CN114578849A - Unmanned aerial vehicle target detection system, target detection method and computer readable storage medium - Google Patents

Abstract

The invention belongs to the technical field of automatic target detection, and discloses an unmanned aerial vehicle target detection system, a target detection method, and a computer-readable storage medium. The unmanned aerial vehicle system; the photoelectric load system used to perform fixed-point observation tasks and fixed observation of the shooting range area; the wireless image transmission system used for the transmission of observation video images using the HHLM transmission module; used for the unmanned aerial vehicle system As well as a wireless remote control system for the control of the photoelectric payload system and a target detection integrated task table for receiving, storing, forwarding and processing video images. The invention provides a target detection system at sea, which is of great significance to improving the level of testing technology and testing equipment, and improving the actual combat capability and training level of destroyers and naval guns. The need is very realistic.

Description

UAV target detection system, target detection method, and computer-readable storage medium

technical field

The invention belongs to the technical field of automatic target detection, and in particular relates to an unmanned aerial vehicle target detection system, a target detection method and a computer-readable storage medium.

Background technique

At present, since the beginning of the 21st century, maritime security issues have attracted more and more attention from various countries. Disputes over maritime territory and maritime rights and interests between neighboring countries, rampant pirate attacks, and safety issues such as freedom of navigation at sea have caused countries to gradually pay attention to the modernization of the navy. . Live ammunition is an important item in naval militarization training. In the competitive assessment of live ammunition, rapid and accurate evaluation of the effect of live ammunition is the key to completing the comprehensive evaluation of combat performance. The daily training level is judged by the hit rate and miss rate after the naval gun is fired, so it is very necessary to measure it accurately and scientifically. The marine shooting range is different from the land shooting range. The target test traces on the land range are easy to capture, while at sea, the projectile falling into the water to stir up a water column is a short process, and the test traces are fleeting and difficult to capture.

Therefore, there is no automatic target detection method applied to the marine shooting range in the prior art, and no method or system capable of realizing automatic and intelligent evaluation of the firing of naval guns at sea.

Through the above analysis, the existing problems and defects of the existing technology are as follows: the existing technology does not have an automatic target detection method applied to a marine shooting range, nor a method or system capable of realizing automatic and intelligent evaluation of the firing of naval guns at sea. The traditional optical measurement method requires manual alignment of two sets of image data, manual measurement, and there is an orthogonal error, the workload is large, and the professionalism is strong, which is not suitable for modern high-intensity training. At the same time, if the large-scale measurement and control equipment with advanced technology, complex structure and high precision in the shooting range is used to obtain data (impact deviation), the cost problem cannot be ignored.

The difficulty of solving the above problems and defects is as follows: Judging from the implementation process of the existing main gun short-range counterattack shooting results, there are still the following shortcomings:

(1) The use of support troops is large, and coordination is difficult. It is necessary to refer to the two support forces of the ship and the target ship, and to deploy multiple cameras, which means that multiple support personnel are required, so it must involve continuous coordination and communication between the support forces and the support personnel.

(2) The recording process requires high requirements and is complicated to implement. Because only the two videos from the reference ship and the target ship can be used to calculate the dispersion error of naval guns, and in order to ensure the consistency of the measurement targets in the video images, the time reference of the two cameras needs to be unified, and it is also necessary to ensure Start recording at the same time, otherwise it will be difficult to achieve objective post-event analysis based on video.

(3) The estimation of the dispersion error is slow and the accuracy is not high. For the post-processing of the video of the impacting water column, after confirming the consistency of the measurement target, use the scale to measure the distance between the impacting water column and the floating target on the computer display screen. , which is not only time-consuming and labor-intensive, but also has low precision.

The significance of solving the above problems and defects is as follows: in the traditional missed target assessment work, two orthogonal cameras are scattered from the shooting direction and distance of the shooting ship and the support ship respectively. After returning to the shore, the two sets of image data are manually aligned. Then measure the distance on the image to calculate the dispersion error, and evaluate the assessment results. The whole process is time-consuming and labor-intensive. The application prospect of the UAV target detection system is broad. It is not only suitable for the target detection task of naval guns shooting at sea, but also suitable for naval guns shooting at shore. With the continuous maturity of UAV target detection technology, the UAV target detection system will be gradually promoted to replace the currently used target detection system. At the same time, based on the video image, the evaluation of the missed target amount of the naval gun shooting was completed. The automatic target detection system of the naval gun shooting UAV was used to obtain the video image of the relative position of the water column of the naval gun shooting and the target from a fixed point at a high altitude, and then transmitted to the digital processing. Module, through digital image processing technology to complete the extraction and positioning of the target and the extraction and positioning of the impact water column, calculate the coordinate position of the impact water column, get the shooting dispersion error to evaluate the shooting score, and finally complete the GUI realization of the missed target evaluation system. . The evaluation time is shortened, the manual workload is reduced, and the evaluation accuracy and timeliness are improved.

SUMMARY OF THE INVENTION

In view of the problems existing in the prior art, the present invention provides a target detection system, a target detection method and a computer-readable storage medium for an unmanned aerial vehicle.

The present invention is realized in this way, an unmanned aerial vehicle target detection system, the unmanned aerial vehicle target detection system includes:

UAV system, photoelectric load system, wireless image transmission system, wireless remote control system and target detection integrated task platform;

UAV system, including quadrotor UAV body, propulsion device, flight control device and power supply device; used to carry the photoelectric load system to the observation position and hover at a fixed point;

The photoelectric load system is mounted on the body of the quadrotor UAV; it includes a CCD camera, an infrared detection device and a pod; it is used to perform fixed-point observation tasks and perform fixed observation of the shooting range area;

The wireless image transmission system is used to transmit the observation video image by using the HHLM type transmission module;

Wireless remote control system for control of UAV system and photoelectric load system;

The target detection comprehensive task platform is used for receiving, storing, forwarding and processing video images.

Further, the wireless remote control system includes:

Accepting module for accepting control instructions;

The decoding module is used to decode the control instruction by using the decoder;

The control module is used to control the UAV system and the optoelectronic load system based on the control instructions obtained by decoding.

Another object of the present invention is to provide a UAV target detection method applied to the UAV target detection system, and the UAV target detection method includes:

Use the photoelectric load to obtain the video image of the water column of the naval artillery, and obtain the video information, obtain the image sequence, play the video information, select the image frame by frame, input the rotation angle of the coordinate system, extract the water column of the projectile and other means. The video image of the water column is analyzed and processed to obtain the hit rate and miss rate of the naval gun shooting, and the shooting evaluation results are obtained.

Further, the UAV target detection method comprises the following steps:

Step 1, generating a control command, and controlling the quadrotor UAV equipped with the photoelectric load to fly to a safe height above the target based on the generated control command, and then hovering; at the same time, based on the generated control command, the UAV is controlled to adjust the photoelectric load shooting attitude;

Step 2, control the photoelectric load based on the control command to record the shooting situation of naval guns in the area of the marine shooting range, continuously obtain video images of the water column caused by the projectile entering the water, and obtain the relative relationship between the impacted water column and the target in the shooting range, that is, the dispersion of naval gun shooting;

Step 3: Convert the acquired video image into an electrical signal, and convert it into a digital signal after sampling, quantization, and encoding using an analog-to-digital converter;

Step 4: Preprocess the converted image to obtain clear and useful image information; and calculate the relative position of the impact water column and the floating target based on the obtained clear and useful image, and calculate the projectile dispersion error, and then calculate the amount of misses by the naval gun. Evaluate.

Further, the safe height calculation method includes:

First, analyze the hovering height of the UAV based on the firing height of the naval gun:

H ≥ h;

Among them, H represents the hovering height of the drone; h represents the height of the projectile when it reaches the highest point; the height h of the projectile when it reaches the highest point is determined by the relationship between the firing angle, firing range and firing height of the naval gun;

Secondly, analyze the hovering height of the UAV based on the observation range of the photoelectric load: according to the type of artillery, the shooting method and the average shooting distance, and according to the corresponding table of the evaluation parameters of the shooting performance of the floating target and the simulated target, determine the range of the judgment interval; The rectangular boundary of the judgment interval determines the hovering height of the UAV; the rectangular boundary of the judgment interval means that when the naval gun shoots at the floating target or the simulated target, according to the corresponding shooting performance evaluation parameter table, it is judged that the shell falls into the effective hit area. Rectangular interval boundary;

Finally, the safe height of the UAV is determined based on the analysis results of the UAV's hovering height, the rectangular boundary limit of the judgment interval, the range of the camera's field of view, and the UAV's own flight height limit.

