阅读说明：本技术 基于图像处理技术的舰炮系统精度检测装置及检测方法 (Image processing technology-based ship cannon system precision detection device and detection method ) 是由 李宙恒 黄金锋 王兴众 龙加军 赵复政 于 2018-09-13 设计创作，主要内容包括：本发明设计了一种基于图像处理技术的舰炮系统精度检测装置及检测方法,图像传感器设置在零飞仪上,图像传感器用于采集炮火轴线前方图像,在统一的时间脉冲信号下,炮火轴线前方图像及其数据信息和跟踪器的目标跟踪图像及其数据信息均传送到计算机,计算机从跟踪器的目标跟踪图像中,根据跟踪器波门选择的范围,提取出跟踪目标,将跟踪目标和炮火轴线前方图像进行模板匹配,根据匹配算法计算相似度并输出匹配结果,在炮火轴线前方图像中定位跟踪器所跟踪目标的位置,计算平均零飞误差。具有目标自动检测与识别,图像处理分析、零飞误差自校正等集成优化特点,能够利用近距离角点目标检测舰炮系统综合误差。(The invention designs a ship gun system precision detection device and a detection method based on an image processing technology, wherein an image sensor is arranged on a zero-flight instrument and used for acquiring an image in front of a gun axis, the image in front of the gun axis and data information thereof as well as a target tracking image and data information thereof of a tracker are transmitted to a computer under a uniform time pulse signal, the computer extracts a tracking target from the target tracking image of the tracker according to the range selected by a wave gate of the tracker, performs template matching on the tracking target and the image in front of the gun axis, calculates similarity according to a matching algorithm and outputs a matching result, positions the target tracked by the tracker in the image in front of the gun axis, and calculates an average zero-flight error. The method has the integrated optimization characteristics of automatic target detection and identification, image processing analysis, zero-flight error self-correction and the like, and can detect the comprehensive error of the gun system by using the near-distance angular point target.)

1. The device is characterized in that an image sensor is arranged on a zero-flight instrument, the image sensor is used for acquiring images in front of the axis of a gun fire, a tracker is fixedly arranged on a ship, under the condition of a uniform time pulse signal, the images in front of the axis of the gun fire and data information of the images in front of the axis of the gun fire as well as target tracking images and data information of the tracker are transmitted to a computer, the computer extracts a tracking target from the target tracking images of the tracker according to the range selected by a gate of the tracker, performs template matching on the tracking target and the images in front of the axis of the gun fire, calculates the similarity according to a matching algorithm and outputs a matching result, positions the position of the tracking target tracked by the tracker in the images in front of the axis of the gun fire, and calculates the azimuth deviation angle e of the tracking target relative to the center of an aiming line of a gun_xAnd a high and low deviation angle e_yAnd eliminating singular points through threshold filtering, and calculating the average zero-flight error.

2. The image processing technology-based gun system precision detection device of claim 1, wherein the computer preprocesses each image, converts the image into RGB space, decomposes an RGB model color image into 3 color components, performs image enhancement on the decomposed 3 color components based on a pixel gray value point operation, and synthesizes a new color image.

3. The image processing technology-based gun system precision detection device according to claim 1 or 2, wherein a power supply and a communication transfer box provide power for the zero flight instrument, and the communication transfer box receives video information sent by an image sensor of the zero flight instrument; providing a time unified signal for a computer through an analog interface, transmitting video and data information to the computer through a network interface, and receiving data information externally sent by the computer through a communication transfer box; the communication adapter box is provided with a communication interface with an external gun system.

4. The method for detecting the accuracy of the gun system based on any one of claims 1 to 3 is characterized by comprising the following steps of firstly, enabling the gun system to enter a zero-flight working mode, and simultaneously acquiring an image in front of a gun axis and data information thereof and a target tracking image of a tracker and data information thereof under a unified time pulse signal;

extracting a tracking target from a target tracking image of the tracker according to the range selected by the wave gate of the tracker, performing template matching on the tracking target and an image in front of the axis of the gun fire, calculating the similarity according to a matching algorithm and outputting a matching result; positioning the position of a target tracked by a tracker in an image in front of the axis of the gunfire; thirdly, processing the image in front of the gun firing axis according to the image recognition based on template matching, positioning the position of the tracking target, and calculating the azimuth deviation angle e of the tracking target relative to the center of the gun muzzle sight line according to the relative position of the tracking target in the image_xAnd a high and low deviation angle e_yAnd eliminating singular points through threshold filtering, and calculating the average zero-flight error.

