CN104732557A - Color point cloud generating method of ground laser scanner - Google Patents

Abstract

The invention discloses a color point cloud generating method of a ground laser scanner. The color point cloud generating method includes arranging a camera externally on the basis of the ground laser scanner, calibrating internal parameters of the camera by a space resection method and external parameters of the camera by a direct linear transformation method, determining internal and external direction elements by a calibration method, accurately controlling angle of scene images acquired by the camera by a scanner rotating platform, and finally calculating successive transformational relation by the formula. By the color point cloud generating method, point cloud of field data acquisition and image texture information are accurately integrated and matched, so that color point cloud is automatically acquired.

Description

A kind of terrestrial Laser scanner colour point clouds generation method

Technical field

The present invention relates to photogrammetric and three-dimensional laser technical field, specifically a kind of terrestrial Laser scanner colour point clouds generation method.

Background technology

Terrestrial Laser scanner can the three-dimensional point cloud of quick obtaining scene, the three-dimensional reconstruction that the data utilizing laser scanner to obtain carry out target has broad application prospects in fields such as ancient building, historical relic's protection, streetscape productions, but the information of some cloud is only limitted to the information such as coordinate, intensity, the RGB information of texture is generally given a some cloud by the acquisition of texture information after image and some cloud being mated by certain means after the image of digital camera acquisition scene, namely obtains colour point clouds; Colour point clouds incorporates the 3D structural information of a cloud and the texture information of image, contributes to the expression of the decipher to cloud data, Object identifying, classification and minutia, can raise the efficiency in the subsequent applications such as point cloud classifications, three-dimensional modeling.

In order to obtain the texture information of testee, the image data of scene is gathered while needing to obtain cloud data, and by certain computing method, the space coordinates of laser scanner and the coordinate systems in image of image data are unified, Here it is some cloud and the registration work of image; The key issue of registration be tied to a cloud as the coordinate of plane space coordinates between the asking for of coordinate conversion parameter R-rotation matrix and t-translation vector; At present, general method is on a cloud and image, manually choose the reference mark of the same name of some respectively, selected number of control points of the same name is more than or equal to 3, conversion parameter is calculated by this group same place, realize the unification of two coordinate systems, thus reaching the object of registration, some cloud can not be synchronous with the registration of impact.

Summary of the invention

The object of the present invention is to provide a kind of terrestrial Laser scanner colour point clouds generation method, to solve the problem proposed in above-mentioned background technology.

For achieving the above object, the invention provides following technical scheme:

A kind of terrestrial Laser scanner colour point clouds generation method, by external camera on terrestrial Laser scanner basis, demarcation is carried out to camera inside and outside parameter, controls the angle of camera acquisition scene image and subsequent conversion relation is calculated, realize the automatic generation of colour point clouds, comprise the following steps:

(1) demarcation of camera intrinsic parameter

After adopting resection method to take some images in controlling filed, image is chosen some reference mark and goes out the object coordinates of reference mark at same coordinate system by total station survey; X will be obtained after calibration of camera internal parameters ₀(principal point x coordinate), y ₀(principal point y coordinate), f _xf _y(focal length), k ₁k ₂(radial distortion parameter), p ₁p ₂parameters such as (tangential distortion parameters);

(2) demarcation of camera and scanner center relative position

Reference mark of the same name chosen respectively by the some cloud adopting direct linear transformation (DLT) method to gather in Calibration Field and image, is obtained six elements of exterior orientation (X of camera by the computing method of DLT ₀, Y ₀, Z ₀, roll, pitch, yaw), X ₀, Y ₀and Z ₀be called the line element of Camera extrinsic number; Roll, pitch and yaw are called the angle element of camera elements of exterior orientation, and the computing formula of DLT is as follows:

$x+\frac{l_{1}X+l_{2}Y+l_{3}Z+l_4}{l_{9}X+l_{10}Y+l_{11}Z+1}=0$

$y+\frac{l_{5}X+l_{6}Y+l_{7}Z+l_8}{l_{9}X+l_{10}Y+l_{11}Z+1}=0$

Wherein, l ₁... l ₁₁for the function of camera internal and external orientation;

