Target detection and identification task-level simulation model construction method for variable-focus photoelectric pod

Abstract

The invention relates to an airborne variable-focus photoelectric pod, and provides a variable-focus photoelectric pod target detection and recognition task-level simulation model construction method in order to solve the problem that the target observation reliability is low in the detection and recognition process of an existing airborne variable-focus photoelectric pod. Sequentially carrying out detection resolving and identification confirmation resolving on all targets in the view field range of each simulation period; the method comprises the following steps: firstly, detecting and resolving all targets in a view field range of a current simulation period, then identifying, confirming and resolving the targets in a detected target list, listing the targets which can be identified and confirmed in an identified and confirmed target list, and correspondingly updating confirmation time starting points of other targets in the detected target list, therefore, detection and identification of the airborne variable-focus photoelectric pod on a ground target are completed, and the physical authenticity and data reliability of a simulation system are effectively improved.

Description

Technical Field

[0001] The present invention relates to an airborne variable-focus optoelectronic pod, and more particularly to a method for constructing a simulation model for the target detection and recognition task level of a variable-focus optoelectronic pod. Background Technique

[0002] The airborne optoelectronic pod technology is an important part of the optoelectronic reconnaissance warning technology and its equipment, and can be used for the observation, precise search, tracking and locking of targets by an air-mobile platform. The optoelectronic imaging of an unmanned aerial vehicle realized by using an airborne optoelectronic pod has technical advantages such as high real-time resolution and good real-time performance, and it is widely used in the reconnaissance on land, at sea, in the air and in space, and its carriers are vehicles, ships, airplanes, satellites, etc.

[0003] At present, for the optoelectronic imaging observation model of an unmanned aerial vehicle, the image-level model introduces many parameters and has high computational complexity, and its practicability is not strong in large-scale entity simulation and deduction. Therefore, the simplified task-level model has become the mainstream of research. However, the currently adopted task-level model only considers the observation range and approximates the observation range as a sector or a circle, which has a large deviation from the actual situation and seriously restricts its application value in simulation.

[0004] Therefore, a task-level model that introduces the target perception probability of an optoelectronic payload in the field of view is needed, which comprehensively considers the observation distance, the target size and the resolution of the optoelectronic pod camera to describe the actual observation performance of the airborne optoelectronic pod. In order to more accurately describe the detection and recognition process of an airborne variable-focus optoelectronic pod for ground targets in large-scale entity simulation, it is necessary to develop a modeling method for the detection and recognition process of an airborne variable-focus optoelectronic pod for ground targets. Summary of the Invention

[0005] The purpose of the present invention is to solve the technical problems that the existing task-level measurement model of the target perception probability of an optoelectronic payload in the field of view has a low observation reliability for the target and cannot accurately give the actual observation performance of the airborne optoelectronic pod, and to propose a method for constructing a simulation model for the target detection and recognition task level of a variable-focus optoelectronic pod.

[0006] The design idea of the present invention is as follows: In each simulation cycle, when the number of pixels occupied by a target exceeds a threshold value, the target can be detected, the target detection probability is calculated and corrected to obtain the actual detection probability, and whether the target is detected is simulated by generating an equiprobability random number to realize the target detection calculation. Then, the time interval from the starting time of the confirmation of the detected target to the current simulation time is compared with the average time required to recognize and confirm a single target to determine whether the target is recognized, and the target recognition calculation is realized.

[0007] To achieve the above object, the technical solution proposed by the present invention is:

[0008] A method for constructing a task-level simulation model for target detection and recognition of a variable-focus optoelectronic pod, which is characterized in that within M simulation cycles, detection calculations and recognition confirmation calculations are sequentially performed on all targets within the field of view of each simulation cycle; it includes the following steps:

[0009] S1. Perform detection calculations on all targets within the field of view of the current simulation cycle;

[0010] Sequentially determine whether each target detected in the current simulation cycle exists in the previous simulation cycle; for the targets that do not exist, calculate their number of pixels and obtain their target detection probabilities; process the target detection probabilities through a probability correction function to obtain the actual detection probabilities, and compare the actual detection probabilities with the equally probable detection random numbers generated in the [0, 1] interval to determine whether the target can be detected; list the detected targets in the list of detected targets to form the list of detected targets for the current simulation cycle;

[0011] S2. Sequentially perform recognition confirmation calculations on the targets in the list of detected targets;

[0012] Determine whether each target in the list of detected targets for the current simulation cycle has been recognized and confirmed in the previous simulation cycle; for the targets that have not been recognized and confirmed, calculate the time interval between the current simulation time and the starting time of the target confirmation; compare this time interval with the average time required to recognize and confirm a target to determine whether the target can be recognized and confirmed; list the targets that can be recognized and confirmed in the list of recognized and confirmed targets, and at the same time update the starting time of the confirmation of the remaining targets in the pre-recognized target list accordingly, so as to complete the detection and recognition of ground targets by the variable-focus optoelectronic pod.