5. The gun system accuracy detection method according to claim 4, characterized in that in the third step, the template matching process specifically includes that the tracked target T is a P × Q template, the image S in front of the gun axis is a search graph which needs to be matched according to the template, the size is M × N, Sij is a subgraph with the size of P × Q covered by the template, i and j are coordinates of a vertex at the upper left corner of the subgraph in the search graph respectively, the search graph is divided into (M-P +1) (N-Q +1) subgraphs with the size of MN, one subgraph with the size same as that of the template graph is obtained from the origin of the search graph, the similarity between the template and the subgraph is calculated according to the matching algorithm, the template is sequentially slid in the search graph in a sliding manner from left to right and from top to bottom, the similarity after each movement is calculated, and finally (M-P +1) (N-Q +1) numerical values are obtained, and comparing and selecting the best matching point.

6. The gun system accuracy detection method of claim 5, characterized in that, in the matching algorithm, an improved MAD matching algorithm is adopted, as follows:

in the formula, the size of the search image S is mxn, and the size of the template image T is pxq; the search ranges of x and y are: x is 0,1 … … M-P + 1; y is 0,1 … … N-Q + 1;

7. The naval gun system precision detection method according to claim 4, characterized in that, in the second step, each image is preprocessed by adopting an improved gray scale stretching transformation algorithm, the image is firstly converted into an RGB space, an RGB model color image is decomposed into 3 color components, the decomposed 3 color components are respectively subjected to image enhancement based on a pixel gray value point operation, and then the image is synthesized into a new color image.

The image enhancement formula for each color component is as follows.

Image enhancement formula:

in the formula, H' and H represent the gray scale values of the enhanced output image and the video input image, respectively. The method comprises the steps of selecting the minimum value of each line on each color component in one frame of original image, accumulating and summing the minimum values of all the lines and averaging the minimum values to obtain the minimum value Min of a target interval of the selected frame of image, selecting the maximum value of each line on each color component in the frame of image in the same way, accumulating and summing the maximum values of all the lines and averaging the maximum value Max of the target interval of the selected frame of image, wherein the maximum value Max of the target interval of the selected frame of image is obtained, so that the situation that the original data value is larger than Max possibly occurs in the target conversion interval in the enhancement formula, if the amplification factor is set to be 255, a lot of data exceeding the gray level range can occur, the gain is serious, too many too bright pixel points of the image can occur, and the amplification factor is set to be 224.

Technical Field

The invention belongs to the technical field of ship precision detection, and particularly relates to a ship and cannon system precision detection device and method based on an image processing technology.

Background

In order to complete a special mission task of a certain official law enforcement vessel, a naval gun system is required to be equipped, and the naval gun system comprises naval guns, tracking equipment, fire control equipment, auxiliary equipment and the like. An important condition for the gun to strike the target accurately in real time is that the zero flight precision (also called tracking and aiming precision) of the gun system meets the requirement of technical indexes, so how to detect the zero flight precision of the gun system quickly and accurately is particularly important. The zero flight precision is an important index for evaluating the battle technical performance of the naval gun system. The zero-flight precision refers to the precision degree of the axis of the gun barrel pointing to the target when the gun system performs static or dynamic tracking aiming in a zero-flight working mode (namely the time of the shot flying from the muzzle to the target is zero, and if the system error is zero, the axis of the gun barrel should directly point to the target). The traditional zero-flight precision detection depends on a zero-flight instrument for detection, and the zero-flight instrument generally comprises a television measuring device, a zero-flight instrument display and control console (or a computer), a communication or power supply transfer box and the like. The method for measuring the zero-flight precision of the gun by using the zero-flight instrument comprises the following steps: and (6) installing and calibrating. Fixing the television measuring device on a naval gun barrel through a special clamp, connecting a cable between the television measuring device and a display control console, and adjusting the optical axis direction of the television measuring device through a calibration method to keep the same as the sighting line of the naval gun; test admission. When a zero-flight test is carried out, according to the imaging characteristics of a target and a background, the brightness and the contrast are adjusted on a display control platform of a zero-flight instrument to enable the target image to be clear, and a zero-flight video image is recorded in real time by the zero-flight instrument; and comparing and processing the recorded zero-flight video image with an image of a tracking target of a tracker (such as a tracking radar or a photoelectric tracker) frame by frame to obtain error data of a stable altitude angle and an unstable azimuth angle of the system part.