Further, the preprocessing of the converted image to obtain clear and useful image information includes:

(1) The converted image is segmented by a segmentation method based on color features, and the target area is identified and extracted;

(2) Using the least squares method to perform curve fitting based on the target information obtained after segmentation to determine the center and radius of the target to locate the target;

(3) Grayscale, median filtering, and contrast enhancement are performed on the extracted and positioned target images.

Further, the calculation of the relative position of the impact water column and the floating target based on the obtained clear and useful image includes:

The threshold value is calculated by adaptive and iterative methods to segment the impact water column image to obtain the impact water column image after removing the background; binarize the obtained impact water column image after removing the background; at the same time, the improved centroid method is used. Perform the positioning of the impacting water column.

Further, the calculation of thresholds by adaptive and iterative methods to segment the water column image includes:

1.1) Count the minimum gray value T _min and the maximum gray value T _max of the water column image, and calculate the binary average value as the initial threshold T:

1.2) Segment the image according to the threshold T, and obtain two pixel sets:

G ₁ ={f(x,y)≥T}, G ₂ ={f(x,y)≤T};

1.3) Calculate the grayscale average values μ ₁ and μ ₂ of the pixel sets G ₁ and G ₂ :

重复步骤1.2)、步骤1.3)、步骤1.4)，直至阈值T收敛到某一范围为止。1.4) Calculate a new threshold based on μ ₁ and μ ₂

Repeat steps 1.2), 1.3), and 1.4) until the threshold T converges to a certain range.

Further, the positioning of the impacting water column using the improved centroid method includes:

The edge detection of the impacting water column is performed, the edge information of the impacting water column is extracted, and the centroid of the impacting water column is calculated by arithmetic mean based on the edge information of the impacting water column.

Further, in step 4, the projectile dispersion error is obtained by the calculation, and the evaluation of the miss distance of naval gun shooting includes:

(1) Calculate the dispersion error:

Based on the obtained position information of the target and the impact water column in the image, the impact dispersion coordinate system with the target as the origin is established, and the dispersion error is calculated by the following formula:

式4.25中

(X_i，Z_i)表示每一个弹着水柱的实际坐标值，

表示弹着水柱群中心坐标，

的值为所有弹着水柱实际坐标值的平均值；0.6745表示或然系数，n表示弹丸总数；in,

In Equation 4.25

(X _i , Z _i ) represents the actual coordinate value of each impacted water column,

Represents the coordinates of the center of the impacting water column group,

The value of is the average value of the actual coordinate values of all impacting water columns; 0.6745 represents the probability coefficient, and n represents the total number of projectiles;

(2) Comparing the calculated values of E _X and E _Z with the standard value K of the impact point dispersion error to determine the shooting results of the naval guns, and obtain the evaluation results of the shooting misses of the naval guns.

Another object of the present invention is to provide a computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, the processor enables the processor to perform the function of the UAV target detection system

Combining all the above technical solutions, the advantages and positive effects of the present invention are: the present invention uses the naval gun shooting UAV automatic target detection system to obtain the video image of the relative position of the water column of the naval gun shooting and the target from a fixed point at a high altitude, It is transmitted to the digital processing module, and the extraction and positioning of the target and the extraction and positioning of the impact water column are completed through digital image processing technology, the coordinate position of the impact water column is calculated, and the shooting dispersion error is obtained. It is of great significance to the improvement of the actual combat capability and training level of the destroyer and frigate naval guns.

The invention adopts the HHLM type wireless image transmission system, and the longest transmission distance can reach 50 kilometers. HHLM microwave image transmission system is a high-performance and high-quality wireless image transmission system specially designed for long-distance or environments without wired transmission conditions. The invention itself is small in size and light in weight, can transmit high-quality video images in real time without distortion, has stable modulation and demodulation performance, bright and clear transmitted images, and is easy to install and debug.

Description of drawings

FIG. 1 is a schematic diagram of a UAV target detection system provided by an embodiment of the present invention.

2 is a schematic structural diagram of a UAV target detection system provided by an embodiment of the present invention;

In the picture: 1. UAV system; 2. Photoelectric load system; 3. Wireless image transmission system; 4. Wireless remote control system;

FIG. 3 is a schematic diagram of a method for detecting a target of an unmanned aerial vehicle provided by an embodiment of the present invention.

FIG. 4 is a flow chart of a method for detecting a target of an unmanned aerial vehicle provided by an embodiment of the present invention.

FIG. 5 is a schematic diagram of the position of a target detection drone provided by an embodiment of the present invention.

FIG. 6 is a graph of a shooting elevation angle and a shooting distance provided by an embodiment of the present invention.

FIG. 7 is a graph of a shooting elevation angle and a shooting height provided by an embodiment of the present invention.

FIG. 8 is a shooting curve diagram of different shooting elevation angles provided by an embodiment of the present invention.

FIG. 9 is a schematic diagram of a determination interval provided by an embodiment of the present invention.

FIG. 10 is a schematic diagram of detection by a target detection system provided by an embodiment of the present invention.

FIG. 11 is a schematic diagram of pinhole imaging provided by an embodiment of the present invention.

FIG. 12 is a flowchart of the automatic extraction and positioning of the impact water column provided by an embodiment of the present invention.

FIG. 13 is a grayscale histogram showing a single peak provided by an embodiment of the present invention.

FIG. 14 is a schematic diagram of an image coordinate system provided by an embodiment of the present invention.

FIG. 15 is a schematic diagram of the principle of calculating the coordinates of the water column of impact provided by an embodiment of the present invention.

FIG. 16 is a schematic diagram of a projectile dispersion coordinate system provided by an embodiment of the present invention.

FIG. 17 is a plane position relationship diagram of a shooting ship and an unmanned aerial vehicle provided by an embodiment of the present invention.

Detailed ways

In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention, but not to limit the present invention.

Aiming at the problems existing in the prior art, the present invention provides an unmanned aerial vehicle target detection system. The present invention will be described in detail below with reference to the accompanying drawings.

As shown in FIG. 1-FIG. 2, the UAV target detection system provided by the embodiment of the present invention includes:

UAV system 1, photoelectric load system 2, wireless image transmission system 3, wireless remote control system 4 and target detection integrated task platform 5;

UAV system 1, including a quadrotor UAV body, a propulsion device, a flight control device and a power supply device; it is used to carry the photoelectric load system to the observation position and hover at a fixed point;

The optoelectronic load system 2 is mounted on the body of the quadrotor UAV; it includes a CCD camera, an infrared detection device and a pod; it is used to perform fixed-point observation tasks and perform fixed observation of the shooting range area;

The wireless image transmission system 3 is used to transmit the observation video image by using the HHLM type transmission module;

The wireless remote control system 4 is used to control the UAV system and the photoelectric load system;

The target detection comprehensive task platform 5 is used for receiving, storing, forwarding and processing video images.

The wireless remote control system 4 provided by the embodiment of the present invention includes:

Accepting module for accepting control instructions;

The decoding module is used to decode the control instruction by using the decoder;

The control module is used to control the UAV system and the optoelectronic load system based on the control instructions obtained by decoding.

As shown in FIG. 3 , the UAV target detection method provided by the embodiment of the present invention includes:

Use the photoelectric load to obtain the video image of the water column of the naval artillery, and obtain the video information, obtain the image sequence, play the video information, select the image frame by frame, input the rotation angle of the coordinate system, extract the water column of the projectile and other means. The video image of the water column is analyzed and processed to obtain the hit rate and miss rate of the naval gun shooting, and the shooting evaluation results are obtained.