[0013] Further, step S1 is specifically:

[0014] S1.1. Determine whether the target Target within the field of view of the current simulation cycle K exists in the list of all targets within the field of view of the previous simulation cycle K - 1, where K = 1, 2, 3,..., M, and M is the number of simulation cycles; I is the total number of all targets within the field of view of the simulation cycle, and i = 1, 2, 3,..., I; the list of all targets within the field of view of each simulation cycle includes target ID, historical highest detection probability, target confirmation starting time, and most recent detection time; i If it does not exist, initialize the historical highest detection probability P of this target Target

[0015] to 0, calculate the number of pixels of this target Target i and execute step S1.2; targeti If it exists, directly calculate this target Target i and execute step S1.2;

[0016] If it exists, directly calculate the number of pixels of this target Targeti For the number of pixels, execute step S1.2;

[0017] S1.2. Determine whether the number of pixels of the target is not less than the set threshold of the minimum detectable number of pixels;

[0018] If not, then the target Target i cannot be detected, then execute step S1.6;

[0019] If so, execute step S1.3;

[0020] S1.3. Determine whether the target Target i exists in the list of detected targets detected in the previous simulation cycle ;

[0021] If it exists, update the detection time t of the current simulation cycle to the most recent detection time t i of the target Target targeti and record the target Target i in the list of detected targets of the current simulation cycle and execute step S1.6;

[0022] If it does not exist, calculate the target detection probability P i of the target Target i and execute step S1.4;

[0023] S1.4. Determine whether the target detection probability P i of the target Target i is greater than its historical highest detection probability P targeti ;

[0024] If not, execute step S1.6;

[0025] If so, process the target detection probability P i through a probability correction function to obtain the actual detection probability P and execute step S1.5;

[0026] S1.5. Generate an equally probable detection random number P s in the interval [0, 1], compare the actual detection probability P with the equally probable detection random number P s and determine whether the target Target i is detected;

[0027] If the equally probable detection random number P s > the actual detection probability P, it means that the target Target i is not detected, and within the current simulation cycle, the target Targeti The target detection probability P i is updated to the historical highest detection probability P targeti , and step S1.6 is executed;

[0028] If the equal-probability detection random number P s ≤ the actual detection probability P, it means that the target Target i is detected. Record the detection time t of the current simulation cycle as the target confirmation time start point t i of the target Target 0targeti and the nearest detection time t targeti , and record the target Target i into the list of detected targets in the current simulation cycle , and then execute step S1.6;

[0029] S1.6. Determine whether the target Target i is the last target within the field of view in the current simulation cycle;

[0030] If not, return to step S1.1 to perform detection and solution for the next target;

[0031] If so, end the detection process and obtain the list of detected targets in the current simulation cycle K

[0032] Furthermore, in step S1.3, the calculation of the target detection probability P i satisfies the following formula:

[0033]

[0034] E = 2.7 + 0.7(N / N 50 )

[0035] where: N is the number of pixels occupied by the target Target i , N 50 is the minimum detectable pixel number with a 50% detection probability.

[0036] Furthermore, in step S1.4, the calculation formula for obtaining the actual detection probability P by processing the target detection probability P i through the probability correction function is:

[0037]

[0038] where: P is the actual detection probability, P i is the target detection probability of the i-th target Target i in the current simulation cycle, P targetiFor the i-th target i The historical highest detection probability.

