阅读说明：本技术 一种低空快速飞行条件下的光电吊舱扫描方法 (Photoelectric pod scanning method under low-altitude rapid flight condition ) 是由 赵君臣 司帆 申亮亮 张如飞 于 2021-07-30 设计创作，主要内容包括：本发明目的在于提供一种低空快速飞行条件下的光电吊舱扫描方法,实时计算图像所代表的空间地理位置,并根据相邻图像的重叠系数来调整伺服框架的摆动角速度,以实现对目标区域快速、大范围和无遗漏地扫描,解决了光电吊舱传统扫描方法搜索效率较低的问题。光电吊舱扫描系统,包括：光学相机、激光测距仪、伺服框架、图像处理模块、基座和POS模块。光学相机还包括快速反射镜。其中,伺服框架按一定的摆动频率驱动光学相机对大范围区域扫描成像,快速反射镜维持光学相机视轴在曝光期间的惯性空间稳定,以清晰成像。(The invention aims to provide a photoelectric pod scanning method under a low-altitude rapid flight condition, which is used for calculating the space geographic position represented by an image in real time and adjusting the swinging angular speed of a servo frame according to the overlapping coefficient of adjacent images so as to realize rapid, large-range and non-leakage scanning of a target area and solve the problem of low searching efficiency of the traditional photoelectric pod scanning method. An electro-optical pod scanning system comprising: the system comprises an optical camera, a laser range finder, a servo frame, an image processing module, a base and a POS module. The optical camera also includes a fast mirror. The servo frame drives the optical camera to scan and image a large area according to a certain swing frequency, and the fast reflector maintains the inertial space of the optical camera visual axis during exposure to be stable so as to clearly image.)

1. A photoelectric pod scanning method under a low-altitude rapid flight condition is characterized by comprising the following specific steps:

first step building photoelectric pod scanning system

An electro-optical pod scanning system comprising: the system comprises an optical camera, a laser range finder, a servo frame, an image processing module, a base and a POS module; the optical camera includes: the device comprises a body, a lens and a quick reflector;

the optical camera is in a cuboid shape and is arranged in the inner platform body of the servo frame to realize the image acquisition function; the camera body is fixedly connected with the camera lens; the quick reflector is a circular lens with a motor drive, and is arranged in the lens to realize an image stabilizing function under the swinging condition;

an optical lens is embedded in the laser range finder, is in a cuboid shape, is arranged in the inner platform body of the servo frame, and is used for measuring the distance of the ground by emitting laser; the servo frame is spherical, the base is cylindrical, and the servo frame is connected with the base through a flange;

second step image processing module determines scan strip width and swing angular velocity

Before the photoelectric pod scanning system enters a scanning area, the image processing module determines the width and the swinging angular speed of a scanning strip;

the swath width L in the parallel flight direction is determined using equation (1):

L＝D×ε/sinθ (1)

determining the servo azimuth frame yaw angular velocity ω using equation (2):

the swath width W in the vertical flight direction is determined using equation (3):

third step fast reflector compensation image motion

After the servo frame continuously swings, the quick reflector swings in the opposite direction when the optical camera is exposed, image motion is compensated, and a scene is imaged clearly;

fourth, the POS module determines the position information

The POS module collects gyro and acceleration information and carries out combined navigation; acquiring satellite signals, determining inertia/satellite coupling, and obtaining longitude, latitude and elevation information;

the fifth step is that the image processing module positions the ground area corresponding to the single frame image

According to the longitude, latitude and elevation information, the azimuth angle and the pitch angle of the servo frame, the image processing module determines the azimuth angle and the pitch angle of the visual axis in the inertial space; setting the zero position coordinate of the internal visual axis of the image as O (x0, y0), the coordinate of the upper left corner point as A (0, 0), the coordinate of the lower left corner point as B (0, H _ pix), the coordinate of the lower right corner point as C (W _ pix, H _ pix), and the coordinate of the upper right corner point as D (W _ pix, 0); w _ pix, H _ pix are effective pixels in the image width and height directions, respectively;

azimuthal deviation ψ of point a with respect to point O_AExpressed by equation (4):

pitch offset of point a relative to point OExpressed by equation (5):

attitude matrix of inertial system rotating to ground scenery direction pointed by point AExpressed by equation (6):

thereby obtaining the azimuth and the pitch angle of the ground scene direction pointed by the point A in the inertial space from the attitude matrix; the azimuth and the pitch angle of the ground scenery direction pointed by B, C and D point in the inertial space are deduced in the same way, and then the zero position of the visual axis and the longitude, the latitude and the elevation of the ground scenery pointed by 4 angle points are determined according to the distance information of the laser range finder, so that the positioning of the ground area corresponding to the single-frame image is realized;

sixthly, the image processing module completes image splicing of the scanning area