As shown in FIG. 4 , the UAV target detection method provided by the embodiment of the present invention includes the following steps:

S101 , generating a control command, and controlling the quadrotor UAV equipped with the photoelectric load to fly to a safe height above the target based on the generated control command, and then hovering; at the same time, based on the generated control command, the UAV is controlled to adjust the photoelectric load shooting attitude;

S102, control the photoelectric load based on the control command to record the shooting situation of the naval guns in the area of the marine shooting range, continuously obtain the video images of the water column caused by the projectile entering the water, and obtain the relative relationship between the impacting water column and the target in the shooting range, that is, the dispersion of the shooting of the naval guns;

S103, convert the acquired video image into an electrical signal, and convert it into a digital signal after sampling, quantization, and encoding using an analog-to-digital converter;

S104, preprocess the converted image to obtain clear and useful image information; and calculate the relative position of the impacting water column and the floating target based on the obtained clear and useful image, and calculate the projectile dispersion error, and evaluate the missed target amount of naval gun shooting .

The safety height calculation method provided by the embodiment of the present invention includes:

First, analyze the hovering height of the UAV based on the firing height of the naval gun:

H ≥ h;

Among them, H represents the hovering height of the drone; h represents the height of the projectile when it reaches the highest point; the height h of the projectile when it reaches the highest point is determined by the relationship between the firing angle, firing range and firing height of the naval gun;

Secondly, analyze the hovering height of the UAV based on the observation range of the photoelectric load: according to the type of artillery, the shooting method and the average shooting distance, and according to the corresponding table of the evaluation parameters of the shooting performance of the floating target and the simulated target, determine the range of the judgment interval; The rectangular boundary of the judgment interval determines the hovering height of the UAV; the rectangular boundary of the judgment interval means that when the naval gun shoots at the floating target or the simulated target, according to the corresponding shooting performance evaluation parameter table, it is judged that the shell falls into the effective hit area. Rectangular interval boundary;

Finally, the safe height of the UAV is determined based on the analysis results of the UAV's hovering height, the rectangular boundary limit of the judgment interval, the range of the camera's field of view, and the UAV's own flight height limit.

The preprocessing of the converted image provided by the embodiment of the present invention to obtain clear and useful image information includes:

(1) The converted image is segmented by a segmentation method based on color features, and the target area is identified and extracted;

(2) Using the least squares method to perform curve fitting based on the target information obtained after segmentation to determine the center and radius of the target to locate the target;

(3) Grayscale, median filtering, and contrast enhancement are performed on the extracted and positioned target images.

The calculation of the relative position of the impact water column and the floating target based on the obtained clear and useful image provided by the embodiment of the present invention includes:

The threshold value is calculated by adaptive and iterative methods to segment the impact water column image to obtain the impact water column image after removing the background; binarize the obtained impact water column image after removing the background; at the same time, the improved centroid method is used. Perform the positioning of the impacting water column.

The embodiment of the present invention provides an adaptive and iterative method to calculate the threshold to segment the water column image, including:

1.1) Count the minimum gray value T _min and the maximum gray value T _max of the water column image, and calculate the binary average value as the initial threshold T:

1.2) Segment the image according to the threshold T, and obtain two pixel sets:

G ₁ ={f(x,y)≥T}, G ₂ ={f(x,y)≤T};

1.3) Calculate the grayscale average values μ ₁ and μ ₂ of the pixel sets G ₁ and G ₂ :

重复步骤1.2)、步骤1.3)、步骤1.4)，直至阈值T收敛到某一范围为止。1.4) Calculate a new threshold based on μ ₁ and μ ₂

Repeat steps 1.2), 1.3), and 1.4) until the threshold T converges to a certain range.

The positioning of the impacting water column using the improved centroid method provided by the embodiment of the present invention includes:

The edge detection of the impacting water column is performed, the edge information of the impacting water column is extracted, and the centroid of the impacting water column is calculated by arithmetic mean based on the edge information of the impacting water column.

The calculation provided by the embodiment of the present invention obtains the projectile dispersion error, and the evaluation of the miss distance of naval gun shooting includes:

(1) Calculate the dispersion error:

Based on the obtained position information of the target and the impact water column in the image, the impact dispersion coordinate system with the target as the origin is established, and the dispersion error is calculated by the following formula:

式4.25中

(X_i，Z_i)表示每一个弹着水柱的实际坐标值，

表示弹着水柱群中心坐标，

的值为所有弹着水柱实际坐标值的平均值；0.6745表示或然系数，n表示弹丸总数；in,

In Equation 4.25

(X _i , Z _i ) represents the actual coordinate value of each impacted water column,

Represents the coordinates of the center of the impacting water column group,

The value of is the average value of the actual coordinate values of all impacting water columns; 0.6745 represents the probability coefficient, and n represents the total number of projectiles;

(2) Comparing the calculated values of E _X and E _Z with the standard value K of the impact point dispersion error to determine the shooting results of the naval guns, and obtain the evaluation results of the shooting misses of the naval guns.

The technical solutions of the present invention will be further described below with reference to specific embodiments.

Example 1:

1. Composition and working principle of UAV target detection system

1.1 System Composition

The components of the UAV target detection system include the UAV system, the photoelectric load, the wireless image transmission system, the wireless remote control system and the target detection integrated task table.

UAV, referred to as UAV, is an unmanned aerial vehicle that is operated by radio remote control equipment and its own program control device. The UAV platform system is the air flight part of the UAV system, including the aircraft body, propulsion device, flight control device, and power supply device. The airborne part of the communication data link, the mission payload, is also installed on the aircraft. The present invention selects four-rotor unmanned aerial vehicle, and uses aviation lithium battery to provide power. Mainly considering that the quadrotor UAV is widely used, it has a good technical foundation, low cost, good maneuverability, strong survivability, and low environmental requirements. It can overcome the observation work that is difficult to carry out manually, and is easy to use. Therefore, it is convenient to hover and record the water column video of the impact point. In this system, the optical measuring platform can be carried to complete the fixed-point observation task in the air, and the fixed observation of the shooting range area can be realized.

The optoelectronic payload consists of a CCD camera, an infrared detection device and a pod. CCD (Charge Coupled Device, charge-coupled device image sensor) is made of a high-sensitivity semiconductor material, which can convert light into electric charges, and convert them into digital signals through an analog-to-digital converter chip. The flash memory inside the camera or the built-in hard disk card is stored, so data can be easily transferred to a computer. It is mounted on a quad-rotor UAV to record the shooting situation of naval guns in the area of the marine shooting range, continuously shoot the water column caused by the projectile entering the water, and obtain the relative relationship between the impacting water column and the target in the shooting range, that is, the dispersion of naval gun shooting.

The wireless image transmission system is used for the timely transmission of images. It converts the video image information captured by the UAV photometric platform into electrical signals, which are sampled, quantized and encoded by the analog/digital converter and converted into digital signals, which are sent to the dynamic The memory is temporarily stored, and then the effective data is wirelessly sent to the information receiving and processing system in real time through the wireless data transmission chip for later image processing and data operation. The invention adopts the HHLM type wireless image transmission system, and the longest transmission distance can reach 50 kilometers. HHLM microwave image transmission system is a high-performance and high-quality wireless image transmission system specially designed for long-distance or environments without wired transmission conditions. The invention itself is small in size and light in weight, can transmit high-quality video images in real time without distortion, has stable modulation and demodulation performance, bright and clear transmitted images, and is easy to install and debug.

The function of the wireless remote control system is to realize the control of the UAV and the photoelectric load. The ground control center sends out instructions, which are transmitted to the receiving module of the wireless remote control system, and the instructions are read out through the decoder to realize the control of the UAV and the control of the photoelectric load.

Target detection comprehensive task platform, whose task is to complete the receiving, storage and forwarding of video images, and effectively process the video images, mainly composed of a PC and some image processing and numerical calculation software. After receiving the information from the image transmission system, the image is effectively preprocessed, interference and noise are removed, clear and useful image information is taken out, and the relative position of the impacting water column and the floating target is calculated according to a reasonable mathematical model. , and calculate the projectile dispersion error.

1.2 Working principle

The traditional evaluation method is relatively cumbersome, with large accidental errors, and the workload of security personnel is large. The automatic target detection system of naval gun shooting UAV can change the drawbacks of traditional target detection work from the acquisition of video image information to the later data processing, reduce working time and improve efficiency. The basic process of system work is roughly divided into three steps, as shown in Figure 1. .

(1) As a platform for carrying photoelectric loads, the UAV carries the photoelectric load and flies to a safe height above the target and then hovers.

(2) The photoelectric load of the UAV adjusts the shooting attitude to carry out the work, and the target is imaged on the target surface by the camera.

