阅读说明：本技术 一种航拍图像任意像素点定位信息获取方法 (Method for acquiring positioning information of any pixel point of aerial image ) 是由 毕国玲 张星祥 聂婷 余达 张艳超 于 2019-07-23 设计创作，主要内容包括：一种航拍图像任意像素点定位信息获取方法,涉及光电成像技术领域,解决现有对无人机的航拍图像进行目标定位时,存在定位精度低的问题,包括坐标系的建立,根据飞机的航拍系统获得航拍图像,并根据惯导系统提供的飞行姿态信息,对航拍图像进行正射校正和求图像上任意点的经纬度三个步骤,通过本发明的方法有效地避免了双目立体视觉定位技术的复杂性和解决了单目相机定位精度低的问题,能够准确地获取图像上任意点的经纬度定位信息,为航拍系统在灾害救援、国土监察、资源调查、军事侦查等方面的应用奠定了基础。(A method for obtaining positioning information of any pixel point of an aerial image relates to the technical field of photoelectric imaging, and solves the problem of low positioning precision when the aerial image of an unmanned aerial vehicle is subjected to target positioning.)

1. A method for acquiring positioning information of any pixel point of an aerial image is characterized by comprising the following steps: the method is realized by the following steps:

firstly, establishing a ground coordinate system Og-XgYgZg, wherein the origin is Og, the positive direction of an Xg axis points to the true north direction of geography, the positive direction of a Zg axis is vertical to the ground plane and faces downwards, and the positive direction of a Yg axis is determined by a right-hand rule;

An included angle between the body coordinate system O-XYZ and the ground coordinate system is used as an attitude angle of the airplane, wherein a yaw angle psi is an included angle between a projection of a body axis OX on a horizontal plane OgXgYg and the ground axis OgXg, a pitch angle theta is an included angle between the body axis OX and the horizontal plane OgXgYg, and a roll angle phi is an angle formed by rotating the symmetrical plane of the airplane around the body axis OX;

Secondly, acquiring an aerial image by an aerial system of the airplane, and performing orthorectification on the aerial image according to flight attitude information provided by an inertial navigation system; the method specifically comprises the following steps:

firstly, establishing a conversion model from a ground coordinate system to a body coordinate system according to attitude information, wherein the conversion model is expressed by the following formula:

The model for converting the body coordinate system to the ground coordinate system is expressed by the following formula:

correcting and converting the aerial images from a body coordinate system to a ground coordinate system, wherein the row direction of the corrected images is the geographical north direction;

step three, acquiring the longitude and latitude of any point on the corrected image;

the longitude and latitude coordinates of the image center point Oa are known, the pixel coordinates of the image center point Oa on the image are (Xa, Ya), and the longitude and latitude coordinates are (Ja, Wa);

Any point B on the image has pixel coordinates of (Xb, Yb) and longitude and latitude coordinates of (Jb, Wb); the horizontal coordinate difference delta x between any two points Ao and B is Xb-Xa, the vertical coordinate difference delta y between two points Ao and B is Yb-Ya, and the azimuth angle gamma between the two points is calculated according to the following formula,

and judging quadrants where the deltax and the deltay are positioned to obtain a final azimuth angle gamma, which is expressed by the following formula:

from the camera intrinsic parameters, AoB is obtained the distance L between two points, which is expressed as:

In the formula, a is the pixel size, f is the focal length, and H is the flying height;

Projecting the distance L to the longitude direction as dx and to the latitude direction as dy, wherein dx and dy are respectively expressed by the following formula:

dx＝L*sin(γ)

dy＝L*cos(γ)

the corrected latitudinal translation radius length ex is expressed by the following equation:

ex＝Ej+(Er-Ej)*(90-Wa)/90

in the formula, Er is the equatorial radius, and Ej is the polar radius;

Obtaining the radius ed of the latitude circle where the Ao point is located, and expressing the radius ed as follows:

ed＝ex*cos(Wa*PI/180)

And acquiring longitude and latitude coordinates of any point B, and respectively expressing the longitude and latitude coordinates as follows:

Jb＝dx/ed*180/PI+Ja

Wb＝dy/ex*180/PI+Wa。

Technical Field

the invention relates to the technical field of photoelectric imaging, in particular to a method for acquiring positioning information of each pixel point of an aerial image.

background

with the continuous development of the unmanned aerial vehicle technology, the unmanned aerial vehicle is widely applied to disaster rescue, national and local supervision, environmental protection, resource investigation, military investigation and the like as an important tool for new generations. The target positioning by using the aerial image of the unmanned aerial vehicle becomes a research hotspot in the field of unmanned aerial vehicles, the position of a positioned target is calculated based on a binocular stereo vision monitoring target positioning technology, and the coordinate information of a target point of a camera coordinate system in a ground coordinate system is acquired based on the conversion among various coordinate systems in the aspect of target positioning shot by a single camera. Due to the fact that the unmanned aerial vehicle is complex in environment in the flying process, the pose angles (pitch angle, yaw angle and roll angle) of three directions and the difference of platforms are large, accurate positioning in an actual system is difficult to achieve.

disclosure of Invention

the invention provides a method for acquiring positioning information of each pixel point of an aerial image, aiming at solving the problem of low positioning precision when the aerial image of an unmanned aerial vehicle is subjected to target positioning.

