CN102831751B - Road high-dangerous slope monitoring method based on double-camera imaging technology - Google Patents

Abstract

The invention discloses a road high-dangerous slope monitoring method based on a double-camera imaging technology, which belongs to the technical field of the slope monitoring method, and the method comprises the following steps of three-dimensionally rebuilding a slope: installing cameras along the length direction of the slope to be monitored, and ensuring that each scheduled monitoring point simultaneously stays within a photography range of two cameras; (2) monitoring the slope: calculating a space coordinate of the monitoring points which are distributed on the slope according to images photographed by the two cameras to obtain a space coordinate set of the monitoring points, periodically calculating a Euclidean distance of three-dimensional spaces of every two adjacent monitoring points, and monitoring the variation situation of the slope; and (3) predicting slope disaster: comparing a monitored slope displacement or slide value with a preset allowed critical value, and processing according to a preset rule. The road high-dangerous slope monitoring method is simple in structure, high in measurement and control precision and low in cost and used for monitoring the high-dangerous slope.

Description

Highway High and dangerous slope method for supervising based on twin camera imaging technique

Technical field

The present invention relates to a kind of slope monitoring method, more particularly, relate in particular to a kind of highway High and dangerous slope method for supervising based on twin camera imaging technique.

Background technology

At present, highway in China is built total amount and has been realized great-leap-forward growth, pressure on transport situation overall relief, on built highway, High and dangerous slope quantity is many, slope failure is one of main Highway Geological Disaster, and widely distributed, occur frequently, very easily cause interruption of communication, directly affect traffic safety.

The problem that highway High and dangerous slope monitoring at present exists is: in side slope, bury sensor underground, the displacement situation of monitoring side slope, in rainy season, this method is difficult to the slope of highway disaster that discovery causes due to damaged or destroyed by flood in time, and data processing personnel need to possess certain professional knowledge.

Summary of the invention

The object of the present invention is to provide a kind of simple in structure, observing and controlling precision is high and the lower-cost highway High and dangerous slope method for supervising based on twin camera imaging technique.

Technical scheme of the present invention is achieved in that a kind of highway High and dangerous slope method for supervising based on twin camera imaging technique, and wherein the method comprises the steps:

(1) side slope three-dimensional is rebuild, and along length of slope direction distributing installation video camera to be monitored, each predetermined monitoring point is in the coverage of two video cameras simultaneously;

(2) slope monitoring, the volume coordinate of the control point of laying in the image calculation side slope of taking according to two video cameras, obtain the volume coordinate point set of control point, regularly calculate the Euclidean distance between the three dimensions of adjacent two control points, monitor domatic situation of change;

(3) side slope hazard forecasting, the slope displacement that monitoring is obtained or slip value and the default critical value allowing compare, and process according to preset rules.

In the above-mentioned highway High and dangerous slope method for supervising based on twin camera imaging technique, the described side slope three-dimensional of step (1) is rebuild, specifically further comprising the steps of:

(a) image of taking according to two video cameras adopts Tsai two step standardizations to build three-dimensional environment, utilize linear transformation method or perspective projection transformation Matrix Solving camera parameters, and calculate orthogonal rotation matrix and the translation matrix of world coordinate system with respect to camera coordinate system;

(b) extract minutiae the image of taking from two video cameras, adopts transformation model and grab sample consistent method based on Rotation and Zoom invariant, and unique point is carried out to Corresponding matching, has realized slope of highway three-dimensional reconstruction.

In the above-mentioned highway High and dangerous slope method for supervising based on twin camera imaging technique, the slope monitoring that step (2) is described, specifically further comprising the steps of: to obtain after the volume coordinate of control point, rebuild three-dimensional slope table face by the method for triangle gridding or parametric surface or implicit surface, and then according to the Euclidean distance between the regular three dimensions that calculates adjacent two control points in three-dimensional side slope surface of rebuilding, monitor domatic situation of change.

In the above-mentioned highway High and dangerous slope method for supervising based on twin camera imaging technique, step (3) is described to be processed and is specially according to preset rules: use status predication according to monitoring result, when monitoring value is during lower than critical value, side slope is in a safe condition; In the time approaching critical value, side slope, in critical conditions, approaches danger, need monitor in real time; In the time exceeding critical value, side slope is in the hole, need take corresponding emergency measure.