(3) The video image information of the water column of naval artillery shells is transmitted to the integrated task platform of the target ship in real time through the wireless transmission module, and is stored and forwarded in real time, and then transmitted to the image acquisition and display board of the video signal through the wireless image transmission system. Digitalization, through the development of comprehensive task software, the video images are processed quickly and accurately, the hit rate and miss rate of naval gun shooting are obtained, and the shooting results are evaluated.

2. Analysis of the target detection task area of the UAV system

In order to ensure the safety of UAVs and meet the requirements of target inspection tasks, it is necessary to define the UAV target inspection task area. When the UAV system inspects the target, the UAV hovers directly above the floating target, so the UAV target inspection task area is the area directly above the floating target, and the main job is to determine the height of the area.

2.1 Analysis of the hovering height of UAV based on the firing height of naval guns

In the process of naval gun firing, the trajectory of the projectile after it exits the barrel is approximately parabolic. When the UAV target detection system is working, the UAV hovers right above the floating target. In order to ensure the safety of the UAV, the height to ensure the safety of the UAV should be determined according to the firing range of the naval gun to prevent unmanned aerial vehicles from being unmanned. The machine was shot down by a projectile. The location design idea of the target detection UAV is shown in Figure 5. It can be seen from Figure 5 that the UAV is almost not threatened by the horizontal direction. The safety threat of the UAV mainly comes from the projectile reaching the top of the floating target. Therefore, the choice of the hovering height of the UAV depends on the shooting process of the naval gun. It is very important, once the wrong location is selected, it will cause serious accidents and make the target inspection work fall into a passive situation.

Analyze the relationship between the firing angle, firing range and firing height of naval guns to find the best hovering height.

It is known that the initial velocity of the 76mm main gun projectile out of the gun barrel is 980 m/s, the gravitational acceleration g = 9.8, the shooting elevation angle is α, and the range of the known naval gun shooting angle is -15°～85°. Using Matlab software to simulate, get the curves of shooting elevation angle and shooting distance, shooting elevation angle and shooting height, and shooting distance and shooting height with different shooting elevation angles. safety range.

First, use Matlab to simulate, and get the curve graph of shooting elevation angle and shooting distance as shown in Figure 6. From the graph of shooting elevation angle and shooting distance in Fig. 6, it can be seen that when the shooting elevation angle is 45°, the shooting distance of the naval gun is the largest; under the condition that the distance between the ship and the floating target is 3 nautical miles (5.556㎞), and the shooting elevation angle is 0 ° ~ 3 °, the shooting distance is within 10 kilometers, which can meet the shooting needs under this condition.

Then use Matlab to simulate, and get the curve diagram of shooting elevation angle and maximum shooting height, as shown in Figure 7. It can be seen from the graph of elevation angle and distance in Figure 744 that as the elevation angle of the shooting increases, the maximum shooting height also increases; between 0° and 10°, the maximum shooting height does not change significantly.

According to the analysis results of the curve graph of shooting elevation angle and shooting distance in Figure 6 and the curve diagram of shooting elevation angle and maximum shooting height in Figure 7, and using Matlab to simulate, the corresponding curves of projectile shooting distance and shooting height at different shooting elevation angles are obtained. The curve The diagram is shown in Figure 8.

According to the shooting curves of different shooting elevation angles in Fig. 8 and the simulation results, the parameters of the maximum height that the projectile can reach and the height when it reaches the top of the floating target under different shooting elevation angles are obtained. The parameter table is shown in Table 1.

Table 1 Projectile shooting distance and height parameter table

According to different shooting elevation angles, projectile shooting distance, height parameter table and graph, the dangerous area of naval gun shooting can be obtained intuitively, so the firing range of naval guns in height can be obtained. According to the needs, the firing distance of the naval gun is 3 nautical miles (5.556㎞). The firing range of the naval gun can be obtained from the projectile firing distance and height parameter table in Table 1 and the firing curves of different firing elevation angles in Figure 7. The range of the firing range is based on the launching ship, and the height is 1.5 times the maximum height of naval gun firing. According to the above information, it can be concluded that the maximum shooting height is 40 meters, then the minimum height of the firing range of the naval gun in height is 1.5 times the maximum shooting height, that is, 40×1.5=60 meters.

Continue to calculate the minimum safe altitude of the drone. When the drone inspects the target, the drone hovers directly above the floating target. It is known that the trajectory of the projectile is approximately a parabola. It is assumed that the height of the projectile when it reaches the highest point is h, and the hovering height of the drone is H. In order to ensure the safety of the drone, H ≥ h. Under the condition that the distance between the ship and the floating target is 3 nautical miles (5.556㎞), the minimum firing range of the 76mm naval gun is 60 meters. The safe hovering height is 60 meters, and the drone is directly above the floating target.

2.2 UAV hovering height analysis based on photoelectric load observation range

The hovering height of the UAV based on the observation range of the photoelectric load refers to the height that meets the requirements of the shooting effect and the shooting image contains the elements required for target detection. Therefore, when determining the working height of the UAV, it is not enough to only consider the height that ensures the safety of the UAV, but also the working height of the UAV that can meet the requirements of the shooting task.

The concept of "decision matrix" is introduced here. The rectangular boundary of the "judgment interval" is the rectangular interval boundary of the projectile that is judged to fall into the effective hit area according to the corresponding shooting performance evaluation parameter table when the naval gun shoots at the floating (reflector) target and the simulated target.

According to the type of artillery, the shooting method and the average shooting distance, the range of the "judgment interval" is determined according to the corresponding table of the evaluation parameters of the shooting performance of the floating (reflector) target and the simulated target. The schematic diagram of the "judgment interval" is shown in Figure 9.

In FIG. 9 , 2X is the distance side length value of the rectangular determination section, and 2Z is the direction side length value of the rectangular determination section. In the evaluation parameter table for the shooting performance of floating targets in the sea, the half values X and Z of the distance and direction side length of the rectangular judgment area are listed. Table 2 shows the evaluation parameters for the shooting performance of floating targets in the sea.

Table 2. Parameters for evaluating the shooting performance of floating targets in the sea

According to this shooting task, by comparing the parameter table, we can get the boundary value of the "judgment interval" rectangle when the shooting distance is 3 nautical miles (5.556㎞), the distance X=114.43 meters, and the direction Z=32.651 meters.

Considering the shooting angle of the camera and the shooting effect, the hovering height of the drone is also determined according to the boundary of the "judgment interval" rectangular box. The design scheme of the target detection system detection is shown in Figure 10. It is known that the field of view of the camera is 94° (ie ADB=94°), then half the angle of view is ACO=BCO=1/2ACB=47°. During the evaluation of live ammunition of naval guns, the effective area for evaluation is the area of the "judgment interval" rectangular frame centered on the floating target O, and the "judgment interval" rectangular frame is inscribed with ⊙O.

According to the above known conditions, the height of the OC can be easily calculated to determine the hovering height of the UAV. Using the Pythagorean theorem: c ² =a ² +b ²

Where a=X=114.43 meters

b=Z=32.651 meters

have to

That is, the radius of ⊙OD is equal to 119 meters, OA=OB=OD=OE=OF=OG=119 meters.

According to the field of view of the camera, the length of OC can be obtained, using the tangent theorem:

Where a=OA=119 meters

θ=∠ACO=47°

have to

That is, OC=b=111 meters. Therefore, the hovering height of the drone is 111 meters.

If the water column caused by the projectile falling outside the "judgment interval" rectangle is observed, the visible range of the camera should be expanded. Assuming that the length and width of the visible area are 1.5 times the length and width of the "judgment interval" rectangular frame, under this condition, the boundary value of the rectangular frame is the distance X=1.5*X=171.645 meters, and the direction Z=Z=48.98 meters. According to the method of solving the hovering height of the UAV in the rectangular box area of the "judgment interval", the hovering height of the UAV under this condition can be obtained in the same way. Finally, the solution obtains OC=166.5 meters, that is, the hovering height of the drone is 166.5 meters.

The ultimate flight height of the drone is 500 meters. The height of the safety of the drone, the rectangular boundary limit of the "judgment zone", the range of the camera's field of view, and the flight height limit of the drone are finally set. The hovering height when the machine is working is 111 meters.