A method for acquiring positioning information of each pixel point of an aerial image is realized by the following steps:

firstly, establishing a ground coordinate system Og-XgYgZg, wherein the origin is Og, the positive direction of an Xg axis points to the geographical north-north direction, the positive direction of a Zg axis is vertical to the ground plane and faces downwards, and the positive direction of a Yg axis is determined by a right-hand rule;

an included angle between the body coordinate system O-XYZ and the ground coordinate system is used as an attitude angle of the airplane, wherein a yaw angle psi is an included angle between a projection of a body axis OX on a horizontal plane OgXgYg and the ground axis OgXg, a pitch angle theta is an included angle between the body axis OX and the horizontal plane OgXgYg, and a roll angle phi is an angle formed by rotating the symmetrical plane of the airplane around the body axis OX;

secondly, acquiring an aerial image by an aerial system of the airplane, and performing orthorectification on the aerial image according to flight attitude information provided by an inertial navigation system; the method specifically comprises the following steps:

Firstly, establishing a conversion model from a ground coordinate system to a body coordinate system according to attitude information, wherein the conversion model is expressed by the following formula:

the model for converting the body coordinate system to the ground coordinate system is expressed by the following formula:

Converting and correcting an image acquired by aerial photography from a body coordinate system to a ground coordinate system, wherein the row direction of the corrected image is the geographical north direction;

Step three, acquiring the longitude and latitude of any point on the corrected image;

the longitude and latitude coordinates of the image center point Oa are known, the pixel coordinates of the image center point Oa on the image are (Xa, Ya), and the longitude and latitude coordinates are (Ja, Wa);

Any point B on the image has pixel coordinates of (Xb, Yb) and longitude and latitude coordinates of (Jb, Wb); the horizontal coordinate difference delta x between any two points Ao and B is Xb-Xa, the vertical coordinate difference delta y between two points Ao and B is Yb-Ya, and the azimuth angle gamma between the two points is calculated according to the following formula,

and judging quadrants where the deltax and the deltay are positioned to obtain a final azimuth angle gamma, which is expressed by the following formula:

From the camera intrinsic parameters, AoB is obtained the distance L between two points, which is expressed as:

in the formula, a is the pixel size, f is the focal length, and H is the flying height;

projecting the distance L to the longitude direction as dx and to the latitude direction as dy, wherein dx and dy are respectively expressed by the following formula:

dx＝L*sin(γ)

dy＝L*cos(γ)

the corrected latitudinal translation radius length ex is expressed by the following equation:

ex＝Ej+(Er-Ej)*(90-Wa)/90

in the formula, Er is the equatorial radius, and Ej is the polar radius;

obtaining the radius ed of the latitude circle where the Ao point is located, and expressing the radius ed as follows:

ed＝ex*cos(Wa*PI/180)

and acquiring longitude and latitude coordinates of any point B, and respectively expressing the longitude and latitude coordinates as follows:

Jb＝dx/ed*180/PI+Ja

Wb＝dy/ex*180/PI+Wa。

the invention has the beneficial effects that:

the method effectively avoids the complexity of a binocular stereo vision positioning technology and solves the problem of low positioning precision of the monocular camera, can accurately acquire longitude and latitude positioning information of any point on an image, and lays a foundation for the application of an aerial photography system in disaster relief, territorial surveillance, resource investigation, military reconnaissance and other aspects.

drawings

FIG. 1 is a schematic diagram illustrating a relationship between a body coordinate system O-XYZ and a ground coordinate system Og-XgYgZg in a method for acquiring positioning information of each pixel point of an aerial image according to the present invention;

FIG. 2 is an orthographic aerial photograph image corrected by pose information in an aerial photograph image pixel point positioning information acquisition method according to the present invention;

fig. 3 is a schematic diagram of a projection relationship between the distances between the Ao point and the B point in the latitude and longitude directions in the method for acquiring the positioning information of each pixel point of the aerial image according to the present invention.

Detailed Description