The image processing module is used for recording the position information of the ground area corresponding to the multi-frame images in the scanning process, determining the overlapping coefficient of adjacent areas according to the side length overlapping condition of the adjacent images and judging whether the adjacent areas are in seamless connection or not; when the adjacent areas are overlapped less, the swing angular speed is reduced; when the adjacent areas are overlapped more, the swinging angular speed is increased; gradually enlarging the boundary of the scanned area through area accumulation; when the geographic coordinates of the scanned area relative to the target area on four endpoints of northeast, southeast, southwest and northwest are more northeast, southeast, southwest and northwest, the target area is completely covered, and the photoelectric pod scanning is completed;

and completing the photoelectric pod scanning under the low-altitude fast flight condition.

2. The electro-optical pod scanning method under low-altitude fast flight conditions as claimed in claim 1, wherein the POS module functions as: and outputting the attitude angle and the geographical position of the pod through the measurement of the gyroscope, the adding table and the satellite receiver.

3. The method for scanning optoelectronic pod in low-altitude fast flight condition as claimed in claim 1, wherein the image processing module functions as: and receiving the image from the optical camera, and finishing image processing and data calculation.

4. The method for scanning the optoelectronic pod under the low-altitude fast flying condition as claimed in claim 1, wherein in the formula (1), D is the flying height from the ground, e is the field of view of the optical camera along the longitudinal axis, and θ is the angle between the visual axis and the horizontal ground.

5. The method for scanning the photoelectric pod under the low-altitude fast flight condition as claimed in claim 1, wherein in the formula (2), m is the imaging times in the process of swinging from one side to the other side, γ is the field of view of the horizontal axis of the optical camera, k is the overlapping coefficient of adjacent images, v is the flight speed of the photoelectric pod scanning system, and L is the flight distance of the azimuth frame in one time of swinging left and right.

6. The electro-optical pod scanning method under low-altitude fast flight condition as claimed in claim 1, characterized in that in the formula (5), f represents the optical camera focal length, d represents the pixel size; the zero position direction of the visual axis is taken as the X axis, the direction represented by the image ordinate is the Y axis, and the attitude matrix of the optical camera is

Technical Field

The invention relates to a photoelectric pod scanning method, in particular to a photoelectric pod scanning method under a low-altitude rapid flight condition.

Background

The photoelectric pod is a photoelectric device with a rotating frame and is installed on carriers such as airplanes, ships, vehicles and the like, and is generally used for detecting and identifying objects such as automobiles, human bodies and the like. The photovoltaic pod can integrate a variety of loads, the primary loads including: optical camera, laser range finder and POS module. The optical camera is used for acquiring a high-definition image of a target in the daytime/at night; the laser range finder emits laser, detects an optical signal reflected by a target, and measures the target distance; the POS module realizes satellite and inertia combined positioning and attitude measurement, and obtains the position and attitude of the photoelectric pod.

Under the condition of low-altitude fast flight, the speed height of the airplane is high, and the photoelectric pod is difficult to observe the ground. Firstly, too large a speed-to-height ratio results in insufficient time to conduct a large-scale scan search; secondly, the moving speed of the ground scenery in the visual field of the photoelectric pod is high, so that the tail of an image is removed during exposure, and clear imaging is not easy to realize; thirdly, within a limited time, seamless connection of the scanning area cannot be achieved, and valuable ground targets are easily missed.

Under low-altitude fast flight conditions, the traditional photoelectric pod scanning method is as follows: the search efficiency is reduced, the ground area which can be scanned by one flight is reduced, and clear imaging of small-range ground scenery is realized by using a frame servo or a quick reflector; the specific area is repeatedly scanned so as to avoid missing the ground target. After the method is adopted, the scanning of the specific area can be finished by multiple times of back-and-forth flight, the searching efficiency is reduced, and the task time is increased.

In order to realize the large-scale searching capability under the low-altitude rapid flight condition, a new scanning method needs to be adopted, and the problem of low searching efficiency of the traditional scanning method of the photoelectric pod is solved.

Disclosure of Invention

The invention aims to provide a photoelectric pod scanning method under a low-altitude rapid flight condition, and solves the problem that the traditional photoelectric pod scanning method is low in searching efficiency.

The photoelectric pod scanning method under the low-altitude rapid flight condition comprises the following specific steps:

first step building photoelectric pod scanning system

An electro-optical pod scanning system comprising: the device comprises an optical camera, a laser range finder, a servo frame, an image processing module, a base and a POS module. The optical camera includes: the camera comprises a body, a lens and a quick reflector.