3. Target extraction and positioning and impact water column measurement

In order to realize the evaluation of the missed target amount of naval gun shooting based on video images, it is necessary to extract the target image information and the impact water column image information from the impact water column image. The target that the naval gun shoots at the sea is a red spherical floating target with a diameter of three meters, which has obvious contrast with the color of the ocean background in the image. Therefore, the color area of interest can be segmented directly by the RGB image of the water column obtained by the drone. , that is, to extract and locate the target.

In order to extract and locate the impact water column, it is necessary to preprocess the input impact water column image first to obtain clear target information. In this chapter, the extraction and positioning method of the target, the relationship between the image distance and the actual distance are first described, and the impact water column image is preprocessed to obtain clearer impact water column information. Then, a threshold segmentation method different from the extraction target is used to extract the impact water column, which realizes the automatic extraction and positioning of the impact water column under simple and complex backgrounds.

3.1 Target extraction and positioning and data preprocessing

Image segmentation is an important step in image processing. It directly determines the accuracy of subsequent positioning. First, the target should be extracted and the target should be separated from the background from the image. Since the red drift target used in naval gun shooting is obviously different from the ocean background, according to the consistency of the target color and the large contrast between the target and the background color, the target is directly divided in the RGB image. Extract it. Therefore, the present invention adopts the segmentation method based on color feature to segment the image.

The discrimination rule adopted by the present invention for target extraction is to discriminate a certain pixel as red according to when the R component in R, G, and B is obviously not smaller than other components, and set a threshold to control the colour of the discrimination condition. The area-based image segmentation mentioned in the present invention is the most critical and the most difficult to define strictly. For RGB color determination, the larger the threshold is set, the smaller the variation range of the target color will be. The tolerance will become smaller; if the threshold is set smaller, the range of target color changes will increase, and the possible segmented targets will increase or segment irrelevant areas.

Subsequently, the position of the target needs to be further determined, and the relative positional relationship between the impact point and the target can be determined only after the position of the target is determined, and the dispersion error can be calculated. The target information obtained by image segmentation is an approximate circle, and the positioning of the target is to obtain the center and radius of the target.

Circle detection is a very important application in digital image processing. There are many methods for extracting the center of a circle or a circle-like object, such as three-point circle method, Hough transform method, curve fitting method, etc.

(1) The three-point circle method is an external area boundary representation method, and the algorithm only pays attention to the boundary information of the target circular area. The disadvantage of this algorithm in practical application is that any three points on the boundary of the target circular area are selected, and the arbitrary selection will lead to strong randomness of the calculation, which has a great influence on the calculation result. But its advantage is that the algorithm can process incomplete and partially defective target circular areas, and the algorithm is relatively simple, easy to code and implement, and requires less computation.

(2) Hough transform method is a kind of clustering classification idea, which performs clustering operation on the pixel information with certain relationship in the image space, and finds the parameter space accumulation corresponding point that can connect these pixels in a certain analytical form. The advantages of this algorithm are strong anti-noise ability and high calculation accuracy. But its disadvantage is: the amount of data that needs to be stored is large, which leads to a slower operation speed of the algorithm.

(3) The surface fitting method is also a method of representing the boundary of the external area. It is usually based on the least squares method to find the best fitting point set and solve it iteratively. The advantage of this algorithm is that the accuracy is high, the influence of each boundary point pixel of the target circular area is considered, and the problem that the three-point selection is too flexible and the randomness is difficult to control in the three-point circle method is improved. However, the disadvantage of this algorithm is that when the target circular area is large and the boundary pixels of the target area are many, the calculation amount of the algorithm is very large, which is not conducive to the rapid positioning of the center of the circle and affects the execution efficiency of the algorithm.

Based on the comparison and analysis of the above methods for extracting the center of various circles or circle-like circles, since the number of pixels of the target after segmentation based on the regional image during the target extraction process is not many, and the target will be partially covered by the waves, in order to satisfy the system's accuracy of the target position To meet the requirements of speed and accuracy, the present invention uses the least squares method to perform a curve fitting algorithm for the detection of the target center.

After a segment of arc is found, the corresponding parameters are calculated by the equation of the circle, and the equation of the circle is shown in Equation 3.1.

(xx _c ) ² +(yy _c ) ² =r ² (Equation 3.1)

where (x _c , y _c ) is the center of the circle, and it is expanded to obtain Equation 3.2.

x ² +y ² +ax+by+c=0 (Equation 3.2)

Among them, a=-2x _c , b=-2y _c , c=x _c ² +y _c ² -r ² , use the least squares method to calculate the parameters a, b, c, so as to obtain the radius of the circle:

The target is fitted according to the coordinates and radius of the target center obtained by the least squares method, and the obtained radius fits well with the radius of the target.

After the target extraction and positioning is completed, the acquired image information must be preprocessed before the water column is extracted. Image preprocessing is the basic process of image processing. The water column video image after shooting is obtained through the acquisition system. In order to measure the miss of the naval gun, the water column of the target is extracted, and some difficulties in the image are overcome, especially for the image shot. When the difference between the impacting water column and the background is small, the extraction of the impacting water column target is easier, so the impacting water column image should be pre-processed.

The process of image preprocessing mainly includes image grayscale, median filtering for noise elimination and image contrast enhancement, etc., to prepare for the identification of the impacting water column.

3.2 The relationship between image distance and actual distance

To evaluate the miss distance of the shell, it is necessary to calculate the impact deviation according to the actual distance. According to the working requirements of the system, the photometric equipment shoots vertically downwards from a high altitude to obtain images including the entire area of the marine shooting range and the water column of the marine buoys. Based on the principles of perspective geometry and photogrammetry, the relationship between the pixels in the captured image and the actual distance is derived and calculated.

The invention calculates according to the imaging principle of the pinhole camera, and considers that the image acquisition of the shooting range area by the camera is proportional to the scaling, and does not consider the error caused by the image distortion to the measurement distance, because the image hitting the water column is beyond the edge of the principle image, which is affected by the lens distortion. less impact.

As shown in Figure 11, O is the center of the lens, a is the object distance, which is the distance from the optical center O to the object plane, and f is the focal length of the lens, which is the distance from the optical center O to the image plane.

It is known that the actual target is a sphere with a diameter of three meters, which can be used as a reference object for non-contact distance measurement. According to the number of pixels occupied by the target in the image, the actual length represented by each pixel can be obtained, that is, the size of the proportional relationship e .

Calculate the distance between the impacting water column and the target, and obtain the actual distance by calculating the Euclidean distance on the images of the centers of the two targets, according to the proportional relationship between the pixel size and the length of the actual object. So the relationship between image pixels and actual distance is

The center coordinates and radius of the target have been determined in the previous description, the target radius is 14.1032 pixels, and the actual radius of the target is 1.5 meters, so e=0.1064 meters/pixel.

3.3 Automatic extraction and positioning of the impact water column

The so-called simple background is ideally an environment that is more favorable for the recognition of the impacting water column, and the ideal condition is the ocean background with only one grayscale in the impacting water column image. If there is a sky background, the gray level of the sky is close to the gray level of the water column, which will make the target segmentation more difficult; if the weather conditions are good and the wind speed is small, the shape of the impacting water column will basically remain a cylinder, and the impact point position can be divided by the water column. The center is representative, and the sea conditions are good, there are less surges on the sea surface, and there is less interference with the target extraction. The impact water column image obtained in this case is called the simple background impact water column image, and the digital image processing is efficient and the error is small. Figure 12 shows the design of the automatic extraction and positioning process of the impacting water column under a simple background.

When a naval gun is fired at the sea, a water column is generated after the projectile enters the water, so the impact deviation can be determined according to the position of the impact water column. The extraction of the impacting water column mainly separates the target from the ocean background, and then filters out the influence of the waves in the ocean. Due to the difference in the contrast between the water column and the background, the target and the background are segmented by a predetermined grayscale threshold.

Suppose the image f(x, y) has some bright objects on a dark background, so the gray levels of object and background pixels are divided into two main modes. A common way to extract objects from the background is to choose a threshold T to separate the two modes. Then, any point (x, y) that satisfies the condition f(x, y) > T is called an object point, and other points are called background points (and vice versa for dark objects on a bright background). The definition of the thresholded (binary) image g(x, y) is as in Equation 4.1.