[0039] Furthermore, step S2 is specifically as follows:

[0040] S2.1. Determine whether the targets in the detected target list of the current simulation cycle K j exist in the recognized and confirmed target list of the previous simulation cycle K - 1 , where the total number of targets in the detected target list of the current simulation cycle K is J, and j = 1, 2, 3, …, J;

[0041] If they exist, execute step S2.3;

[0042] If they do not exist, calculate the time interval t’ between the current simulation time and the start time of target confirmation of this target j , and execute step S2.2;

[0043] S2.2. Set the average time required to recognize and confirm a target as T recognize , and compare the time interval t’ calculated in step S2.1 with the average time T recognize ;

[0044] If the time interval t’ ≥ the average time T recognize , then add this target j to the recognized and confirmed target list T recognized , and at the same time correspondingly update the start time t of confirmation of the remaining targets in the detected target list , and execute step S2.3; 0targetj

[0045] If the time interval t’ < the average time T recognize , then execute step S2.3;

[0046] S2.3. Determine whether this target j is the last target in the detected target list of the current simulation cycle ;

[0047] If not, return to step S2.1 to perform the recognition and confirmation solution for the next target;

[0048] If so, end the recognition and confirmation solution process to obtain the recognized and confirmed target list T of the current simulation cycle recognized , thereby completing the detection and recognition of ground targets by the zoom electro-optical pod.

[0049] Further, in step S2.2, the confirmation time start point t of the remaining targets in the updated detected target list is updated according to the following formula: 0targetj

[0050] t 0targetj = t 0targetj + T recognize k .

[0051] Advantages of the present invention:

[0052] A method for constructing a task-level simulation model for target detection and recognition of a variable-focus optoelectronic pod of the present invention calculates the target detection probability through the number of pixels occupied by the target, introduces a probability correction function to correct the target detection probability, and obtains the actual detection probability of the target in the current simulation period, solving the problem that the total detection probability of the target being detected is higher than the actual value due to the detection calculation of the target in multiple consecutive simulation periods after entering the field of view once, greatly reducing the system calculation amount and improving the calculation efficiency of the simulation system; by generating an equiprobable random number and comparing it with the actual detection probability of the target to determine whether the target is detected, and for the detected target, by comparing the time interval from its confirmation time start point to the current simulation time with the average time required to identify and confirm a single target, it is determined whether the target is recognized, effectively realizing the simulation of the detection and recognition of ground targets by an airborne variable-focus optoelectronic pod, and improving the physical authenticity and reliability of the simulation system. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 is a flowchart of the detection calculation in an embodiment of a method for constructing a task-level simulation model for target detection and recognition of a variable-focus optoelectronic pod of the present invention;

[0054] Figure 2 is a flowchart of the recognition and confirmation calculation in an embodiment of a method for constructing a task-level simulation model for target detection and recognition of a variable-focus optoelectronic pod of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0055] In M simulation periods, all target lists within the field of view in each simulation period contain information such as {target ID, historical highest detection probability, target confirmation time start point, and most recent detection time}. In this embodiment, there are a total of 5 targets within the field of view in the k-th simulation period. Among them, Target1 has been detected, Target2 and Target3 existed within the field of view in the previous simulation period K-1 but were not detected, and Target4 and Target5 are newly emerged targets within the field of view in the current simulation period K.

[0056] The target list T of the 5 targets within the field of view in the k-th simulation period k is as follows:

[0057] T k = {Target1, Target2, Target3, Target4, Target5}

[0058] Target1 = {1, 0, 309, 300}

[0059] Target2 = {2, 0.5, 0, 0}

[0060] Target3 = {3, 0.5, 0, 0}

[0061] Target4 = {4, 0, 0, 0}

[0062] Target5 = {5, 0, 0, 0}

[0063] The solution calculation of the task-level model for detecting and identifying ground targets by a zoom electro-optical pod can be divided into two steps: detection solution calculation and identification confirmation solution calculation.

[0064] A method for constructing a task-level simulation model for detecting and identifying targets by a zoom electro-optical pod includes the following steps:

[0065] S1, as Figure 1 shown, perform detection solution calculation on all targets within the field of view of the current simulation cycle;

[0066] Successively judge whether each detected target Target i in the current simulation cycle K exists in the previous simulation cycle; calculate the number of pixels of the target that does not exist, and obtain its target detection probability; process the target detection probability through a probability correction function to obtain the actual detection probability, and compare the actual detection probability with an equal-probability detection random number generated in the interval [0, 1] to judge whether the target can be detected; list the detected targets in the list of detected targets, and form the list of detected targets in the current simulation cycle;

[0067] Step S1 specifically includes the following steps:

[0068] S1. Judge whether the target Target within the field of view of the current simulation cycle K i exists in the list of all targets within the field of view of the previous simulation cycle K - 1, where K = 1, 2, 3,..., M, and M is the number of simulation cycles; I is the total number of all targets within the field of view of the simulation cycle, and i = 1, 2, 3,..., I; the list of all targets within the field of view of each simulation cycle contains target ID, historical highest detection probability, target confirmation time start point, and most recent detection time;