The present invention adopts after said method, by twin camera, three-dimensional imaging is carried out in monitoring point, obtains the real scene of slope of highway, the spatial variations of putting by Monitoring and Controlling, the displacement or the sliding condition that calculate slope of highway, obtain in time side slope disaster information, and carry out hazard forecasting.Thereby can substantially realize real-time, controlled high precision monitor, various harm be resolved at shady bud state, can effectively carry out Real-Time Monitoring to side slope.

Brief description of the drawings

Below in conjunction with the embodiment in accompanying drawing, the present invention is described in further detail, but do not form any limitation of the invention.

Fig. 1 is the process flow diagram of the inventive method;

Fig. 2 is the laying schematic diagram of video camera of the present invention;

Fig. 3 (a) is the tessellated virgin state schematic diagram of papery in the specific embodiment of the invention;

Fig. 3 (b) is one of tessellated deformation process of papery in the specific embodiment of the invention;

Fig. 3 (c) is two of the interior tessellated deformation process of papery of the specific embodiment of the invention;

Fig. 3 (d) is three of the interior tessellated deformation process of papery of the specific embodiment of the invention;

Fig. 3 (e) is four of the interior tessellated deformation process of papery of the specific embodiment of the invention;

Fig. 3 (f) is the monitoring structure schematic diagram corresponding with Fig. 3 (a);

Fig. 3 (g) is the monitoring structure schematic diagram corresponding with Fig. 3 (b);

Fig. 3 (h) is the monitoring structure schematic diagram corresponding with Fig. 3 (c);

Fig. 3 is (i) the monitoring structure schematic diagram corresponding with Fig. 3 (d);

Fig. 3 (j) is the monitoring structure schematic diagram corresponding with Fig. 3 (e);

Embodiment

Consult shown in Fig. 1, the highway High and dangerous slope method for supervising based on twin camera imaging technique of the present invention, the method comprises the steps:

(1) side slope three-dimensional is rebuild, and concrete steps are:

(a) along length of slope direction distributing installation video camera to be monitored, each predetermined monitoring point is in the coverage of two video cameras simultaneously;

(b) image of taking according to two video cameras adopts Tsai two step standardizations to build three-dimensional environment, utilize linear transformation method or perspective projection transformation Matrix Solving camera parameters as initial value, and calculate orthogonal rotation matrix and the translation matrix of world coordinate system with respect to camera coordinate system;

(c) video camera is regularly taken the multi-source image of acquisition, need carry out the matching processing of data, extract minutiae from image, adopts transformation model and grab sample consistent method based on Rotation and Zoom invariant, unique point is carried out to Corresponding matching, realized slope of highway three-dimensional reconstruction.

(2) slope monitoring, is exactly mainly according to three-dimensional imaging principle, and concrete steps are:

(a), according to the volume coordinate of the control point of laying in the image calculation side slope of two video cameras shootings, obtain the volume coordinate point set of control point;

(b), according to the volume coordinate point set of the control point that obtains, rebuild three-dimensional slope table face by the method for triangle gridding or parametric surface or implicit surface;

(c) regularly calculate the Euclidean distance between the three dimensions of adjacent two control points, monitor domatic situation of change;

(3) side slope hazard forecasting, concrete steps are:

(a) critical value of slope displacement monitoring being obtained or slip value and permission compares.

(b) use status predication, when monitoring value is during lower than critical value, side slope is in a safe condition; In the time approaching critical value, side slope, in critical conditions, approaches danger, need monitor in real time; In the time exceeding critical value, side slope is in the hole, need take corresponding emergency measure, to reduce or to prevent the massive losses that slope instability brings.