Therefore, the most difficult part of threshold segmentation is how to determine the threshold T. Threshold calculation methods are usually divided into two types: global threshold and basic adaptive threshold.

The global threshold is the most commonly used threshold calculation method, especially when the gray histogram distribution of the image presents two peaks, the global threshold can obviously divide the target and background components to obtain a more ideal image segmentation effect. The basic adaptive threshold is a relatively basic image adaptive segmentation method. It generally performs threshold segmentation based on the characteristics of the image pixel itself and the grayscale change of its neighborhood, and then realizes the binarization of the grayscale image. This method fully considers the characteristics of each pixel neighborhood, so it can generally better highlight the boundaries of the target and the background.

The background of the bouncing water column is the sea, and the color of the water column is white. Ideally, the contrast between the impacting water column and the background is relatively large, and the maximum inter-class difference method can be used to separate the impacting water column and the background. However, due to the influence of the environment during image acquisition, such as the reflection of the ocean surface, the shooting angle of the photometric platform, and the density of the water column, the grayscale histogram shows a single peak state. Figure 13 shows the grayscale of the water column. Histogram, which exhibits a unimodal state.

According to the characteristics of the single-peak grayscale histogram, the adaptive threshold calculation steps adopted by the present invention are as follows:

(1) Initial value. The minimum gray value T _min and the maximum gray value T _max of the water column image are counted, and the binary average value is calculated as the initial threshold value T.

(2) Segmentation. The image is segmented according to the threshold T, and two sets of pixels are obtained.

G ₁ ={f(x,y)≥T}, G ₂ ={f(x,y)≤T} (Equation 4.6)

(3) Mean. The grayscale average values μ ₁ and μ ₂ of the pixel sets G ₁ and G ₂ are calculated.

重复步骤2、3、4，直至阈值T收敛到某一范围为止。(4) Iteration. Calculate a new threshold based on μ ₁ and μ ₂

Repeat steps 2, 3, and 4 until the threshold T converges to a certain range.

The water column images are segmented by calculating thresholds by adaptive and iterative methods.

After the image is binarized by the above methods, the segmentation of the impacting water column and the background is better achieved, but there are still some interference noises. These noises are caused by the closeness of the gray level of some areas in the background and the impacting water column. The false small target points formed after binarization, or the uneven grayscale of the hitting water column, also cause some interference noise in the water column area in the binary image. The existence of these noises affects the location of the impact point in the next extraction of the impact water column. Aiming at these problems, the binary image is filtered by mathematical morphological operations.

Mathematical morphology theory differs from the traditional view on numerical modeling and analysis, mainly using structural elements to analyze and detect images. Mathematical morphology consists of a series of algebraic operations in morphology, including four basic operations: Dilation, Erosion, Opening, and Closing.

The expansion operation merges all background points that an object touches into the object and expands the bounds of the target object to the outside. Using this operation, the space in the object can be filled; the erosion operation can eliminate the target boundary points and shrink the target boundary inward, and the small and meaningless objects in the image can be eliminated by this operation. The filtering of the binary image realizes the elimination of noise through two stages: the first stage is to perform the closing operation on the binary image, the purpose is to eliminate the black dots in the impacting water column, so that the impacting water column area is more complete. The second stage is to open the image, the purpose is to remove the small interfering noise in the background and remove the burr on the edge of the water column, and smooth the edge of the water column.

After obtaining the image of the impacting water column, it is necessary to obtain the center coordinates of the impacting water column. After the previous image processing, the interference noise in the image is basically eliminated, and the water column can be used for positioning. Because the impacting water column is affected by gravity and wind, the image captured from high altitude is not necessarily a regular circle. The present invention uses the centroid method to obtain the center of gravity of the irregular image as the center of the impacting water column. The centroid method refers to using the horizontal and vertical coordinates of the centroid of the impacting water column to represent the position in the image. Suppose an image f(x, y), x _k is the x-coordinate of the k-th pixel, y _k is the y-coordinate of the k-th pixel, and f(x _k , y _k ) is the gray value of the k-th pixel . The total number of pixels in the center of mass coordinates (x, y) is n, then the formula is

Since the gray values of all pixels in the whole image are involved in the calculation, each pixel in the area not related to the target needs to be calculated, and for an image with irregular contours, the outermost pixel position The image centroid calculation has a greater impact. Therefore, the centroid algorithm can be improved by first finding the row and column positions where the edge pixels that satisfy the target are located, and then performing an arithmetic mean to obtain the centroid of the corresponding target image.

Edge refers to the collection of pixels with step change in surrounding pixel gray level or roof change. Edge detection is mainly to measure, detect and locate the gray level change of the image, which is to use gradient to segment the image according to the discontinuity of the image. . This derivative is usually approximated using the difference in grayscale values over smaller neighborhoods in the image. The following is a 3*3 neighborhood:

Z₁ Z₂ Z₃ Z₄ Z₅ Z₆ Z₇ Z₈ Z₉

Simple edge detection is to use first-order differential operators to perform convolution operations on the original image, such as Roberts operator, Sobel operator, and Prewitt operator.

(1) Roberts operator is one of the simplest operators, and it is an operator that uses local difference operators to find edges.

Its gradient is approximately

g _x = Z ₉ -Z ₅ (Equation 4.11)

g _y = Z ₈ -Z ₆ (Equation 4.12)

和g_y：

The convolution operator is: g _x :

and g _y :

(2) The Sobel operator approximates the two components of the gradient as

g _x = (Z ₇ +2Z ₈ +Z ₉ )-(Z ₁ +2Z ₂ +Z ₃ )) (Equation 4.13)

g _y =(Z ₃ +2Z ₆ +Z ₉ )-(Z ₁ +2Z ₄ +Z ₇ ) (Equation 4.14)

和g_y：

The convolution operator is: g _x :

and g _y :

(3) Prewitt operator

Using the Prewitt operator is computationally simpler than the Sobel operator, but the resulting noise may be slightly larger. Its gradient is approximately

g _x =(Z ₇ +Z ₈ +Z ₉ )-(Z ₁ +Z ₂ +Z ₃ ) (Equation 4.15)

g _y =(Z ₃ +Z ₆ +Z ₉ )-(Z ₁ +Z ₄ +Z ₇ ) (Equation 4.16)

和g_y：

Convolution operator: g _x :

and g _y :

Through edge detection, the edge information of the target is extracted, and the position of the centroid is obtained.

4 Off-target assessment

4.1 Calculation and error analysis of dispersion error

Missing amount, also known as impact deviation. The shooting process of the weapon system will be affected and affected by various accidental factors, which will continuously produce corresponding errors, so that the impact point deviates from the target. When evaluating misses, the dispersion error, that is, the probability deviation or mean square error of the impact point relative to the dispersion center, is used to characterize the firing density. When shooting at a target on a horizontal plane, the projectile dispersion is characterized by the distance probability deviation Ex and the direction probability deviation Ez or by the relative probability deviation Ex/x and Ez/x. The smaller the deviation of the impact point (explosion point) from the average impact point or dispersion center, the smaller the projectile dispersion of the weapon, that is, the higher the firing density.

Before calculating the dispersion error, considering that the image coordinate system may not match the projectile dispersion coordinate system, it is necessary to discuss the possible coordinate transformation.

The image information obtained by the CCD camera is converted into a digital image after analog-to-digital conversion. The coordinates of a digital image are expressed as (u, v), which represent image coordinate values in pixel units. Another coordinate system (x, y) is also required, representing the position of this pixel in the image in physical units. The origin of the coordinate system is a certain point O ₁ in the image, the x-axis is parallel to the u-axis, and the y-axis is parallel to the v-axis.

In the (x, y) coordinate system, the origin O ₁ is generally located at the center of the image, which is the intersection of the image plane and the optical axis of the camera. Assuming that O ₁ is at the (u ₀ , v ₀ ) position of the (u, v) coordinate system, the physical size of each pixel in the x-axis direction is Δ _x , and the physical size in the Y-axis direction is Δ _y , then The relationship between the two coordinate systems is represented by any pixel in the image as shown in Equation 4.17 and Equation 4.18.

In the present invention, it is necessary to establish an image coordinate system O ₁ -xy with the target as the origin, the u-axis as the x-axis, and the v-axis as the y-axis.

The formula for calculating the water column coordinates of the bullet is deduced below.