(3) scanner rotation platform control camera is taken pictures and recording angular parameter

Scanner device is carrying out the collection of a cloud and image data after the demarcation of inside and outside parameter, and the drainage pattern that terrestrial Laser scanner LS300 sets is the scanning first carrying out a cloud, carries out the shooting of image after completing again; From X-axis, launch Article 1 sweep trace when scanner carries out the collection of a cloud, base platform gets back to X-axis after the rotation of 360 °; LS300 controls camera afterwards be that 60 °, 180 ° and 300 ° of three positions carry out taking and are recorded in image file by angle parameter with X-axis respectively, and in follow-up operating process, angle parameter is directly obtained by filename parsing; It is obtained by the angle element of three of camera outer parameters that rotation matrix calculates, and computing formula is as follows:

$R=\left[\begin{array}{ccc}{a}_1& b_1& c_1\\ a_2& b_2& c_2\\ a_3& b_3& c_3\end{array}\right]$

Wherein,

α ₁＝cosαcosγ-sinαsinβsinγ；α ₂＝-cosαsinγ-sinαsinβcosγ；α ₃＝-cosαcosβ；

b ₁＝cosβsinγ；b ₂＝cosβcosγ；b ₃＝sinβ；c ₁＝sinαcosγ+cosαsinβsinγ；

c ₂＝-sinαsinγ+cosαsinβcosγ；c ₂＝cosαcosβ；

(4) solution asks rotation matrix, and asks the pixel coordinate on image that a cloud is corresponding according to collinearity equation solution

After obtaining rotation matrix, in three dimensions, namely each object space point can be mapped with a pixel coordinate, a pixel coordinate corresponding to each some cloud is obtained by collinearity equation, convolution algorithm is carried out according to interpolation method by this pixel coordinate, obtain a rgb value, namely complete the making of colour point clouds after rgb value is assigned to a cloud, collinearity equation formula is as follows:

$\left[\begin{array}{c}{x}_0\\ y_0\\ -f\end{array}\right]=R(\mathrm{\α},\mathrm{\β},\mathrm{\γ})\left[\begin{array}{c}X\\ Y\\ Z\end{array}\right]+\left[\begin{array}{c}{t}_1\\ t_2\\ t_3\end{array}\right]$

Compared with prior art, the invention has the beneficial effects as follows: the coordinate position in camera and scanner is fixed by the present invention on LS300 terrestrial Laser scanner basis, simultaneously by the angle that motor control camera is taken pictures, also namely determining takes pictures photographs the position of light and attitude instantaneously; In advance the intrinsic parameter of camera is demarcated, achieved the autoregistration operation of a cloud and image, finally complete a fusion for cloud texture information, realize the automatic generation of colour point clouds.

Accompanying drawing explanation

Accompanying drawing 1 is the terrestrial Laser scanner LS300 outside drawing of independent research;

Accompanying drawing 2 is a cloud data dynamic interior of building being scanned to rear acquisition;

Accompanying drawing 3-5 is three images that camera goes out to obtain three angles;

Accompanying drawing 6 is calibration of camera internal parameters software interface;

Accompanying drawing 7 is the space coordinates of scanner inside definition;

Accompanying drawing 8 is camera inside and outside parameter;

Accompanying drawing 9 is the colour point clouds generated.

Embodiment

Below in conjunction with the accompanying drawing in the embodiment of the present invention, be clearly and completely described the technical scheme in the embodiment of the present invention, obviously, described embodiment is only the present invention's part embodiment, instead of whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art, not making the every other embodiment obtained under creative work prerequisite, belong to the scope of protection of the invention.