Although the conventional detection method can relatively quickly measure the zero flight precision of the system through electronic equipment, the conventional detection method still has certain defects: (1) the traditional zero flight instrument records video images through a television measuring device for analysis processing, but the selection and the identification of a tracking target still need to be carried out in a manual mode, so that the selected point and an actual tracking point can cause deviation in direction and height;

(2) the recorded video image cannot be automatically compared with a target tracking image of the tracker, and the data is manually read to influence the detection precision;

(3) the detection result only exists on the single zero-flight instrument device, cannot be recorded by the ship and cannon system, and is not beneficial to maintenance and development of the ship and cannon system.

Disclosure of Invention

The invention aims to solve the technical problem of providing a device and a method for detecting the accuracy of a gun system based on an image processing technology, which have the integrated optimization characteristics of automatic target detection and identification, image processing analysis, zero-flight error self-correction and the like, and can detect the comprehensive error of the gun system by utilizing a near-distance angular point target.

The technical scheme adopted by the invention for solving the technical problems is as follows: firstly, providing a ship-gun system precision detection device based on an image processing technology, wherein an image sensor is arranged on a zero-flight instrument and used for acquiring a gun axis front image, a tracker is fixedly arranged on a ship, the gun axis front image and data information thereof, a target tracking image of the tracker and data information thereof are transmitted to a computer under a uniform time pulse signal, the computer extracts a tracking target from the target tracking image of the tracker according to the range selected by a gate of the tracker, performs template matching on the tracking target and the gun axis front image, calculates similarity according to a matching algorithm and outputs a matching result, positions the position of the target tracked by the tracker in the gun axis front image, and calculates an azimuth deviation angle e of the tracking target relative to the center of a gun muzzle sight line_xAnd a high and low deviation angle e_yAnd eliminating singular points through threshold filtering, and calculating the average zero-flight error.

According to the technical scheme, the computer preprocesses each image, converts the image into an RGB space, decomposes an RGB model color image into 3 color components, performs image enhancement on the decomposed 3 color components respectively based on the point operation of pixel gray values, and synthesizes the color components into a new color image.

According to the technical scheme, the power supply and the communication transfer box provide power for the zero flight instrument, and the communication transfer box receives video information sent by the image sensor of the zero flight instrument; providing a time unified signal for a computer through an analog interface, transmitting video and data information to the computer through a network interface, and receiving data information externally sent by the computer through a communication transfer box; the communication adapter box is provided with a communication interface with an external gun system.

The zero-flight instrument is attached with an adjusting ring, and the adjusting ring matched with the zero-flight instrument is selected according to naval guns with different calibers.

The invention also discloses a method for detecting the accuracy of the gun-based system, which comprises the following steps that firstly, the gun-based system enters a zero-flight working mode, and simultaneously acquires the front image and the data information of the axis of the gun and the target tracking image and the data information of the tracker under the condition of a uniform time pulse signal; the adoption frequency can be set according to the needs of users, and image recording based on uniform time information is realized.