As shown in Figure 15. In the image coordinate system O ₁ -xy, the center coordinates of the impacting water column m are (x _m , y _m ), and the impacting water column world coordinate system O'-X'Y (the X and Z axes of this coordinate system are not the same as the The coordinates of the center of the impacting water column M are (X′ _M , Z′ _M ), where m is the image of M in the image. e is the proportional relationship between the actual distance and the distance on the map.

X′ _M = x _m ·e (Equation 4.19)

Z′ _M = y _m ·e (Equation 4.20)

Since the target is far away from the shooting ship, the X-axis of the impact dispersion is not easy to determine, so there may be a situation where the photographed image coordinate system does not match the impact dispersion coordinate system, that is, there is the impact water column coordinate system O'-X'Y' The process of rotating relative to the scatter coordinate system O-XZ. As shown in Figure 16. (X _M , Z _M ) are the coordinates of the impact water column in the impact dispersion coordinate system.

Rotation formula of Cartesian coordinate system

X _M =X′ _M ·cos(θ)+Z′ _M ·sin(θ) (Equation 4.21)

Z _M = -X′ _M ·sin(θ)+Z′ _M ·cos(θ) (Equation 4.22)

Θ is the rotation angle of the image coordinate system relative to the world coordinate system reflected by the image coordinate system. The rotation angle can be calculated according to the course of the drone, the course of the shooting ship, and the firing broadside, provided that the x-axis of the image coordinate system is the same as the direction of the drone. The formula for calculating Θ is shown in Equation 4.23.

Θ=TC2-TC1-Q (Equation 4.23)

In the formula, TC1 and TC2 are the headings of the shooting ship and the UAV, respectively, and Q is the shooting broadside. These parameters are available from drones and shooters.

Through the extraction and positioning of the target and the extraction and positioning of the impact water column, the position information of the target and the impact water column in the image is obtained, and a projectile dispersion coordinate system with the target as the origin is established. The dispersion error is calculated by using the coordinate data of the impact water column derived from the formula of the impact water column coordinates in the previous section. The formula for calculating the dispersion error is shown in Equations 4.24 and 4.25.

式4.25中

In Equation 4.24

In Equation 4.25

表示弹着水柱群中心坐标，它的值为所有弹着水柱实际坐标值的平均值，0.6745是或然系数，n为弹丸总数。(X _i , Z _i ) represents the actual coordinate value of each impacted water column,

Indicates the center coordinate of the impacting water column group, its value is the average value of the actual coordinate values of all the impacting water columns, 0.6745 is the probability coefficient, and n is the total number of projectiles.

According to the evaluation standard of naval gun shooting results from the shooting table of naval gun shooting, there is a standard value _K for the dispersion _error of projectiles.

The calculation of the dispersion error is completed. According to the knowledge of the error theory and the whole system design process, the main sources of the error are analyzed as follows:

(1) Preparation stage: The X-axis of the photoelectric load image is not parallel to the UAV heading direction and is not fixed; the flying height is unreasonable, which makes the shooting angle poor.

(2) Shooting stage: When the photoelectric load of the UAV is shooting, due to the long distance of the UAV from the control platform and the complex working electromagnetic environment, the photoelectric load of the UAV will have a certain deviation in its control when it is working; The photoelectric load itself will cause errors, the camera accuracy is not high and the lens distortion caused by the lens manufacturing process, or the vibration generated by the wind flow during the shooting process will be transmitted to the optical system through the fixed point of the pod's seat, all of which will affect the imaging quality. influences. Secondly, when shooting, because the image obtained by the photoelectric load cannot be proportionally scaled to the actual sea level, there is an error in calculating the dispersion distance of the impacted water column.

(3) Image transmission stage: the acquired image is affected by the performance of the image transmission system after the wireless transmission process, and the quality of the image after transmission will bring errors to the subsequent image processing.

(4) Image processing stage: The quality of the image preprocessing will affect the extraction accuracy of the subsequent target and the impact water column. The positioning accuracy of the target and the impact water column is directly related to the size of the error in the dispersion distance of the impact water column.

(5) Calculation stage: The video image and the actual sea area are not completely obtained by proportional scaling, but when the distance on the map is converted to the actual distance, the calculation is proportional to the actual distance, which will cause certain errors. In addition, the determination of the number of significant digits and the rounding of the numbers in the calculation of the dispersion distance of the impact water column can cause errors in the results.

4.2 The software design of shooting misses evaluation software based on MATLAB GUI

Want to use the rapid modeling and powerful computing power of the MATLAB tool to complete the post-event rapid analysis of the video collected by the target inspection, including the reading of the naval gun shooting video, information acquisition, image sequence acquisition, playback, pause, and interactive bullet impact distance. calculation and other functions. And use the graphical user interface (GUI) provided by MATLAB to realize a visual object-oriented operation interface, and realize the selection, processing and result display of video images and frame-by-frame pictures required for the assessment of shooting misses, as well as the calculation of dispersion errors. and other functions.

The system stores the acquired video images of naval gun shooting in the form of video frames, and selects images from the video frames to process the impact water column image when determining the impact point position. results are judged.

The specific steps are shown in Figure 3.

In the design of the system, in order to remind the user to operate according to the specification and reduce the wrong operation, a part about the system introduction has been added in the software design, and the operation process can be viewed from the software interface. At the same time, if the operation steps are different from the designed process, the system will issue a prompt message.

The GUI (Graph User Interfaces) platform provided by MATLAB is used to realize the design of the video image-based naval gun shooting missed target evaluation system of the present invention. The platform provides a variety of control tools for interface design, and the user can provide friendly In the interactive mode, realize the operation interface required for convenient and fast design. According to the implementation method of GUI, it is divided into the following three steps.

(1) Create a blank GUI.

(2) Layout and functional design of system modules.

After completing the interface design required by the system, after saving the interface design, the MATLAB GUI will generate two files related to the interface design.

.Fig file: This file includes the GUI's image window and a full description of all sub-objects and property values for all objects, including user controls and axes.

.M file: It contains all the code needed to run the GUI, control the GUI and decide how the GUI responds to the user's actions.

(3) Compile the callback function.

The user can write the callback functions required by the GUI components within the framework of the M file generated by the GUI design. The M file contains a series of sub-functions: main function, Opening function, Output function and callback function. It should be noted that the main function cannot be modified, otherwise the GUI interface initialization will fail.

Test the system according to the operation steps, run the program, enter the missed target amount evaluation interface, and input the known image coordinate system and the rotation angle of the shot dispersion coordinate system. The ideal angle is 0. You can view the operating steps of the system by clicking "System Introduction":

After completing the GUI implementation of the missed target quantity assessment system, after loading the video information and obtaining the video information and image sequence, you can view the video frame-by-frame image information in the current path.

Select and load the clear image information of the impacting water column for image processing, click the button "Extracting the impacting water column", the image display area displays the identified target, select the position of the impacting water column with the mouse cursor, and display the mark in the display area to get the impact point. Direction and distance deviation from the target are displayed. As shown in Figure 5-6, perform a water column extraction operation.

The image data used in the GUI implementation is intercepted from the video image obtained by the target ship, which reflects the distance distribution of the projectile. The result is 61.99m. Requirements for off-target measurement.

According to the same operation process, select the shots that can best reflect the water column characteristics of the shot in turn to extract the water column until all the information of the water column is obtained. Finally, click "Dispersion Error Calculation" to get the dispersion error of this group of shots. Therefore, the shooting performance is determined by using the dispersion error data.

5. Assessing the missed target of naval gun shooting plays an indispensable role in the performance appraisal of naval gun weapons and military competition training for naval forces. The present invention first introduces the composition and working principle of the UAV target inspection system and the UAV target inspection task area. The automatic target detection system obtains the video image of the relative position of the impact water column and the target fired by the naval gun from a fixed point at high altitude, transmits it to the digital processing module, and completes the extraction and positioning of the target and the impact water column through digital image processing technology. The coordinate position of the impacting water column is used to obtain the dispersion error of the shooting to evaluate the shooting results, and finally the GUI implementation of the evaluation system for the missed target amount is completed.