Refer to Fig. 1-9, a kind of terrestrial Laser scanner colour point clouds generation method, by external camera on terrestrial Laser scanner basis, demarcation is carried out to camera inside and outside parameter, controls the angle of camera acquisition scene image and subsequent conversion relation is calculated, realize the automatic generation of colour point clouds, comprise the following steps:

(1) demarcation of camera intrinsic parameter

As shown in Figure 1, equipment adopts LS300 terrestrial Laser scanner, external camera on LS300 terrestrial Laser scanner basis, after adopting resection method to take some images in controlling filed, image is chosen some reference mark and goes out the object coordinates of reference mark at same coordinate system by total station survey; X will be obtained after calibration of camera internal parameters ₀(principal point x coordinate), y ₀(principal point y coordinate), f _xf _y(focal length), k ₁k ₂(radial distortion parameter), p ₁p ₂parameters such as (tangential distortion parameters);

(2) demarcation of camera and scanner center relative position

Reference mark of the same name chosen respectively by the some cloud adopting direct linear transformation (DLT) method to gather in Calibration Field and image, is obtained six elements of exterior orientation (X of camera by the computing method of DLT ₀, Y ₀, Z ₀, roll, pitch, yaw), X ₀, Y ₀and Z ₀be called the line element of Camera extrinsic number; Roll, pitch and yaw are called the angle element of camera elements of exterior orientation, and the computing formula of DLT is as follows:

$x+\frac{l_{1}X+l_{2}Y+l_{3}Z+l_4}{l_{9}X+l_{10}Y+l_{11}Z+1}=0$

$y+\frac{l_{5}X+l_{6}Y+l_{7}Z+l_8}{l_{9}X+l_{10}Y+l_{11}Z+1}=0$

Wherein, l ₁... l ₁₁for the function of camera internal and external orientation;

(3) scanner rotation platform control camera is taken pictures and recording angular parameter

Scanner device is carrying out the collection of a cloud and image data after the demarcation of inside and outside parameter, and the drainage pattern that terrestrial Laser scanner LS300 sets is the scanning first carrying out a cloud, carries out the shooting of image after completing again; The coordinate system of scanner inside as shown in Figure 7, launches Article 1 sweep trace when scanner carries out the collection of a cloud from X-axis, and base platform gets back to X-axis after the rotation of 360 °; LS300 controls camera afterwards be that 60 °, 180 ° and 300 ° of three positions carry out taking and are recorded in image file by angle parameter with X-axis respectively, and in follow-up operating process, angle parameter is directly obtained by filename parsing; It is obtained by the angle element of three of camera outer parameters that rotation matrix calculates, and computing formula is as follows:

$R=\left[\begin{array}{ccc}{a}_1& b_1& c_1\\ a_2& b_2& c_2\\ a_3& b_3& c_3\end{array}\right]$

Wherein,

α ₁＝cosαcosγ-sinαsinβsinγ；α ₂＝-cosαsinγ-sinαsinβcosγ；α ₃＝-cosαcosβ；

b ₁＝cosβsinγ；b ₂＝cosβcosγ；b ₃＝sinβ；c ₁＝sinαcosγ+cosαsinβsinγ；

c ₂＝-sinαsinγ+cosαsinβcosγ；c ₂＝cosαcosβ；

(4) solution asks rotation matrix, and asks the pixel coordinate on image that a cloud is corresponding according to collinearity equation solution

After obtaining rotation matrix, in three dimensions, namely each object space point can be mapped with a pixel coordinate, a pixel coordinate corresponding to each some cloud is obtained by collinearity equation, convolution algorithm is carried out according to interpolation method by this pixel coordinate, obtain a rgb value, namely complete the making of colour point clouds after rgb value is assigned to a cloud, collinearity equation formula is as follows:

$\left[\begin{array}{c}{x}_0\\ y_0\\ -f\end{array}\right]=R(\mathrm{\α},\mathrm{\β},\mathrm{\γ})\left[\begin{array}{c}X\\ Y\\ Z\end{array}\right]+\left[\begin{array}{c}{t}_1\\ t_2\\ t_3\end{array}\right]$