Extracting a tracking target from a target tracking image of the tracker according to the range selected by the wave gate of the tracker, performing template matching on the tracking target and an image in front of the axis of the gun fire, calculating the similarity according to a matching algorithm and outputting a matching result; positioning the position of a target tracked by a tracker in an image in front of the axis of the gunfire; in some environments, the reliability of the method for automatically identifying the target by the computer is poor, for example, a complex background is arranged on an image, other elements similar to the tracked target are arranged on the image, and the like. At this time, a method of manually selecting the target is required. In the test process, a tracking target is manually selected on a computer, and a box is arranged on the target, so that the detection system can identify the tracking target on the recorded image according to the box selected area. Thirdly, processing the image in front of the axis of the gun fire according to the image recognition based on template matching, positioning the position of the tracking target, and calculating the relative position of the tracking target in the image by combining the parameters of a zero-flight instrument cameraAzimuth deviation angle e of center of line of sight_xAnd a high and low deviation angle e_yAnd eliminating singular points through threshold filtering, and calculating the average zero-flight error.

A set of zero-fly-error values e can be calculated for each set of images acquired_xAnd e_yDepending on the frequency setting used, hundreds of images may be captured. After calculation, there are hundreds of sets of zero-flight error data. These raw zero-flight error data are further processed as needed.

According to the technical scheme, in the third step, the template matching process specifically comprises the steps of tracking a target T as a P multiplied by Q template, taking an image S in front of a gunfire axis as a search map needing to be matched according to the template, taking M multiplied by N, Sij as subgraphs covered by the template and having the size of P multiplied by Q, i and j as coordinates of a vertex at the upper left corner of the subgraph in the search map, dividing the search map into (M-P +1) (N-Q +1) subgraphs with the size of MN, obtaining one subgraph with the same size as the template map from the origin of the search map, calculating the similarity between the template and the subgraph according to a matching algorithm, sliding the template in the search map in sequence in a sliding mode from left to right and from top to bottom, calculating the similarity after each movement, and finally obtaining (M-P +1) (N-Q +1) numerical values, and comparing and selecting the best matching point.

According to the technical scheme, an improved MAD matching algorithm is adopted in the matching algorithm, the algorithm complexity is low, and the operation amount can be effectively reduced. As follows:

in the formula, the size of the search image S is mxn, and the size of the template image T is pxq; the search ranges of x and y are: x is 0,1 … … M-P + 1; y is 0,1 … … N-Q + 1;

for traversal matches, the offset is the average of the search sub-images at (i, j),is the average of the template images.

Images acquired by the camera cannot be directly used for template matching due to interference of various conditions (environmental influences such as weather, illumination and the like) and random noise. Therefore, the image needs to be preprocessed before template matching, noise and irrelevant information interference are reduced, and target features are enhanced and completely stored as much as possible.

According to the technical scheme, the second step comprises the step of preprocessing each image by adopting an improved gray scale stretching transformation algorithm, so that the purposes of reducing the calculated amount and realizing rapid processing are achieved. Firstly, converting an image into an RGB space, decomposing an RGB model color image into 3 color components, respectively carrying out image enhancement on the 3 decomposed color components based on the point operation of pixel gray values, and then synthesizing into a new color image.

The image enhancement formula for each color component is as follows.

Image enhancement formula:

in the formula, H' and H represent the gray scale values of the enhanced output image and the video input image, respectively. The method comprises the steps of selecting the minimum value of each line on each color component in one frame of original image, accumulating and summing the minimum values of all the lines and averaging the minimum values to obtain the minimum value Min of a target interval of the selected frame of image, selecting the maximum value of each line on each color component in the frame of image in the same way, accumulating and summing the maximum values of all the lines and averaging the maximum value Max of the target interval of the selected frame of image, wherein the maximum value Max of the target interval of the selected frame of image is obtained, so that the situation that the original data value is larger than Max possibly occurs in the target conversion interval in the enhancement formula, if the amplification factor is set to be 255, a lot of data exceeding the gray level range can occur, the gain is serious, too many too bright pixel points of the image can occur, and the amplification factor is set to be 224.