In the description of the present invention, unless otherwise stated, "plurality" means two or more; the terms "upper", "lower", "left", "right", "inner", "outer" The orientation or positional relationship indicated by , "front end", "rear end", "head", "tail", etc. are based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, not An indication or implication that the referred device or element must have a particular orientation, be constructed and operate in a particular orientation, is not to be construed as a limitation of the invention. Furthermore, the terms "first," "second," "third," etc. are used for descriptive purposes only and should not be construed to indicate or imply relative importance.

It should be noted that the embodiments of the present invention may be implemented by hardware, software, or a combination of software and hardware. The hardware portion may be implemented using special purpose logic; the software portion may be stored in memory and executed by a suitable instruction execution system, such as a microprocessor or specially designed hardware. Those of ordinary skill in the art will appreciate that the apparatus and methods described above may be implemented using computer-executable instructions and/or embodied in processor control code, for example on a carrier medium such as a disk, CD or DVD-ROM, such as a read-only memory Such code is provided on a programmable memory (firmware) or a data carrier such as an optical or electronic signal carrier. The device of the present invention and its modules can be implemented by hardware circuits such as very large scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, etc., or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., It can also be implemented by software executed by various types of processors, or by a combination of the above-mentioned hardware circuits and software, such as firmware.

The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited to this. Any person skilled in the art is within the technical scope disclosed by the present invention, and all within the spirit and principle of the present invention Any modifications, equivalent replacements and improvements made within the scope of the present invention should be included within the protection scope of the present invention.

Claims (10)

1. An unmanned aerial vehicle system of examining target, its characterized in that, unmanned aerial vehicle system of examining target includes:

the system comprises an unmanned aerial vehicle system, a photoelectric load system, a wireless image transmission system, a wireless remote control system and a target detection comprehensive task platform;

the unmanned aerial vehicle system comprises a four-rotor unmanned aerial vehicle body, a propulsion device, a flight control device and a power supply device; the device is used for carrying the photoelectric load system to an observation position and hovering at a fixed point;

the photoelectric load system is carried on the four-rotor unmanned aerial vehicle body; the device comprises a CCD camera, an infrared detection device and a nacelle; the system is used for executing a fixed-point observation task and carrying out fixed observation on a target range area;

the wireless image transmission system is used for transmitting the observed video images by adopting the HHLM type transmission module;

the wireless remote control system is used for controlling the unmanned aerial vehicle system and the photoelectric load system;

and the target detection comprehensive task platform is used for receiving, storing, forwarding and processing the video image.

2. The drone borescope system of claim 1, wherein the wireless remote control system comprises:

the receiving module is used for receiving the control instruction;

the decoding module is used for decoding the control instruction by using a decoder;

and the control module is used for controlling the unmanned aerial vehicle system and the photoelectric load system based on the control instruction obtained by decoding.

3. An unmanned aerial vehicle target detection method applied to the unmanned aerial vehicle target detection system according to any one of claims 1-2, wherein the unmanned aerial vehicle target detection method comprises the following steps:

the method comprises the steps of obtaining a video image of a water column shot by a ship gun by utilizing a photoelectric load, analyzing and processing the video image of the water column shot by the ship gun by obtaining video information, obtaining an image sequence, playing the video information, selecting images frame by frame, inputting a coordinate system rotation angle, extracting the water column shot by the ship gun and other means to obtain a hit rate and a miss rate of shooting of the ship gun, and obtaining a shooting evaluation result.

4. The drone borescope method of claim 3, wherein the drone borescope method comprises the steps of:

generating a control instruction, and controlling the quad-rotor unmanned aerial vehicle carrying the photoelectric load to fly to a safe height above a target and then hover based on the generated control instruction; meanwhile, controlling the unmanned aerial vehicle to adjust the shooting attitude of the photoelectric load based on the generated control instruction;

controlling the photoelectric load to record the shooting condition of the naval gun in the offshore target area based on the control instruction, continuously acquiring a water column video image stimulated by the shot entering water, and acquiring the relative relation between the shot water column and the target in the target area, namely the scattering condition of naval gun shooting;

converting the obtained video image into an electric signal, and converting the electric signal into a digital signal after sampling, quantizing and encoding by using an analog/digital converter;

step four, preprocessing the converted image to obtain clear and useful image information; and calculating the relative position of the shot water column and the floating target based on the obtained clear and useful image, calculating to obtain a shot scattering error, and evaluating the shot miss amount of the naval gun.

5. The drone target detection method of claim 4, wherein the safe height calculation method comprises:

firstly, analyzing the hovering height of the unmanned aerial vehicle based on the shot-firing height of the ship cannon:

H≥h；

wherein H represents the hovering height of the drone; h represents the height of the projectile when it reaches the highest point; the height h of the projectile when reaching the highest point is determined by the relationship among the firing angle, the firing distance and the firing height of the naval gun;

secondly, analyzing the hovering height of the unmanned aerial vehicle based on the photoelectric load observation range: determining the range of the obtained judgment interval according to the type, the shooting method and the average shooting distance of the artillery and the corresponding table of the evaluation parameters of the floating targets and the simulated target shooting scores; determining the hovering height of the unmanned aerial vehicle according to the judgment interval rectangular boundary; the rectangular boundary of the judgment interval is the rectangular boundary of the cannonball falling into the effective hit area according to the corresponding shooting score evaluation parameter table when the cannonball shoots the floating target and the simulated target;

and finally, determining the safety height of the unmanned aerial vehicle based on the hovering height analysis result of the unmanned aerial vehicle launched by the ship cannon, the judgment interval rectangular boundary limit, the field angle range of the camera and the flight height limit of the unmanned aerial vehicle.

6. The unmanned aerial vehicle target detection method of claim 4, wherein the preprocessing of the converted image to obtain clear and useful image information comprises:

(1) segmenting the converted image by adopting a segmentation method based on color characteristics, and identifying and extracting a target region;

(2) performing curve fitting on the basis of the target information obtained after segmentation by adopting a least square method to determine the circle center and the radius of the target for positioning the target;

(3) and carrying out graying, median filtering and contrast enhancement on the extracted and positioned target image.

7. The drone target detection method of claim 4, wherein calculating the relative position of the bouncing water column and the floating target based on the obtained clear and useful image comprises:

calculating a threshold value through a self-adaption and iteration method to segment the image of the impinging water column to obtain the image of the impinging water column with the background removed; carrying out binarization processing on the obtained image of the bouncing water column after the background is removed; meanwhile, the positioning of the bouncing water column is carried out by utilizing an improved centroid method.

8. The drone boresight method of claim 7, wherein the segmenting the ballistic water column image by calculating the threshold values by an adaptive and iterative method comprises:

1.1) counting the minimum gray value T of the shot water column image_minMaximum gray value T_maxCalculating a binary average value as an initial threshold value T:

1.2) segmenting the image according to the threshold value T to obtain two pixel sets which are respectively:

G₁＝{f(x，y)≥T}，G₂＝{f(x，y)≤T}；

1.3) computing the set of pixels G₁And G₂Gray scale average value mu of₁And mu₂：

1.4) according to μ₁And mu₂Calculating a new threshold value

Step 1.2), step 1.3), step 1.4) are repeated until the threshold T converges to a certain range.

9. The drone borescope method of claim 4, wherein the positioning of the bouncing water column using the modified centroid method comprises:

performing edge detection on the bouncing water column, extracting edge information of the bouncing water column, and performing arithmetic mean calculation on the basis of the edge information of the bouncing water column to calculate the mass center of the bouncing water column;

in the fourth step, the step of calculating to obtain the shot scattering error and the step of evaluating the shot miss amount of the gun comprises the following steps:

(1) the calculation of the dispersion error is performed:

based on the obtained position information of the target and the shot water column in the image, a shot scattering coordinate system with the target as an origin is established, and scattering errors are calculated by the following formula:

wherein,

in formula 4.25

(X_i，Z_i) The actual coordinate values of each of the impinging water columns are indicated,

representing the coordinates of the center of the group of bouncing water columns,

the value of (A) is the average value of the actual coordinate values of all the bouncing water columns; 0.6745 represents the likelihood factor, n represents the total number of projectiles;

(2) comparing calculated E_X，E_ZDetermining the shooting result of the gun and the gun shooting miss amount evaluation result according to the value of the difference and the impact point scattering error standard value K.

10. A computer-readable storage medium storing a computer program which, when executed by a processor, causes the processor to perform the functions of the drone targeting system of any one of claims 1-2.

Images