AR (augmented reality) graphic method for HUD (head Up display) orthostatic live-action display
Technical Field
The invention relates to an augmented reality graph conversion technology, in particular to an AR graph algorithm which obtains an AR graph conversion function relationship through a multivariate multiple linear relationship model function and can realize orthostatic real-scene display of shot images at different angles.
Background
The Advanced Driving assistance System (Advanced Driving assistance System) senses the surrounding environment at any time in the Driving process of the automobile by using various sensors arranged on the automobile, collects data, identifies, detects and tracks static and dynamic objects, and performs systematic operation and analysis by combining with map data of a navigator, so that a driver can be made to perceive possible dangers in advance, and the comfort and the safety of automobile Driving are effectively improved.
Augmented Reality (AR) is a technology for calculating the position and angle of a camera image in real time and adding a corresponding image, and the aim of the technology is to sleeve a virtual world on a screen in the real world and perform interaction.
At present, an ADAS advanced driving assistance system cannot be fully realized by a vehicle-mounted head-up display optical system, and the main defects are as follows:
1. the existing traditional vehicle-mounted head-up display system directly projects a display picture which is irrelevant to a road picture (an actual world) in front, that is, the traditional vehicle-mounted head-up display image is not fused with the external actual world. The perspective view angle of the real world object observed by the human eyes is characterized by being large and small, the image in the traditional head-up display is not subjected to corresponding coordinate conversion to accord with the perspective effect of the human eyes, so that the image projected by the head-up display is inconsistent with the coordinate of the real image, the human eyes watch the image in the head-up display, the two images cannot be well fused and attached, and the function of enhancing the display is not provided, so that if the content of the head-up display is fused with the real world, the image characteristic required to be projected accords with the perspective effect of the real world object on the human eyes.
2. Static and dynamic objects with a certain visual angle, such as pedestrians and vehicles on the side, or vehicles behind, cannot be identified, detected and tracked, and in this case, on one hand, the existing head-up display system cannot display the image real scene in the right position, so that subjective misjudgment is easily generated, and the other party lacks necessary warning behaviors, thereby having a certain potential safety hazard.
Disclosure of Invention
The invention aims to solve the technical problems and provides an AR (augmented reality) graphic method for HUD (head Up display) normal live-action display.
In order to solve the above prior art problems, the technical scheme of the invention is as follows:
an AR graphic method for HUD orthomorphic real scene display, the AR graphic algorithm includes constructing a reference icon library stored by using an orthomorphic real scene gridding format, the AR graphic algorithm adopts the following steps:
firstly, respectively establishing a coordinate system of an initial graph and a positive real scene graph;
secondly, constructing a multivariate multiple linear relation model function between coordinate systems;
thirdly, selecting coordinate points in the orthostatic real scene gridding format of the icons in the multiple reference icon libraries and corresponding coordinate points in the initial coordinate graph to substitute the multiple linear relation model function, solving to obtain parameter values of the multiple linear relation model function, and further obtaining a numerical conversion function;
and fourthly, converting all the corresponding coordinate points in the initial coordinate graph through the numerical function relationship, thereby obtaining the normal live-action image corresponding to the initial coordinate graph.
The normal real scene gridding format of the icons in the reference icon library is a two-dimensional or three-dimensional mode;
when the normal real scene gridding format of the icons in the reference icon library is in a two-dimensional mode, the AR graphic algorithm comprises the following steps:
setting a coordinate system of the initial coordinate graph as a coordinate I, and representing the middle point of the image as (X)1,Y1)、(X2,Y2)、(X3,Y3)……(Xn,Yn),
The coordinate system of the icons in the reference icon library is a coordinate II, and the middle point of the image is represented as (Xa)1,Ya1)、(Xa2,Ya2)、(Xa3,Ya3)……(Xan,Yan);
Setting the coordinate point in the first coordinate and the X-axis coordinate and the Y-axis coordinate of the coordinate point in the second coordinate to be a multiple linear relation, wherein the multiple linear relation is at least 3 times, and when the multiple linear relation is selected to be a cubic linear relation, obtaining a functional relation model of the coordinates as follows:
Xn = AXan 3 + BXan 2 + CXan + DYan 3 + EYan 2 + FYan + G,
Yn = HXan 3 + IXan 2 + JXan + KYan 3 + LYan 2 + MYan + M;
selecting coordinate points in the normal real-scene gridding format of the icons in a plurality of reference icon libraries and corresponding coordinate points in the initial coordinate graph to substitute the coordinate points into the fourteen-element cubic equation, and selecting 14 characteristic points to obtain the values of the parameters A-M;
and (IV) substituting the parameter values into the functional relation model to obtain a numerical functional relation, and converting all corresponding coordinate points in the initial coordinate graph through the numerical functional relation to obtain an orthonormal coordinate image corresponding to the initial coordinate graph.
