CN101819678A - Calibration method of three-dimensional virtual image of driving analog system - Google Patents

Abstract

The invention discloses a calibration method of a three-dimensional virtual image of a driving analog system, which completes the calibration of a target object in a virtual scene by respectively solving out a distance PL between every two projected target objects, a height PK of projected target objects and a height PH from the bottom of each projected target object to the ground based on a calibration template of a three-dimensional virtual scene of the driving analog system established by three-dimensional virtual image modeling software in the prior art, and completes the calibration of a displayed target object in the virtual scene by solving out the amplification factor of the displayed target object in the virtual scene. The invention can carry out calibration on object spatial dimension and effective sighting distance in the three-dimensional virtual image of the driving analog system in a faster, more comprehensive, more accurate and more objective way and simultaneously solves the problems of only individual-specificity and generality deficiency in the static virtual image calibration technology of the traditional driving analog system.

Description

The scaling method of three-dimensional virtual image of driving analog system

Technical field

The present invention relates to a kind of image scaling method, the scaling method of especially a kind of driving simulation system three-dimensional virtual image.

Background technology

Research driving simulation system is the drive simulating person-automobile-road traffic behavior of a kind of energy, and can make the driver obtain to feel close virtual emulation equipment with the real vehicle driving.Because in driving analog system, the driver is by the information of eyes acquisition more than 90%, so the technology of three-dimensional virtual image is a most important component in the drive simulation technology, must make driver's vision near real impression.

At present, three-dimensional virtual image in the driving analog system is to generate through computing machine, projector amplifies, a plurality of links such as screen demonstration are finished, therefore, virtual image that the driver sees and the actual bigger difference of scene existence, and randomness is bigger, particularly when using the multichannel image display system to show, the stitching error of its image can further strengthen.How to adjust image viewpoint and image transfer, the virtual image that the driver is seen approaches real situation, and corresponding scaling method must be arranged.But still do not have at present a cover complete three-dimensional virtual image of driving analog system is carried out the method that authenticity is demarcated.

The research driving analog system is that it is necessary for the driver and provides one to drive real as far as possible driving environment and impression with actual as a kind of prerequisite of scientific research instrument.Therefore, for the research driving analog system, must demarcate the authenticity of its three-dimensional virtual image.

Summary of the invention

Purpose of the present invention overcomes the prior art above shortcomings exactly, and a kind of scaling method of three-dimensional virtual image of driving analog system is provided.

The present invention is based on following principle:

One, in the three-dimensional virtual image of driving analog system, the demarcation that the bulk of object is carried out.

As shown in Figure 1, object (two mark post 1 and 11) is arranged on real border W, be L apart, the spacing of binocular viewpoint 3,31 is a, the distance of viewpoint 3,31 and real border W is S, between binocular 3,31 and real border W, establish driving analog system screen image, be empty border M, the distance of empty border M and binocular 3,31 is d, object on the real border W (two mark post 1 and 11) is projection target thing (two mark posts 2,21) in empty border M, level interval between the projection target thing (2,22) among the empty border M is PL, and then according to similar triangle theory, PL is suc as formula (1):

PL＝d×L/S+a-a×d/S????(1)

In like manner, as shown in Figure 5, the height of object 1 is H among the real border W, and it is projected as 2 in empty border M, and the standoff height in empty border M is made as PK, viewpoint is 3, viewpoint 3 is S to the distance of real border W object 1, and the distance of the projection target thing 2 of viewpoint 3 in the empty border M is d, and empty border M goes up projection object 2 bottoms and is made as PH to ground distance simultaneously, driver's viewpoint 3 is h to ground distance, then according to similar triangle theory respectively the computing formula of PK and PH is:

PK＝d×H/S??????(2)

PH＝h-h×d/S????(3)

Thus, driving analog system is the object with reality---road traffic scenery is presented on the screen image by virtual mode.Suppose road traffic scenery as shown in Figure 1 real border W and the object 1 just in the reality, driving analog system screen image is as shown in Figure 1 empty border M just, object 1,11 among the then real border W is a projection target thing 2,21 just in empty border M, and the driving analog system screen image that binocular 3,31 is seen is that the spacing PL of two projection target things 2,21 among the empty border M just should equal the PL value by formula (1) calculating; In like manner, the height PK of the projection target thing 2 in the driving analog system screen image just should equal the PK value by formula (2) calculating; Projection target thing 2 bottoms just should equal the pH value that calculated by formula (3) in the driving analog system screen image to ground distance PH.This has just guaranteed the authenticity of driving analog system screen image at horizontal space.

Two, in the three-dimensional virtual image of driving analog system, to the effectively demarcation carried out of sighting distance of object (mainly referring to traffic sign, label).