[0069] If it does not exist, initialize the target Target iThe historical highest detection probability P targeti is 0, and calculate the number of pixels of this target Target i , and execute step S1.2;

[0070] If it exists, directly calculate the number of pixels of this target Target i , and execute step S1.2;

[0071] S1.2. Judge whether the number of pixels of this target is not less than the set threshold of the minimum detectable number of pixels;

[0072] If not, this target Target i cannot be detected, and execute step S1.6;

[0073] If so, execute step S1.3;

[0074] S1.3. Judge whether this target Target i exists in the list of detected targets detected in the previous simulation cycle[[ID=2,6]] ;

[0075] If it exists, update the detection time t of the current simulation cycle to the most recent detection time t i of this target Target targeti , and record this target Target i into the list of detected targets of the current simulation cycle , and execute step S1.6;

[0076] If it does not exist, calculate the target detection probability P i of this target Target i , and execute step S1.4;

[0077] S1.4. Judge whether the target detection probability P i of this target Target i is greater than its historical highest detection probability P targeti ;

[0078] If not, execute step S1.6;

[0079] If so, process the target detection probability P i through the probability correction function to obtain the actual detection probability P, and execute step S1.5;

[0080] S1.5. Generate an equally probable detection random number P s in the interval [0, 1], compare the actual detection probability P with the equally probable detection random number P s , and judge whether this target Target i is detected;

[0081] If the equal-probability detection random number P s > the actual detection probability P, it means that the target Target i has not been detected. Update the target detection probability P i of this target Target i in the current simulation cycle to the historical highest detection probability P targeti , and execute step S1.6;

[0082] If the equal-probability detection random number P s ≤ the actual detection probability P, it means that the target Target i has been detected. Record the detection time t of the current simulation cycle as the target confirmation time starting point t i and the most recent detection time t 0targeti of this target Target targeti , and record this target Target i into the list of detected targets in the current simulation cycle , and then execute step S1.6;

[0083] S1.6. Determine whether this target Target i is the last target within the field of view in the current simulation cycle;

[0084] If not, return to step S1.1 to perform detection and solution for the next target;

[0085] If so, end the detection process and obtain the list of detected targets in the current simulation cycle K

[0086] In this embodiment, the number of target pixels calculated in step S1.1 are respectively: Target1: 20, Target2: 10, Target3: 18, Target4: 15, Target5: 8. The minimum detectable pixel number is 10, so Target5 cannot be detected.

[0087] In this embodiment, in step S1.3, the calculation of the target detection probability P i satisfies the following formula:

[0088]

[0089] E = 2.7 + 0.7(N / N 50 )

[0090] where: N is the number of pixels occupied by this target Target i , N 50 is the minimum detectable pixel number, with a detection probability of 50%.

[0091] Since Target1 was detected in the previous cycle, it is directly added to the list of detected targets in the current simulation cycle. No further calculations are performed. Based on the number of target pixels of Target2, Target3, and Target4, the calculated target detection probabilities are 0.5, 0.91, and 0.82 respectively.

[0092] In step S1.4 of this embodiment, the probability correction function is:

[0093]

[0094] The detection probability P2 of Target2 is equal to its historical highest detection probability and cannot be detected in the current simulation cycle K.

[0095] The historical highest detection probabilities of Target3 and Target4 are 0.5 and 0 respectively. The corrected actual detection probabilities P of Target3 and Target4 are calculated to be 0.82 and 0.82.

[0096] Equally probable detection random numbers P for Target3 and Target4 are generated in the interval [0, 1]. s They are 0.56 and 0.25. Both 0.25 and 0.56 are less than 0.82, so Target3 and Target4 are detected in the current simulation cycle K.

[0097] All the detected targets finally calculated in this embodiment are T k The target Target1, Target3, and Target4 in it, obtaining the list of detected targets in the current simulation cycle K. It is:

[0098]

[0099] Target1 = {1, 0, 312, 300}

[0100] Target2 = {3, 0.5, 312, 312}

[0101] Target3 = {4, 0, 312, 312}

[0102] S2. As Figure 2 shown, the targets in the pre-detected target list are sequentially identified and confirmed for calculation. In the target list of detected targets in the current simulation cycle

[0103] Judge Whether the target in it has been recognized and confirmed in the previous simulation cycle; calculate the time interval between the current simulation time and the start time of the target confirmation for the unrecognized and unconfirmed target; compare this time interval with the average time required to recognize and confirm a target to determine whether the target can be recognized and confirmed; list the target that can be recognized and confirmed in the recognized and confirmed target list, and at the same time update the start time of the confirmation of the remaining targets in the pre-recognized target list accordingly, so as to complete the detection and recognition of the ground target by the variable-focus optoelectronic pod.