The optical camera is in a cuboid shape and is arranged in the servo frame inner platform body to realize the image acquisition function. The camera body is fixedly connected with the camera lens; the fast reflector is a circular lens with motor drive, and is arranged in the lens to realize the image stabilization function under the swinging condition.

The embedded optical lens of laser range finder is the cuboid shape, arranges servo frame inner stage body in, realizes the range finding to ground through transmitting laser. The servo frame is spherical, the base is cylindrical, and the servo frame is connected with the base through a flange.

The POS module has the functions of: and outputting the attitude angle and the geographical position of the pod through the measurement of the gyroscope, the adding table and the satellite receiver.

The image processing module has the functions of: and receiving the image from the optical camera, and finishing image processing and data calculation.

Second step image processing module determines scan strip width and swing angular velocity

Before the photoelectric pod scanning system enters a scanning area, the image processing module determines the width of a scanning strip and the swinging angular speed.

The swath width L in the parallel flight direction is determined using equation (1):

L＝D×ε/sinθ (1)

in the formula (1), D is the flying height from the ground, epsilon is the longitudinal axis view field of the optical camera, and theta is the included angle between the view axis and the horizontal ground.

Determining the servo azimuth frame yaw angular velocity ω using equation (2):

in the formula (2), m is the imaging frequency in the process of swinging from one side to the other side, gamma is the field of view of a transverse shaft of the optical camera, k is the overlapping coefficient of adjacent images, upsilon is the flight speed of the photoelectric pod scanning system, and L is the flight distance in one time of swinging the azimuth frame left and right.

The swath width W in the vertical flight direction is determined using equation (3):

third step fast reflector compensation image motion

After the servo frame continuously swings, the fast reflector swings in the opposite direction when the optical camera is exposed, image motion is compensated, and the scenery is clearly imaged.

Fourth, the POS module determines the position information

The POS module collects gyro and acceleration information and carries out combined navigation; satellite signals are collected, inertia/satellite coupling is determined, and longitude, latitude and elevation information is obtained.

The fifth step is that the image processing module positions the ground area corresponding to the single frame image

And determining the azimuth and the pitch angle of the visual axis in the inertial space by the image processing module according to the longitude, latitude and elevation information and the azimuth and the pitch angle of the servo frame. Let the zero position coordinate of the viewing axis in the image be O (x0, y0), the upper left corner point coordinate be A (0, 0), the lower left corner point coordinate be B (0, H _ pix), the lower right corner point coordinate be C (W _ pix, H _ pix), and the upper right corner point coordinate be D (W _ pix, 0). W _ pix, H _ pix are effective pixels in the image width and height directions, respectively.

Azimuthal deviation ψ of point a with respect to point O_AExpressed by equation (4):

pitch offset of point a relative to point OExpressed by equation (5):

in formula (5), f represents the focal length of the optical camera, and d represents the pixel size. The zero position direction of the visual axis is taken as the X axis, the direction represented by the image ordinate is the Y axis, and the attitude matrix of the optical camera is

Attitude matrix of inertial system rotating to ground scenery direction pointed by point AExpressed by equation (6):

therefore, the azimuth and the pitch angle of the ground scene direction pointed by the point A in the inertial space are obtained from the attitude matrix. And the azimuth and the pitch angle of the ground scenery direction pointed by B, C and the D point in the inertial space are deduced in the same way, and the longitude, the latitude and the elevation of the visual axis zero position and the ground scenery pointed by 4 angular points are determined according to the distance information of the laser range finder, so that the positioning of the ground area corresponding to the single-frame image is realized.

Sixthly, the image processing module completes image splicing of the scanning area

The image processing module determines the overlapping coefficient of adjacent areas according to the side length overlapping condition of the adjacent images for the ground area position information corresponding to the multi-frame images in the recording and scanning process, and judges whether the adjacent areas are connected seamlessly. When the adjacent areas are overlapped less, the swing angular speed is reduced; when adjacent regions overlap more, the rocking angular velocity is increased. The boundaries of the scanned area are gradually enlarged by area accumulation. When the geographic coordinates of the scanned area relative to the target area at the four endpoints of northeast, southeast, southwest and northwest are more northeast, southeast, southwest and northwest, indicating that the target area is completely covered, the electro-optical pod scanning is completed.

And completing the photoelectric pod scanning under the low-altitude fast flight condition.

The invention adopts the fast reflector to compensate the image motion to realize continuous sweep, realizes the full coverage of the area by single-frame image positioning and multi-frame image seamless splicing, solves the defects of long time, omission of the area and the like of the traditional scanning method under the condition of low-altitude fast flight, and improves the searching efficiency.