As shown in Figure 2, the display-object thing such as the traffic mark board 4 of numeral, literal or symbol arranged on its that in real border W a fixed measure to be arranged be t, driver's viewpoint 3 is SL with the distance of sign board 4.When viewpoint 3 is just seen the content of sign board 4 clearly, SL is the maximum sighting distance of vision in the reality, it is to be the Projection Display object among the empty border M that sign board 4 is projected to driving analog system screen image by same size, from empty border M, can see the Projection Display object clearly---effective viewpoint of virtual signage board 4 is 5, be called the maximum effectively sighting distance of simulation system apart from VS between viewpoint 5 and the virtual signage board 4, because computer picture, projector resolution, the influence of factors such as picture quality, effectively sighting distance VS is inevitable less than maximum sighting distance SL, that is: VS＜SL, then definition:

SR=VS/SL (SR is smaller or equal to 1) (4)

SR is called the viewing ratio of driving simulator, and SR is more little, illustrates that the explanation quality of virtual image is poor more; When SR illustrates that near 1 the resolution of virtual image is near truth.

For the driving analog system of determining, SR fixes, and for the effective sighting distance VS that guarantees virtual signage board 4 equates with actual maximum sighting distance SL, just the size of sign board 4 must be amplified to 6, if the sign board after amplifying is of a size of T, according to similar triangle theory:

T＝t×SL/VS＝Wt????(5)

W＝SL/VS????????????(6)

Wherein: t is the sign board physical size, the sign board size of T for amplifying, and W is the enlargement ratio of display-object thing in driving analog system screen image.Amplify by driving analog system screen image thus, guaranteed the driver in empty border under the maximum sighting distance SL of reality, can see clearly and actual identical traffic sign information, thereby guarantee the effectively authenticity of sighting distance of object in the three-dimensional virtual image of driving analog system (mainly referring to traffic sign, label).

The present invention finishes by following technical proposal: a kind of scaling method of three-dimensional virtual image of driving analog system, comprise the three-dimensional virtual scene demarcation model of setting up driving analog system with the three-dimensional virtual image modeling software of prior art, it is characterized in that comprising the following steps:

A, demarcate on the model at three-dimensional virtual scene, set a center, this center is the binocular viewpoint, distance between the binocular viewpoint is a, and be that the arc surface that a radius is R is set in the center of circle with this viewpoint, this arc surface is real border, establish object on real border, the height of object is H, and the distance between the object is L in twos, radius R be viewpoint to object apart from S, between binocular viewpoint and object a driving analog system screen image similar to real border is set, promptly empty border is after the object in the real border is projected in empty border, the height of its projection target thing is PK, projection target thing bottom is PH apart from floor level, and the distance between the projection target thing is PL in twos, visual point to the projection target object distance from being d, visual point is h apart from floor level, then:

Obtain in the empty border distance P L between the projection target thing in twos by formula (1):

PL＝d×L/S+a-a×d/S????(1)

Obtain projection target object height degree PK in the empty border by formula (2):

PK＝d×H/S????????(2)

Obtain projection target thing bottom in the empty border apart from floor level PH by formula (3):

PH＝h-h×d/S????????(3)

In the formula: L, S, H, a, h, d are known number;

PL, PK, the pH value of B, projection target thing that above-mentioned steps A is tried to achieve, mark is gone in the empty border respectively, the object in the real border is presented in the empty border by projection, thereby finishes the demarcation of object in empty border;

C, the arc surface of demarcating on the model at the three-dimensional virtual scene of steps A is real side overseas, it is on its of t numeral to be arranged that selectivity is provided with fixed measure, the display-object thing of literal or symbol, the maximum sighting distance of display-object thing is SL in visual point and the real border, the display-object thing projects to and is the Projection Display object in the empty border, can see at visual point that effective sighting distance of Projection Display object is VS in the empty border clearly, for the driving analog system of determining, SR fixes, for the effective sighting distance VS that guarantees Projection Display object in the empty border equates with the maximum sighting distance SL of display-object thing in the real border, display-object thing size will be enlarged into T, according to similar triangle theory:

T＝t×SL/VS＝Wt????(5)

W＝SL/VS????????????(6)

Wherein: t is the physical size of display-object thing, the display-object thing size of T for amplifying, and W is the enlargement ratio of display-object thing in empty border, thereby finishes the demarcation of display-object thing in empty border.

The present invention compared with prior art has following advantage and effect: adopt such scheme, can be faster, more comprehensively, more accurate, more objectively the object space size in the three-dimensional virtual image of driving analog system and effective sighting distance are demarcated, only solved simultaneously in the existing driving analog system static virtual image calibration technique at individuality, lacked the problem of versatility.