Embodiment 1

The external camera model selecting erection is Nikon D700, by taking multiple photographs of multiple attitude after layout calibration field, calculates intrinsic parameter after choosing reference mark, chooses reference mark and calculates interface and see accompanying drawing 6; Namely the staking-out work of all parameters is completed after the demarcation of scanner center relative position completing camera;

Accompanying drawing 2 is the some clouds adopting LS300 one place's buildings to be scanned to rear acquisition, has scanned the image that rear control external camera takes three angle positions, has seen accompanying drawing 3-5; The Parameter File of scanner synchronous recording is shown in accompanying drawing 6, and the parameter according to these records calculates rotation matrix and the translation parameters of often opening photo:

$R=\left[\begin{array}{ccc}{a}_1& b_1& c_1\\ a_2& b_2& c_2\\ a_3& b_3& c_3\end{array}\right],T=\left[\begin{array}{c}{t}_1\\ t_2\\ t_3\end{array}\right]$

Each some cloud in cloud data calculates a corresponding location of pixels by collinearity equation, generates the colour point clouds in accompanying drawing 9.

Principle of the present invention: for texture information that acquisition point cloud is corresponding has to pass through the registration of a cloud and image, the elements of exterior orientation (X of image ₀, Y ₀, Z ₀, α, beta, gamma) be camera shooting moment position and attitude.Wherein X ₀, Y ₀, Z ₀represent the coordinate of photo centre in reference frame, using the reference frame of the coordinate system of scanner as camera; α, beta, gamma describes three angle elements in photographic light flux space; Point cloud and the method for registering of image mainly solve the spatial transformation parameter between cloud and image, mapping mode comprises rigid transformation and non-rigid transformation two kinds, the present invention selects rigid transformation, namely put between cloud and image and only there is rotation, Pan and Zoom conversion, its transforming function transformation function relation is expressed as follows:

$\left[\begin{array}{c}{x}_0\\ y_0\\ -f\end{array}\right]=R(\mathrm{\α},\mathrm{\β},\mathrm{\γ})\left[\begin{array}{c}X\\ Y\\ Z\end{array}\right]+\left[\begin{array}{c}{t}_1\\ t_2\\ t_3\end{array}\right]$

Wherein, (x ₀, y ₀,-f) ^tthe three-dimensional coordinate of the image space coordinate system that texture is corresponding, (X, Y, Z) ^tbe the three-dimensional coordinate in a cloud, R (α, beta, gamma) is the rotation matrix between the image space auxiliary coordinates of image and the three-dimensional system of coordinate at scanner center; For obtaining the transformational relation of image space coordinate system and scanner coordinate system, must by the method determination internal and external orientation of demarcating, the intrinsic parameter of camera is demarcated and is adopted space resection's method, the external parameters calibration of camera adopts direct linear transformation's method, the shooting angle of camera is accurately controlled by scanner rotation platform, a cloud and texture information can accurately merge by the some cloud of field data acquisition and image after operating process of the present invention, realize the automatic acquisition of colour point clouds.

To those skilled in the art, obviously the invention is not restricted to the details of above-mentioned one exemplary embodiment, and when not deviating from spirit of the present invention or essential characteristic, the present invention can be realized in other specific forms.Therefore, no matter from which point, all should embodiment be regarded as exemplary, and be nonrestrictive, scope of the present invention is limited by claims instead of above-mentioned explanation, and all changes be therefore intended in the implication of the equivalency by dropping on claim and scope are included in the present invention.Any Reference numeral in claim should be considered as the claim involved by limiting.

In addition, be to be understood that, although this instructions is described according to embodiment, but not each embodiment only comprises an independently technical scheme, this narrating mode of instructions is only for clarity sake, those skilled in the art should by instructions integrally, and the technical scheme in each embodiment also through appropriately combined, can form other embodiments that it will be appreciated by those skilled in the art that.