The Some Related Algorithms of rebuilding about side slope three-dimensional below:

(1) orthogonal rotation matrix R and translation matrix T are:

$R=\left(\begin{array}{ccc}{r}_1& r_2& r_3\\ r_4& r_5& r_6\\ r_7& r_8& r_9\end{array}\right)---\left(1\right)$

$T=\left[\begin{array}{c}{t}_x\\ t_y\\ t_z\end{array}\right]---\left(2\right)$

(2) Corresponding matching of control point

The laying of video camera as shown in Figure 2, supposes that left video camera O-xyz is positioned at the initial point place of world coordinate system and without spin, left image coordinate is O _l-X _ly _l, focal length is f _l, right camera coordinates is O _r-x _ry _rz _r, image coordinate is O _r-X _ry _r, focal length is f _r.Two video cameras are taken the same control point P of slope of highway at synchronization, this coordinate under world coordinates is: (x, y, z), the coordinate under right camera coordinate system is (x _r, y _r, z _r), be respectively (X at the coordinate of the left and right plane of delineation _l, Y _l), (X _r, Y _r).Have according to video camera Perspective transformation model:

$S_l\left[\begin{array}{c}{X}_l\\ Y_l\\ 1\end{array}\right]=\left[\begin{array}{ccc}{f}_l& 0& 0\\ 0& {\mathrm{<figure-callout\; id="0"\; label="f\; l"\; filenames="HDA00002097789100012.png"\; state="\{\{state\}\}">f</figure-callout>}}_l& 0\\ 0& 0& 1\end{array}\right]\left[\begin{array}{c}x\\ y\\ z\end{array}\right]---\left(3\right)$

$S_r\left[\begin{array}{c}{X}_r\\ Y_r\\ 1\end{array}\right]=\left[\begin{array}{ccc}{\mathrm{<figure-callout\; id="0"\; label="f\; r"\; filenames="HDA00002097789100012.png"\; state="\{\{state\}\}">f</figure-callout>}}_r& 0& 0\\ 0& {\mathrm{<figure-callout\; id="0"\; label="f\; r"\; filenames="HDA00002097789100012.png"\; state="\{\{state\}\}">f</figure-callout>}}_r& 0\\ 0& 0& 1\end{array}\right]\left[\begin{array}{c}{x}_r\\ y_r\\ z_r\end{array}\right]---\left(4\right)$

O _r-x _ry _rz _rand can be expressed as by space transition matrix M between O-xyz

$\left[\begin{array}{c}{x}_r\\ y_r\\ z_r\end{array}\right]=M\left[\begin{array}{c}x\\ y\\ z\\ 1\end{array}\right]=\left[\begin{array}{cccc}{r}_1& r_2& r_3& t_x\\ r_4& r_5& r_6& t_y\\ r_7& r_8& r_9& t_z\end{array}\right]\left[\begin{array}{c}x\\ y\\ z\\ 1\end{array}\right],M=\left[R\right|T]---\left(5\right)$

In formula, $R=\left[\begin{array}{ccc}{r}_1& r_2& r_3\\ r_4& r_5& r_6\\ r_7& r_8& r_9\end{array}\right]$ ， $T=\left[\begin{array}{c}{t}_x\\ t_y\\ t_z\end{array}\right]$ , be O _r-x _ry _rz _rand the translation transformation vector between rotation matrix and initial point between O-xyz.

The external parameter of supposing left and right cameras is respectively left video camera R _l, T _lwith right video camera R _r, T _r, for any spatial point, suppose that its inhomogeneous coordinate under world coordinate system, left camera coordinate system and right camera coordinate system is respectively x _w, x _l, x _r, have:

$x_l=R_l{x}_w+T_l,x_r=R_r{x}_w+T_r---\left(6\right)$

Cancellation x _w, have

.Therefore, the geometric relationship R between two video cameras, T can be expressed as:

$R=R_r{R}_l^{-1},T=T_r-R_r{R}_l^{-1}{T}_l---\left(7\right)$

By formula (3), (4), (5) have, and for the spatial point in O-xyz coordinate system, the corresponding relation between 2 video camera image planes points is:

$S_r\left[\begin{array}{c}{X}_r\\ Y_r\\ 1\end{array}\right]=\left[\begin{array}{cccc}{f}_r{r}_1& f_r{r}_2& f_r{r}_3& f_r{t}_x\\ f_r{r}_4& f_r{r}_5& f_r{r}_6& f_r{t}_y\\ r_7& r_8& r_9& t_z\end{array}\right]\left[\begin{array}{c}z{X}_l/f_l\\ z{X}_l/f\\ 1\end{array}\right]---\left(8\right)$

Thus, spatial point three-dimensional coordinate can be expressed as:

$\begin{array}{c}x=z{X}_l/f_l\\ y=z{Y}_l/f_l\\ z=\frac{f_l(f_r{t}_x-X_r{t}_z)}{X_r(r_7{X}_l+r_8{Y}_l+f_l{r}_9)-f_r(r_1{X}_l+r_2{Y}_l+f_l{r}_3)}\\ =\frac{f_l(f_r{t}_y-Y_r{t}_z)}{Y_r(r_7{X}_l+r_8{Y}_l+f_l{r}_9)-f_r(r_4{X}_l+r_5{Y}_l+f_l{r}_6)}\end{array}---\left(9\right)$

The control point volume coordinate calculating with above-mentioned formula.