Description of drawings

Fig. 1 is the bulk calibration principle synoptic diagram of object in empty border;

Fig. 2 shows the enlargement ratio calibration principle synoptic diagram of thing in empty border for target;

Fig. 3 demarcates the model vertical view for three-dimensional virtual image;

Fig. 4 demarcates the model front elevation for three-dimensional virtual image;

Fig. 5 demarcates the model side view for three-dimensional virtual image;

Embodiment

Embodiment 1

A, the three-dimensional virtual scene of at first setting up driving analog system with the three-dimensional modeling software of prior art are demarcated model, its front view as shown in Figure 4, vertical view as shown in Figure 3, side view is as shown in Figure 5.

B, as shown in Figure 3, setting a center on above-mentioned three-dimensional virtual scene demarcation model is binocular viewpoint 3,31, it is the driver's binocular in the driving simulator 8, binocular viewpoint 3,31 distance is a=0.065m, and be that to set a radius be that the arc surface of R is real border W in the center of circle with this viewpoint, the upright mark post red and white 1 of the six roots of sensation is set on real border W from left to right, 11,12,13,14,15 (being object), the height of mark post is H=3m, mark post 1,11, mark post 12,13, mark post 14, distance between 15 is L=3m, mark post 11,12, mark post 13, distance between 14 is J=3.13m, radius R be viewpoint 3 to each mark post apart from S=10m, between each mark post and visual point 3 a driving analog system screen image similar to real border W being set is empty border M, mark post 1 among the real border W, 11,12,13,14,15 are projection mark post 2 in empty border M, 21,22,23,24,25, distance L among the real border W between mark post, the projector distance of J in empty border M is respectively PL, PJ, the height of projection mark post 2 in empty border M is PK, projection mark post 2 bottoms are PH apart from floor level, visual point 3 is d to the distance of projection mark post, and visual point is h=1.2m apart from floor level;

In C, data substitution formula (1), formula (2), the formula (3) with above-mentioned steps 2,

Between same projection screen mark post apart from theoretical value:

PL＝d/3.3+0.065-0.0065d＝0.065+0.296d????(7)

Between the adjacent mark post of two projection screens apart from theoretical value:

PJ＝d/3.2+0.065-0.007d＝0.065+0.306d????(8)

At the bottom of the projection screen mark post with the ground distance theoretical value:

PH＝1.2-0.12d????????????????(9)

Projection screen mark post standoff height theoretical value:

PK＝0.3d????????????????????(10)

D, with driver's viewpoint to screen be divided into some values apart from d, it is as shown in table 1 then to obtain the pairing parameter theory value of each d value according to above-mentioned theory value computing formula (7)～(10):

Table 1 calibrating parameters theoretical value question blank

??d(m)	Spacing PL (m) between the mark post 1,2,3,4,5,6	Spacing PJ (m) between the mark post 2,3,4,5	??PH??(m)	??PK(m)
					??0.4	??0.18	??0.19	??1.15	??0.12
??0.6	??0.24	??0.25	??1.13	??0.18
					??0.8	??0.30	??0.31	??1.10	??0.24
??1	??0.36	??0.37	??1.08	??0.30
					??1.2	??0.42	??0.43	??1.06	??0.36
??1.4	??0.48	??0.49	??1.03	??0.42
					??1.6	??0.54	??0.55	??1.01	??0.48
??1.8	??0.60	??0.62	??0.98	??0.54
					??2	??0.66	??0.68	??0.96	??0.60
??2.2	??0.72	??0.74	??0.94	??0.66

??d(m)	Spacing PL (m) between the mark post 1,2,3,4,5,6	Spacing PJ (m) between the mark post 2,3,4,5	??PH??(m)	??PK(m)
					??2.4	??0.78	??0.80	??0.91	??0.72
??2.6	??0.83	??0.86	??0.89	??0.78
					??2.8	??0.89	??0.92	??0.86	??0.84
??3	??0.95	??0.98	??0.84	??0.90
					??3.2	??1.01	??1.04	??0.82	??0.96
??3.4	??1.07	??1.11	??0.79	??1.02
					??3.6	??1.13	??1.17	??0.77	??1.08
??3.8	??1.19	??1.23	??0.74	??1.14
					??4	??1.25	??1.29	??0.72	??1.20
??4.2	??1.31	??1.35	??0.70	??1.26
					??4.4	??1.37	??1.41	??0.67	??1.32
??4.6	??1.43	??1.47	??0.65	??1.38
					??4.8	??1.49	??1.53	??0.62	??1.44
??5	??1.55	??1.60	??0.60	??1.50

Wherein, PL: the distance between the mark post 1,11,12,13,14,15 is projected in the projection spacing on the corresponding screen; PJ: the distance between the mark post 11,12,14,14 is projected in the projection spacing on the corresponding screen, PK: the length of projection mark post on the screen; PH: projection mark post bottom is to ground distance on the screen;

E, the theoretical value demarcation of above-mentioned steps 4 is made three-dimensional virtual image on three-dimensional virtual scene demarcation model, project on the driving analog system screen, adjusting driver's viewpoint position makes it to overlap or basic the coincidence with the geometric center of projection screen, the spacing PL of overcasting staff on screen, PJ and mark post length PK and mark post bottom are to ground distance PH, with its value respectively with corresponding theoretical value relatively, if do not satisfy error range, can adjust viewpoint, the projector scaling, parameters such as screen distance, in the error range that is controlled at permission, thereby finish the demarcation of object in empty border, specifically:

(1) measure driver's viewpoint to screen apart from d, according to this distance value, the theoretical value of finding PL, PJ, PK, PH from table 1 is standby.

