corresponding response is made according to the comparison result.

4. The AI-based special operation training management method according to claim 3 is characterized in that: The specific way to get the half-time safety score is: the total score of the special operation test and passing score Get the passing score ratio and pass the qualifying score ratio and the total score of the steps tested Get half-way safety score , where k is a regulation factor greater than 1.

5. The AI-based special operation training management method according to claim 4 is characterized in that: The specific response to the result of comparing the actual score with the half-way safety score is: if the actual score is greater than or equal to the half-way safety score, it will be estimated that the candidate has a high probability of passing the exam based on his performance in the first half of the exam, and no adjustment will be made; if the actual score is less than the half-way safety score, it will be estimated that the candidate will find it difficult to complete the exam with a passing score based on his performance in the first half of the exam, and the wind intensity of the wind equipment will be reduced.

6. The AI-based special operation training management method according to claim 5 is characterized in that: An additional score for environmental adaptability is added to the test score. After the test reaches the middle node, a corresponding score is given based on whether the wind strength of the wind power equipment is reduced in the second half of the test. When the candidate reaches the middle node in the test steps, if the candidate's actual score is greater than or equal to the half-way safety score, the wind equipment will not be adjusted, and the candidate will be given an environmental adaptability score; if the candidate's actual score is less than the half-way safety score, resulting in the wind equipment reducing the wind speed in the second half of the test, the candidate will not be given an environmental adaptability score.

7. The AI-based special operation training management method according to claim 6 is characterized in that: If the candidate has completed half of the test steps and the actual score is seriously unsatisfactory, the environmental factor bonus strategy will be implemented to increase the upper limit of the environmental adaptation score that the candidate can obtain; The environmental factor bonus strategy includes setting a half-way unsatisfactory score. When the examinee's examination steps reach the middle node, the examinee's actual score at this time is compared with the half-way unsatisfactory score, and a corresponding decision is made based on the comparison result.

8. The AI-based special operation training management method according to claim 7 is characterized in that: The specific decision made based on the comparison between the candidate's actual score and the unsatisfactory score of the first half is: if the candidate's actual score is greater than or equal to the unsatisfactory score of the first half, the environmental factor bonus strategy will not be implemented; if the candidate's actual score is less than the unsatisfactory score of the first half, it will be estimated that the probability that the candidate can pass the exam based on the performance in the first half of the exam is extremely low, then the environmental factor bonus strategy will be implemented, the wind intensity of the wind equipment will be increased, and after the candidate completes this high-altitude special operations test, the candidate will be given an additional environmental adaptability score and further additional points will be added.

9. An AI-based special operation training and examination management system, characterized in that: According to any one of claims 1 to 8, an AI-based special operation training management method comprises: The motion capture module includes configuring test tools and equipment at the test site, deploying multiple cameras at the test site to shoot the test area at different angles to capture the test takers' movements, postures, positions, and key items involved during the test; The process monitoring module includes building a data set, including the annotation of key target objects and operation postures in each step of the test process, and using an AI-based posture estimation model to train the target detection task. The posture estimation model is used to train and verify the details of the examinee's operation, and identify the target objects and operations in the test steps. The estimation adjustment module includes setting an intermediate node in the total number of steps of the test, so that the examinee executes the qualified estimation strategy when completing half of the steps of the test. The qualified estimation strategy is to estimate whether the examinee can pass the test when completing all the steps of the test, and adjust the simulated high-altitude operation dynamic background accordingly according to the estimation result; The score recording module includes judging the completion status of the test steps according to the target detection results. Specifically, when a step is detected to be completed, the current frame image is automatically captured and saved to the specified directory, and the timestamp, step information and final score are recorded at the same time.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement an AI-based special operation training management method as described in any one of claims 1 to 8.

Images