The reference icon library is used for inputting different road sign icons or traffic sign icons or pedestrian icons or vehicle icons, and then storing the input road sign icons or traffic sign icons or pedestrian icons or vehicle icons in a normal real scene gridding format to obtain the reference icon library containing a large number of reference icons.
The invention discloses an AR (augmented reality) graph method for HUD (head Up display) normal live-action scene display, which has the following beneficial effects:
1. the AR graphic algorithm can obtain a numerical graphic conversion function relationship by setting a specific reference object coordinate conversion system and a multivariate multiple linear relationship model function, thereby realizing the normal live-action display of shot images at different angles;
2. the AR graph algorithm can realize two-dimensional or three-dimensional graph conversion, the precision of the AR graph algorithm can be adjusted according to needs, meanwhile, the AR graph algorithm is reversible, and various special graph conversion effects can be realized.
Description of the drawings:
FIG. 1 is a schematic diagram of an initial image of an AR graphics method for HUD orthostatic live view display according to the present invention;
FIG. 2 is a schematic diagram of gridding of orthographic scene icons in a reference graphic library of the AR graphic method for HUD orthographic scene display according to the present invention;
fig. 3 is a comparison chart of the initial image conversion and the conversion of the normal live-action icon in the AR graphics method for HUD normal live-action display according to the present invention.
The specific implementation mode is as follows:
the invention is further illustrated by the following examples:
example (b):
an AR graphic method for HUD orthomorphic real scene display, the AR graphic algorithm includes constructing a reference icon library stored by using an orthomorphic real scene gridding format, the AR graphic algorithm adopts the following steps:
firstly, respectively establishing a coordinate system of an initial graph and a positive real scene graph;
secondly, constructing a multivariate multiple linear relation model function between coordinate systems;
thirdly, selecting coordinate points in the orthostatic real scene gridding format of the icons in the multiple reference icon libraries and corresponding coordinate points in the initial coordinate graph to substitute the multiple linear relation model function, solving to obtain parameter values of the multiple linear relation model function, and further obtaining a numerical conversion function;
and fourthly, converting all the corresponding coordinate points in the initial coordinate graph through the numerical function relationship, thereby obtaining the normal live-action image corresponding to the initial coordinate graph.
The normal real scene gridding format of the icons in the reference icon library is a two-dimensional or three-dimensional mode;
when the normal real scene gridding format of the icons in the reference icon library is in a two-dimensional mode, the AR graphic algorithm comprises the following steps:
setting a coordinate system of the initial coordinate graph as a coordinate I, and representing the middle point of the image as (X)1,Y1)、(X2,Y2)、(X3,Y3)……(Xn,Yn),
The coordinate system of the icons in the reference icon library is a coordinate II, and the middle point of the image is represented as (Xa)1,Ya1)、(Xa2,Ya2)、(Xa3,Ya3)……(Xan,Yan);
Setting the X-axis coordinate and the Y-axis coordinate of the coordinate point in the first coordinate and the coordinate point in the second coordinate to be a multiple linear relation, wherein the multiple linear relation is at least 3 times, and when the multiple linear relation is selected to be a cubic linear relation, obtaining a functional relation model of the coordinates as follows:
Xn = AXan 3 + BXan 2 + CXan + DYan 3 + EYan 2 + FYan + G,
Yn = HXan 3 + IXan 2 + JXan + KYan 3 + LYan 2 + MYan + M;
selecting coordinate points in the normal real scene gridding format of the icons in a plurality of reference icon libraries and corresponding coordinate points in the initial coordinate graph to substitute the coordinate points into the fourteen-element cubic equation, and selecting 14 characteristic points to obtain the values of the parameters A-M;
and (IV) substituting the parameter values into the functional relation model to obtain a numerical functional relation, and converting all corresponding coordinate points in the initial coordinate graph through the numerical functional relation to obtain an orthonormal coordinate image corresponding to the initial coordinate graph.
The reference icon library is used for inputting different road sign icons or traffic sign icons or pedestrian icons or vehicle icons, and then storing the input road sign icons or traffic sign icons or pedestrian icons or vehicle icons in a normal real scene gridding format to obtain the reference icon library containing a large number of reference icons.
The present invention has been described in detail, and it should be understood that the detailed description and specific examples, while indicating the preferred embodiment of the invention, are intended for purposes of illustration only and are not intended to limit the scope of the invention.