[0104] Step S2 specifically includes the following steps:

[0105] S2.1. Determine whether the target Target in the detected target list of the current simulation cycle K j exists in the recognized and confirmed target list of the previous simulation cycle K-1, where the total number of targets in the detected target list of the current simulation cycle K is J, and j = 1, 2, 3,..., J;

[0106] If it exists, execute step S2.3;

[0107] If it does not exist, calculate the time interval t' between the current simulation time and the start time of the target Target j confirmation, and execute step S2.2;

[0108] S2.2. Set the average time required to recognize and confirm a target as T recognize , and compare the time interval t' calculated in step S2.1 with the average time T recognize ;

[0109] If the time interval t' ≥ average time T recognize , then add the target Target j to the recognized and confirmed target list T recognized , and at the same time update the start time of the confirmation t of the remaining targets in the detected target list 0targetj , and execute step S2.3;

[0110] Update the start time of the confirmation t of the remaining targets in the detected target list 0targetj using the following update calculation formula:

[0111] t 0targetj = t 0targetj + T recognize ;

[0112] If the time interval t' < average time T recognize , then execute step S2.3;

[0113] S2.3. Determine the target Target j Whether it is the last target in the list of detected targets in the current simulation cycle ;

[0114] If not, return to step S2.1 to perform the recognition and confirmation calculation for the next target;

[0115] If so, end the recognition and confirmation calculation process to obtain the list T of recognized and confirmed targets in the current simulation cycle recognized , thus completing the detection and recognition of ground targets by the variable-focus optoelectronic pod.

[0116] In this embodiment, Target1, Target2, and Target3 are not in the list of detected targets .

[0117] In step S2.1, the current time of Target1 is 312 seconds, and the interval between the current simulation time and the start time of the target confirmation time of Target1 is calculated to be 12 seconds. The average time T recognize required to confirm a target is 10 seconds. Target1 is recognized, and at the same time, the start time of the confirmation time of Target3 is updated to 322 seconds.

[0118] In this embodiment, the list T of recognized and confirmed targets in the current simulation cycle is finally calculated recognized to include the target Target1, thus completing the detection and recognition of ground targets by the airborne variable-focus optoelectronic pod.

Claims

1. A method for constructing a task-level simulation model for target detection and recognition of a variable-focus optoelectronic pod, characterized in that In M simulation cycles, detection and resolution and identification and confirmation resolution are successively performed on all targets within the field of view of each simulation cycle; the steps are as follows: S1. Perform detection and resolution on all targets within the field of view of the current simulation cycle; Successively determine whether each target detected in the current simulation cycle exists in the previous simulation cycle; calculate the number of pixels of the target that does not exist, and obtain its target detection probability; process the target detection probability through a probability correction function to obtain the actual detection probability, and compare the actual detection probability with an equal-probability detection random number generated in the interval [0, 1] to determine whether the target can be detected; list the detected targets in the list of detected targets to form the list of detected targets in the current simulation cycle; S2. Successively perform identification and confirmation resolution on the targets in the list of detected targets; Determine whether each target in the list of detected targets in the current simulation cycle has been identified and confirmed in the previous simulation cycle; calculate the time interval between the current simulation time and the start time of the target confirmation time for the target that has not been identified and confirmed; compare the time interval with the average time required to identify and confirm a target to determine whether the target can be identified and confirmed; list the targets that can be identified and confirmed in the list of identified and confirmed targets, and at the same time update the start time of the confirmation time of the remaining targets in the pre-identified target list accordingly, so as to complete the detection and identification of ground targets by the variable-focus optoelectronic pod.

2. The method for constructing a simulation model at the mission level for target detection and recognition of a variable-focus optoelectronic pod according to claim 1, wherein, The specific steps of step S1 are as follows: S1.

1. Determine whether the target Target within the field of view in the current simulation cycle K exists in the list of all targets within the field of view in the previous simulation cycle K-1, where K = 1, 2, 3, …, M, and M is the number of simulation cycles; I is the total number of all targets within the field of view in the simulation cycle, and i = 1, 2, 3, …, I; the list of all targets within the field of view in each simulation cycle contains the target ID, the historical highest detection probability, the starting point of the target confirmation time, and the most recent detection time. i ​ If not, initialize the target Target i The historical highest detection probability P targeti is 0, and calculate the number of pixels of the target Target i and execute step S1.2; If it exists, directly calculate the target Target i the number of pixels, and execute step S1.2; S1.