The computing formula L of the Euclidean distance between the three dimensions of adjacent two control points is:

$L=\sqrt{{(x_1-{x^{\&prime;}}_1)}^2+{(x_2-{x^{\&prime;}}_2)}^2+...+{(x_n-{x^{\&prime;}}_n)}^2}---\left(10\right)$

X ₁, x ₂... x _nfor having laid the three-dimensional point set of control point, x in side slope ₁＇, x ₂＇ ... x _n＇ is the point set of last corresponding control point.

Experimental example

For the feasibility of checking this patent application in slope of highway monitoring, adopting papery gridiron pattern is Deformation Experiments material, and each grid is of a size of 30*30 millimeter, by manually its edge being pushed, and the tessellated deformation of monitoring papery.Simulation process is if Fig. 3 (a) is to 3(j) as shown in, wherein Fig. 3 (a) and (b), (c), (d), (e) are respectively the tessellated deformation process of interior papery, Fig. 3 (f), (g), (h), (i), (j) be respectively the monitored results of answering in contrast, black point set is the three-dimensional point set of the control point of laying, and other Dark grey region is to rebuild the triangle mesh curved surface obtaining.Through error analysis, the average error of this experiment is about 2 millimeters.

Above-described embodiment is used for the present invention that explains, instead of limits the invention, and in the protection domain of spirit of the present invention and claim, the present invention is made to any amendment and change, all falls into protection scope of the present invention.

Claims (3)

1. the highway High and dangerous slope method for supervising based on twin camera imaging technique, is characterized in that, the method comprises the steps:

(1) side slope three-dimensional is rebuild, and concrete steps are:

(a) along length of slope direction distributing installation video camera to be monitored, each predetermined monitoring point is in the coverage of two video cameras simultaneously;

(b) image of taking according to two video cameras adopts Tsai two step standardizations to build three-dimensional environment, utilize linear transformation method or perspective projection transformation Matrix Solving camera parameters as initial value, and calculate orthogonal rotation matrix and the translation matrix of world coordinate system with respect to camera coordinate system;

(c) video camera is regularly taken the multi-source image of acquisition, need carry out the matching processing of data, extract minutiae from image, adopts transformation model and grab sample consistent method based on Rotation and Zoom invariant, unique point is carried out to Corresponding matching, realized slope of highway three-dimensional reconstruction;

(2) slope monitoring, the volume coordinate of the control point of laying in the image calculation side slope of taking according to two video cameras, obtain the volume coordinate point set of control point, regularly calculate the Euclidean distance between the three dimensions of adjacent two control points, monitor domatic situation of change;

(3) side slope hazard forecasting, the slope displacement that monitoring is obtained or slip value and the default critical value allowing compare, and process according to preset rules.

2. the highway High and dangerous slope method for supervising based on twin camera imaging technique according to claim 1, it is characterized in that, the slope monitoring that step (2) is described, specifically further comprising the steps of: to obtain after the volume coordinate of control point, rebuild three-dimensional slope table face by the method for triangle gridding or parametric surface or implicit surface, and then according to the Euclidean distance between the regular three dimensions that calculates adjacent two control points in three-dimensional side slope surface of rebuilding, monitor domatic situation of change.

3. according to the highway High and dangerous slope method for supervising based on twin camera imaging technique according to claim 1, it is characterized in that, step (3) is described to be processed and is specially according to preset rules: use status predication according to monitoring result, when monitoring value is during lower than critical value, side slope is in a safe condition; In the time approaching critical value, side slope, in critical conditions, approaches danger, need monitor in real time; In the time exceeding critical value, side slope is in the hole, need take corresponding emergency measure.

Images