(2) PL, PJ, PK, the pH value in the virtual image on survey sheet 3 projection screens with the theoretical value contrast, obtained relative error respectively, if do not satisfy error range (≤5%), can follow these steps to adjust:

A) virtual image is carried out whole scaling, satisfy the virtual image imaging effect and be covered with whole screen;

B) the PL value and the PK value of mark post 3,4 in the measurement screen image are searched driver's viewpoint to screen d value according to PL value and PK value in table 1;

C) adjust driver's viewpoint position and make it to satisfy b) middle d value;

D) according to b) the corresponding pH value of middle d value, adjust screen integral image height by the mode of lifting driver viewpoint height;

E) angle of a left side, preceding right screen and middle screen makes it to satisfy pH value before the adjustment;

F, the arc surface outside, real border on the three-dimensional virtual scene demarcation model of steps A, if on its of three fixed size numeral is arranged, the display-object thing of literal or symbol---information label 4, wherein, 41 is digital signage, be positioned at a preceding left side, 42 is the literal label, before being arranged in, 43 is the symbol label, be positioned at the preceding right side, lay respectively at mark post 1,11,12,13,14, between 15, label 4 is adjustablely can know the sighting distance of seeing signage content to driver's viewpoint 3 apart from SL, label 4 projects in the empty border, itself and driver's viewpoint 3 adjustable apart from VS adjusted this sighting distance, makes the driver see the content of projection sign board in the empty border clearly, can obtain the VS value, according to formula (5), formula (6):

T＝t×SL/VS＝Wt????(5)

W＝SL/VS????????????(6)

Can draw the enlargement ratio of label 4 in empty border, thereby finish the demarcation of label 4 in empty border.

Claims (1)

1. the scaling method of a three-dimensional virtual image of driving analog system comprises the three-dimensional virtual scene demarcation model of setting up driving analog system with the three-dimensional virtual image modeling software of prior art, it is characterized in that comprising the following steps:

A, demarcate on the model at three-dimensional virtual scene, set a center, this center is the binocular viewpoint, distance between the binocular viewpoint is a, and be that the arc surface that a radius is R is set in the center of circle with this viewpoint, this arc surface is real border, establish object on real border, the height of object is H, and the distance between the object is L in twos, radius R be viewpoint to object apart from S, between binocular viewpoint and object a driving analog system screen image similar to real border is set, promptly empty border is after the object in the real border is projected in empty border, the height of its projection target thing is PK, projection target thing bottom is PH apart from floor level, and the distance between the projection target thing is PL in twos, visual point to the projection target object distance from being d, visual point is h apart from floor level, then:

Obtain in the empty border distance P L between the projection target thing in twos by formula (1):

PL＝d×L/S+a-a×d/S??????????(1)

Obtain projection target object height degree PK in the empty border by formula (2):

PK＝d×H/S???????????????????(2)

Obtain projection target thing bottom in the empty border apart from floor level PH by formula (3):

PH＝h-h×d/S?????????????????(3)

In the formula: L, S, H, a, h, d are known number;

PL, PK, the pH value of B, projection target thing that above-mentioned steps A is tried to achieve, mark is gone in the empty border respectively, the object in the real border is presented in the empty border by projection, thereby finishes the demarcation of object in empty border;

C, the arc surface of demarcating on the model at the three-dimensional virtual scene of steps A is real side overseas, it is on its of t numeral to be arranged that selectivity is provided with fixed measure, the display-object thing of literal or symbol, the maximum sighting distance of display-object thing is SL in visual point and the real border, the display-object thing projects to and is the Projection Display object in the empty border, can see at visual point that effective sighting distance of Projection Display object is VS in the empty border clearly, for the driving analog system of determining, SR fixes, for the effective sighting distance VS that guarantees Projection Display object in the empty border equates with the maximum sighting distance SL of display-object thing in the real border, display-object thing size will be enlarged into T, according to similar triangle theory:

T＝t×SL/VS＝Wt?????????????????????(5)

W＝SL/VS????????????????????????????(6)

Wherein: t is the physical size of display-object thing, the display-object thing size of T for amplifying, and W is the enlargement ratio of display-object thing in empty border, thereby finishes the demarcation of display-object thing in empty border.

Images