2. Determine whether the number of pixels of the target is not less than the set threshold of the minimum detectable number of pixels; Otherwise, the target i cannot be detected, and step S1.6 is executed; If so, execute step S1.3; S1.

3. Determine the target i whether it exists in the list of detected targets detected in the previous simulation cycle ; If it exists, the detection time t of the current simulation cycle is updated to the target Target i The most recent detection time t targeti Target i List of detected targets recorded in the current simulation cycle , execute step S1.6; If not, calculate the target Target i of the target detection probability P i , and execute step S1.4; S1.

4. Determine the target Target i 's target detection probability P i and check if it is greater than its historical highest detection probability P targeti ; If not , execute step S1.6; If so, process the target detection probability P through a probability correction function i to obtain the actual detection probability P, and execute step S1.5; S1.

5. Generate an equiprobable detection random number P in the interval [0, 1] s , compare the actual detection probability P with the equiprobable detection random number P s , and determine whether the target Target i is detected; If the equal-probability detection random number P s > the actual detection probability P, it means that the target i has not been detected. Update the target detection probability P i of this target during the current simulation cycle i to the historical highest detection probability P targeti , and execute step S1.6; If the equal-probability detection random number P s ≤ the actual detection probability P, it means that the target i is detected. Record the detection time t of the current simulation cycle as the target confirmation time start point t i of the target 0targeti and the most recent detection time t targeti , and record the target i in the list of detected targets of the current simulation cycle , and then execute step S1.6; S1.6, determine the target i Whether it is the last target within the field of view of the current simulation cycle; If not, return to step S1.1 to perform detection and resolution on the next target; If so, end the detection process and obtain the list of detected targets in the current simulation cycle K 3. The method for constructing a task-level simulation model for target detection and identification of a variable-focus optoelectronic pod according to claim 2, wherein: In step S1.3, the calculation of the target detection probability P i satisfies the following formula: E = 2.7 + 0.7(N / N 50 ) where: N is the target i The number of pixels occupied, N 50 is the minimum detectable pixel number, with a detection probability of 50%.

4. The method for constructing a task-level simulation model for target detection and identification of a variable-focus optoelectronic pod according to claim 3, wherein: In step S1.4, the target detection probability P is corrected by the probability correction function. i The calculation formula for the actual detection probability P obtained by processing is: Where: P is the actual detection probability, P i is the target detection probability of the i-th target Target i in the current simulation cycle, and P targeti is the historical highest detection probability of the i-th target Target i .

5. The method for constructing a task-level simulation model for target detection and recognition of a variable-focus optoelectronic pod according to claim 4, characterized in that: The specific steps of step S2 are as follows: S2.

1. Determine the list of detected targets in the current simulation cycle K Target j Is there a target in the list of identified targets in the previous simulation cycle K-1? The list of detected targets in the current simulation cycle K is The total number of targets in is J, j = 1, 2, 3, ..., J; If it exists, execute step S2.3; If it does not exist, calculate the current simulation time and the target Target j Confirm the time interval t between the time starting points and execute step S2.2; S2.

2. Set the average time required to identify and confirm a target as T recognize , and compare the time interval t calculated in step S2.1 with the average time T recognize Make a comparison; If the time interval t′ ≥ the average time T recognize , then the target j Included in the identification and confirmation target list T recognized At the same time, the list of detected targets is updated accordingly. The confirmation time starting point t of the remaining targets in 0targetj , execute step S2.3; If the time interval t'<average time T recognize , then execute step S2.3; S2.

3. Determine the target j whether it is the last target in the detected target list for the current simulation cycle; If not, return to step S2.1 to perform identification and confirmation resolution on the next target; If so, end the recognition and confirmation calculation process to obtain the recognition and confirmation target list T for the current simulation cycle recognized , thus completing the detection and recognition of ground targets by the zoom electro-optical pod 6. The method for constructing a task-level simulation model for target detection and identification of a variable-focus optoelectronic pod according to claim 5, wherein: In step S2.2, the confirmation time start point t of the remaining targets in the updated detected target list 0targetj is updated according to the following formula: t 0targetj = t 0targetj + T recognize .

Images