Claims (2)

1. a terrestrial Laser scanner colour point clouds generation method, by external camera on terrestrial Laser scanner basis, demarcation is carried out to camera inside and outside parameter, controls the angle of camera acquisition scene image and subsequent conversion relation is calculated, realize the automatic generation of colour point clouds, it is characterized in that, comprise the following steps:

(1) demarcation of camera intrinsic parameter

After adopting resection method to take some images in controlling filed, image is chosen some reference mark and goes out the object coordinates of reference mark at same coordinate system by total station survey; X will be obtained after calibration of camera internal parameters ₀(principal point x coordinate), y ₀(principal point y coordinate), f _xf _y(focal length), k ₁k ₂(radial distortion parameter), p ₁p ₂parameters such as (tangential distortion parameters);

(2) demarcation of camera and scanner center relative position

Reference mark of the same name chosen respectively by the some cloud adopting direct linear transformation (DLT) method to gather in Calibration Field and image, is obtained six elements of exterior orientation (X of camera by the computing method of DLT ₀, Y ₀, Z ₀, roll, pitch, yaw), X ₀, Y ₀and Z ₀be called the line element of Camera extrinsic number; Roll, pitch and yaw are called the angle element of camera elements of exterior orientation, and the computing formula of DLT is as follows:

$x+\frac{l_{1}X+l_{2}Y+l_{3}Z+l_4}{l_{9}X+l_{10}Y+l_{11}Z+1}=0$

$y+\frac{l_{5}X+l_{6}Y+l_{7}Z+l_8}{l_{9}X+l_{10}Y+l_{11}Z+1}=0$

Wherein, l ₁... l ₁₁for the function of camera internal and external orientation;

(3) scanner rotation platform control camera is taken pictures and recording angular parameter

Scanner device is carrying out the collection of a cloud and image data after the demarcation of inside and outside parameter, and the drainage pattern that terrestrial Laser scanner LS300 sets is the scanning first carrying out a cloud, carries out the shooting of image after completing again; From X-axis, launch Article 1 sweep trace when scanner carries out the collection of a cloud, base platform gets back to X-axis after the rotation of 360 °; LS300 controls camera afterwards be that 60 °, 180 ° and 300 ° of three positions carry out taking and are recorded in image file by angle parameter with X-axis respectively, and in follow-up operating process, angle parameter is directly obtained by filename parsing; It is obtained by the angle element of three of camera outer parameters that rotation matrix calculates, and computing formula is as follows:

$R=\left[\begin{array}{ccc}{a}_1& b_1& c_1\\ a_2& b_2& c_2\\ a_3& b_3& c_3\end{array}\right]$

Wherein,

a ₁＝cosαcosγ-sinαsinβsinγ；a ₂＝-cosαsinγ-sinαsinβcosγ；a ₃＝-cosαcosβ；

b ₁＝cosβsinγ；b ₂＝cosβcosγ；b ₃＝sinβ；c ₁＝sinαcosγ+cosαsinβsinγ；

c ₂＝-sinαsinγ+cosαsinβcosγ；c ₂＝cosαcosβ；

(4) solution asks rotation matrix, and asks the pixel coordinate on image that a cloud is corresponding according to collinearity equation solution

After obtaining rotation matrix, in three dimensions, namely each object space point can be mapped with a pixel coordinate, a pixel coordinate corresponding to each some cloud is obtained by collinearity equation, convolution algorithm is carried out according to interpolation method by this pixel coordinate, obtain a rgb value, namely complete the making of colour point clouds after rgb value is assigned to a cloud, collinearity equation formula is as follows:

$\left[\begin{array}{c}{x}_0\\ y_0\\ -f\end{array}\right]=R(\mathrm{\α},\mathrm{\β},\mathrm{\γ})\left[\begin{array}{c}X\\ Y\\ Z\end{array}\right]+\left[\begin{array}{c}{t}_1\\ t_2\\ t_3\end{array}\right]$

2. terrestrial Laser scanner colour point clouds generation method according to claim 1, is characterized in that, the model of described terrestrial Laser scanner is LS300.