Detailed Description

The photoelectric pod scanning method under the low-altitude rapid flight condition comprises the following specific steps:

first step building photoelectric pod scanning system

An electro-optical pod scanning system comprising: the device comprises an optical camera, a laser range finder, a servo frame, an image processing module, a base and a POS module. The optical camera includes: the camera comprises a body, a lens and a quick reflector.

The optical camera is in a cuboid shape and is arranged in the servo frame inner platform body to realize the image acquisition function. The camera body is fixedly connected with the camera lens; the fast reflector is a circular lens with motor drive, and is arranged in the lens to realize the image stabilization function under the swinging condition.

The embedded optical lens of laser range finder is the cuboid shape, arranges servo frame inner stage body in, realizes the range finding to ground through transmitting laser. The servo frame is spherical, the base is cylindrical, and the servo frame is connected with the base through a flange.

The POS module has the functions of: and outputting the attitude angle and the geographical position of the pod through the measurement of the gyroscope, the adding table and the satellite receiver.

The image processing module has the functions of: and receiving the image from the optical camera, and finishing image processing and data calculation.

Second step image processing module determines scan strip width and swing angular velocity

Before the photoelectric pod scanning system enters a scanning area, the image processing module determines the width of a scanning strip and the swinging angular speed.

The swath width L in the parallel flight direction is determined using equation (1):

L＝D×ε/sinθ (1)

in the formula (1), D is the flying height from the ground, epsilon is the longitudinal axis view field of the optical camera, and theta is the included angle between the view axis and the horizontal ground.

Determining the servo azimuth frame yaw angular velocity ω using equation (2):

in the formula (2), m is the imaging frequency in the process of swinging from one side to the other side, gamma is the field of view of a transverse shaft of the optical camera, k is the overlapping coefficient of adjacent images, upsilon is the flight speed of the photoelectric pod scanning system, and L is the flight distance in one time of swinging the azimuth frame left and right.

The swath width W in the vertical flight direction is determined using equation (3):

third step fast reflector compensation image motion

After the servo frame continuously swings, the fast reflector swings in the opposite direction when the optical camera is exposed, image motion is compensated, and the scenery is clearly imaged.

Fourth, the POS module determines the position information

The POS module collects gyro and acceleration information and carries out combined navigation; satellite signals are collected, inertia/satellite coupling is determined, and longitude, latitude and elevation information is obtained.

The fifth step is that the image processing module positions the ground area corresponding to the single frame image

And determining the azimuth and the pitch angle of the visual axis in the inertial space by the image processing module according to the longitude, latitude and elevation information and the azimuth and the pitch angle of the servo frame. Let the zero position coordinate of the viewing axis in the image be O (x0, y0), the upper left corner point coordinate be A (0, 0), the lower left corner point coordinate be B (0, H _ pix), the lower right corner point coordinate be C (W _ pix, H _ pix), and the upper right corner point coordinate be D (W _ pix, 0). W _ pix, H _ pix are effective pixels in the image width and height directions, respectively.

Azimuthal deviation ψ of point a with respect to point O_AExpressed by equation (4):

pitch offset of point a relative to point OExpressed by equation (5):

in formula (5), f represents the focal length of the optical camera, and d represents the pixel size. The zero position direction of the visual axis is taken as the X axis, the direction represented by the image ordinate is the Y axis, and the attitude matrix of the optical camera is

Attitude matrix of inertial system rotating to ground scenery direction pointed by point AExpressed by equation (6):

therefore, the azimuth and the pitch angle of the ground scene direction pointed by the point A in the inertial space are obtained from the attitude matrix. And the azimuth and the pitch angle of the ground scenery direction pointed by B, C and the D point in the inertial space are deduced in the same way, and the longitude, the latitude and the elevation of the visual axis zero position and the ground scenery pointed by 4 angular points are determined according to the distance information of the laser range finder, so that the positioning of the ground area corresponding to the single-frame image is realized.

Sixthly, the image processing module completes image splicing of the scanning area

The image processing module determines the overlapping coefficient of adjacent areas according to the side length overlapping condition of the adjacent images for the ground area position information corresponding to the multi-frame images in the recording and scanning process, and judges whether the adjacent areas are connected seamlessly. When the adjacent areas are overlapped less, the swing angular speed is reduced; when adjacent regions overlap more, the rocking angular velocity is increased. The boundaries of the scanned area are gradually enlarged by area accumulation. When the geographic coordinates of the scanned area relative to the target area at the four endpoints of northeast, southeast, southwest and northwest are more northeast, southeast, southwest and northwest, indicating that the target area is completely covered, the electro-optical pod scanning is completed.

And completing the photoelectric pod scanning under the low-altitude fast flight condition.