The invention has the following beneficial effects: the method has the integrated optimization characteristics of automatic target detection and identification, image processing analysis, zero-flight error self-correction and the like, and can detect the comprehensive error of the gun system by utilizing the near-distance angular point target. The device has the characteristics of simple use, convenience in carrying, high reliability and the like, is suitable for naval cannons with different calibers, and can be expanded to be applied to other devices needing similar tests, such as water cannons, multifunctional launching devices and the like.

Drawings

The invention will be further described with reference to the accompanying drawings and examples, in which:

FIG. 1 is a schematic structural diagram of a naval gun precision detection device in the embodiment of the invention;

FIG. 2 is a diagram of interface relationships in an embodiment of the present invention;

FIG. 3 is a flowchart of a method for precision detection according to an embodiment of the present invention;

FIG. 4 is a flow chart of template matching in an embodiment of the present invention;

FIG. 5 is a schematic diagram of template matching according to an embodiment of the present invention;

FIG. 6 is a schematic diagram of a zero-fly-error embodiment of the present invention;

FIG. 7 is a schematic diagram of an interface of an accuracy testing apparatus according to an embodiment of the present invention;

FIG. 8 is a schematic diagram of an error analysis interface in an embodiment of the invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

In a ship cannon precision detection system, how to realize the integration optimization of functions such as interface adaptation, target automatic identification, image processing analysis, zero-flight error self-correction and the like meets the requirements of less personnel configuration, simple operation and use and equipment integration sharing, and is a key technical problem solved by the invention.

In the embodiment of the invention, as shown in fig. 1, an image sensor is installed on the zero flight instrument and is responsible for acquiring a video image in front of the axis of the gun fire; the adjusting ring is attached, and the adjusting ring which is suitable for the cannon with different calibers can be selected according to the cannon. The power supply and communication transfer box can provide power for the zero-flight instrument and receive video information sent by the image sensor of the zero-flight instrument; providing a time unified signal for a computer through an analog interface, providing video and data information through a network interface, and receiving data information externally sent by the computer; and the communication interface with an external gun system is provided. The computer has the control function of the detection system, is responsible for initially calibrating the installation position of the zero-flight instrument, controls various functions in the implementation of the zero-flight test, processes, analyzes and obtains the zero-flight error of the image, and can be in data communication with the gun system.

Interface relationship: the digital interface design technology is used for digitally designing various interfaces of the system, defining and requiring functions, names, input and output parameters and the like of the interfaces, achieving the standardization and normalization of the interfaces and meeting the current integration requirement and the requirement of later upgrading and expanding.

The precision detection device needs to have two interfaces, namely an analog interface and a network interface, with the gun system, as shown in fig. 2. The simulation interface transmits a uniform frequency signal, and auxiliary equipment in the naval gun system provides a high-precision frequency signal; the network interface transmits video information and data information, receives a target tracking video transmitted by the tracker, receives a control command message of the gun system, such as 'start/stop zero flight working mode', 'system tracking good' and the like, receives a uniform time message sent by the gun system, and sends an equipment state message and a zero flight error data message to the gun system.

Applicability: the precision detection device can be suitable for naval guns with different calibers, and can be expanded to be applied to other equipment which needs similar tests, such as water cannons, multifunctional launching devices and the like. The zero-flight instrument is attached with an adjusting ring, and the adjusting ring which is suitable for the zero-flight instrument can be selected according to naval guns with different calibers. By adopting precision machining and high-precision axial positioning design, the height of a mechanical shaft of the zero-flight instrument is ensured to be consistent with that of an optical axis of the camera, the zero-flight instrument is ensured to be inserted into the gun barrel in place, and the height of the optical axis of the camera is consistent with that of the axis of the gun barrel.

When the detection test starts, the computer in the precision detection device can simultaneously acquire two paths of video signals. When the ship and cannon system enters a zero-flight working mode, a control signal is sent to the detection device, related data are sampled, and the adopted frequency can be set according to the needs of users.

As shown in fig. 3, a computer in the precision detection apparatus acquires two paths of video signals, one path is video information that a muzzle points to the front and is acquired by an image sensor on a zero flight instrument, and the other path is video information that a target of a tracker tracks.

And then, image preprocessing is carried out, filtering and image enhancement are adopted to reduce image noise and irrelevant information interference, and a template matching method is adopted to automatically identify the image template of the null flight instrument, wherein the image has a complex background, and other elements similar to the tracked target are arranged on the image. At this time, a method of manually selecting the target is required. In the test process, a tracking target is manually selected on a computer, and a box is arranged on the target, so that the detection system can identify the tracking target on the recorded image according to the box selected area.

Then according to the image processing result, calculating the azimuth deviation angle e of the tracking target relative to the center of the gun muzzle aiming line_xAnd a high and low deviation angle e_yI.e., zero flight error, as shown in fig. 6, and send the data pair out to the gun system.

As shown in fig. 4, in the template matching flowchart, firstly, a tracking target is extracted from a target tracking image of a tracker through an original wave gate on the image, the tracking target is used as a template, the template and a zero-flight instrument image (an image in front of a gun axis) are preprocessed by using an improved gray scale stretching transformation algorithm to form a gray scale information set, then, the similarity is calculated according to the template matching algorithm, and the tracking target is positioned on the zero-flight instrument image according to a matching result. The template matching process is shown in fig. 5.

As shown in fig. 7, the interface of the precision detection device is schematically divided into seven regions on the display interface. The equipment connection display area is used for displaying the connection state of equipment such as a zero flight instrument and a ship and cannon system, has a message debugging function, and can display and debug network messages. The image is provided with a display area, and the time for starting and stopping receiving the video is selected, so that the focal length, the view field and the like of the null plane picture can be adjusted. And (3) setting a selection mode of the template in a zero-flight detection display area: automatic comparison or manual selection; setting the sampling frequency of pictures and data, setting 5 common frequencies by the system, and also being set by a user, and intercepting a single-frame video image; the output data and mode can be set. The data display area can display the azimuth deviation, the elevation deviation and the recorded sample number, analyze the data, make an error analysis curve, calculate the correction amount and the deviation distribution and the like. And the zero-flight instrument video display area is used for displaying the television video on the zero-flight instrument. And the tracker video display area is used for displaying the television video on the tracker, such as the television video of a photoelectric tracker, the television video of a tracking radar and the like. And an operation prompt display area for displaying the operation state of each step, feeding back error information and the like.

As shown in fig. 8, the error analysis interface diagram can perform curve analysis on error data, adjust the coordinate axis display mode, and display the average deviation dispersion value and the variance deviation dispersion, the mouse position, and the like.

After the traditional zero-flight instrument measures the zero-flight error, the result cannot be accessed to the gun system, and the inspection and the maintenance of the gun system in the whole life period are not facilitated. According to the invention, through methods of designing an interface, formulating a communication protocol, editing a software module and the like, a zero-flight detection result can be sent to a ship gun system, the functions of storing, checking, carrying out statistical analysis and self-correcting the detection result are realized, and a closed loop is formed.

After the zero-flight detection is finished, the detection result is triggered and sent on the computer, the detection result can be sent to the fire control equipment of the ship gun system in a data form through the network, the fire control equipment receives the data and then stores the data into the local server, the result data can be called through the zero-flight detection interface, the data is checked, counted and analyzed in a curve mode, and the data can be automatically corrected to shooting data resolution of the fire control equipment according to requirements. The verification is carried out during the subsequent firing practice, and through the detection-verification closed-loop mode, the personnel maintenance burden is reduced, and the firing precision of the naval gun system is continuously optimized.

The invention realizes the interface adaptation of the zero-flight instrument and the gun system through a digital interface design technology, so that the zero-flight instrument can carry out information interaction on the gun system, realizes the automatic identification and the zero-flight error detection on a tracking target through an image processing technology, cuts, optimizes and integrates various flows and human-computer interfaces with independent functions such as zero-flight error detection, fire control software correction and the like through a software design method, and achieves a self-correcting zero-flight error closed-loop processing process.

It will be understood that modifications and variations can be made by persons skilled in the art in light of the above teachings and all such modifications and variations are intended to be included within the scope of the invention as defined in the appended claims.