WO2010069288A1

WO2010069288A1 - Method for visualization of a three-dimensional structure on a display area

Info

Publication number: WO2010069288A1
Application number: PCT/DE2009/001671
Authority: WO
Inventors: Matthias Wegner
Original assignee: Eads Deutschland Gmbh
Priority date: 2008-12-19
Filing date: 2009-11-25
Publication date: 2010-06-24
Also published as: DE102008064066A1

Abstract

The invention relates to a method for visualization of a three-dimensional structure on a display area, wherein the surface of the three-dimensional structure is broken down into individual area elements (FE) which are displayed on the display area in central perspective. The area elements (FE) are coloured with at least three different colours, wherein the distribution of the individual colours is between the area elements (FE) of the structure surface is chosen randomly.

Description

Method for visualizing a three-dimensional structure on a display surface

The invention relates to a method for visualizing a three-dimensional structure on a flat or curved display surface according to the preamble of patent claim 1.

With the availability of highly accurate digital elevation databases (DDTB = Detailed Digital Terrain Database) and high-resolution RADAR and ladar systems (eg HELLAS ^® EADS Germany GmbH), are increasingly called "Synthetic Vision" ads in the cockpits of modern aircraft for Commitment. A "synthetic vision" display is primarily the synthetic, ie artificial, representation of the surrounding terrain structure with all its significant elevations - the so-called soil profile (usually supplemented by further symbolism). Such a synthetic "outside view" is intended to additionally support the pilot during the flight, in particular in invisible weather, and thus ultimately increase flight safety (see, for example, DE 102004051625A1).

In US 6,121,972 a method is described for displaying a 3-dimensional terrain structure from an elevation database with a central projection on a display area. The individual surface elements of the terrain structure are assigned specific colors as a function of the respective elevation height.

The main problem with a "synthetic vision" terrain presentation is the fact that a three-dimensional geometric structure (the terrain) must be displayed on a 2-dimensional display plane. This is done via a central projection, ie a geometric image that corresponds to human vision. However, there is a risk that - if the solution is unfavorable - the geometric depth of the terrain shown is insufficiently reproduced or even lost.

To visualize the depth of three-dimensional terrain structures on a display equidistant escape and distance lines have been used for some time (Fig. 1). These escape and distance lines follow the terrain structure and break up the terrain into mostly equal-sized squares or rectangles, which are intended to convey the impression of depth in the perspective shortening (for example, DE 102007014015 A1, DE 4314811 A1).

The concept of escape and distance lines is established and is almost universally used in the visualization of 3-dimensional terrain structures. It generally gives quite a good depth impression of the terrain, but has its weaknesses - especially when only a grid of simple lines for deep visualization is used.

Often the terrain elevation in "Synthetic Vision Displays" e.g. color coded depending on the current altitude of the aircraft (e.g., red for very large terrain elevations). This type of presentation gives the pilot a good idea of the altitude of the surrounding terrain relative to its altitude, but the depth impression is usually less satisfactory to poor, despite the use of escape and distance lines.

Attempts to improve the depth of the terrain in the display by merely increasing the number of escape and distance lines and / or introducing additional diagonal lines do not show the desired gain in depth information to the viewer. Although a fine-meshed network of escape and distance lines (and possibly diagonal lines) better maps the geometry of the terrain, it does not always contribute to improving the impression of depth for the viewer. A common way to work around this problem is the Use of points or dashed lines for the escape and distance lines. In most cases, such a representation provides a better depth impression, despite fine-meshed escape and distance lines.

However, the possibilities of achieving a good depth impression of the depicted terrain solely on the basis of a grid of escape and distance lines are ultimately limited. To get around this limitation, you move away from simple lines to regular surface elements (so-called patches). For this purpose, FIGS. 2 and 3 show examples of a planar environment (FIG. 2) as well as of a three-dimensional terrain profile (FIG. 3). The terrain is thereby decomposed into equally sized, alternately colored surface elements, which in central perspective, i. from an observer's point of view, in a two-dimensional display plane. The result is a regular, checkerboard-like texturing of the terrain. The surface elements of the texturing are limited by the original escape and distance lines - these lines may or may not be retained in addition to the area elements (see FIG. 2).

The disadvantages of this checkerboard-type texturing for solving the problem of "depth problems" are twofold:

a) There are generally so-called visual artifacts in the far area of the checkerboard pattern. The reason for this lies in the fact that - due to the perspective distortion - an ever smaller checkerboard pattern is imaged over an equidistant display pixel grid. This effect is well known in signal theory and digital image processing and is mathematically described by Shannon's sampling theorem.

b) The decisive disadvantage of a checkerboard pattern for the visual mediation of the depth of an imaged terrain lies in the fact that dynamic changes in the observer's position (eg when flying over the terrain) are only inadequately perceived or not at all if the viewing direction is nearly identical-if the display display is not permanently and highly concentrated by the pilot. In other words, it always looks pretty much the same - especially when you have moved left / right or forward / backward at the discrete grid spacing of the chess bed pattern, without consciously and intentionally following that movement.

Due to the complex requirements of the pilots during the flight, it is not possible or at least very unlikely that the pilot can observe a "Synthetic Vision" display permanently and highly concentrated. In the case of the above-mentioned known solutions to the "depth problem", visual orientation problems can therefore arise for the reasons given, which in turn can adversely affect the decision-making ability of the pilot and thus his missionary freedom of action in the possibly critical seconds. The present invention has for its object to solve the problems described in connection with the visualization of depth in a display representation.

This object is achieved with the subject of claim 1. Advantageous embodiments are the subject of dependent claims.

The advantage of the method according to the invention lies in the safe and easy and rapid detection of the depth by the observer - especially when applied to three-dimensional terrain profiles (eg from digital elevation databases). The present invention allows a much greater "naturalness" in the display as alternative proposals. For this reason, it is able to improve corresponding displays in so-called "Synthetic Vision" displays, as they are increasingly used in modern aircraft. Thus, the pilot can be better supported or relieved and this helps the Flight safety - especially for critical missions or flight profiles - increase.

The present invention allows the pilot to always clearly and quickly recognize and visually detect the depth of the terrain display on a display, even if the pilot can not permanently and highly concentrated monitor the display of the terrain, because in the meantime other things in the cockpit his attention claim. In this respect, the present invention not only provides an improved depth impression than the known and established concepts, but is also more flexible with regard to the operational use in the cockpit.

The invention will be explained in more detail with reference to a concrete embodiment with reference to FIG .. Show it:

Fig. 1 Escape and distance lines to represent the depth of a plane in a two-dimensional display representation (prior art);

2 shows a checkerboard pattern for visualizing the depth of a plane in a two-dimensional display representation (prior art);

3 shows another example of the checkerboard representation of a terrain profile (prior art);

Fig. 4 shows the representation of a terrain profile according to the invention.

To convey a good depth impression of the depicted three-dimensional terrain on a display, a planar representation of the three-dimensional structure structured by surface elements appears more suitable than a corresponding representation consisting of escape and distance lines. In a planar representation, the escape and distance lines are not lost because they limit the individual surface elements. The surface elements are usually the Better support the depth impression of the terrain shown - especially if the escape and distance lines are very close together or if additional information codes (such as the terrain altitude) are included in the display.

According to the invention, the area elements with n (n ≥ 3) become different

Colors (hereafter referred to as "representation colors", which may also be gray tones) are colored, whereby the distribution of the individual colors on the surface elements is random.

The surface elements can be additionally limited by escape and distance lines.

The random distribution of the representation colors on the surface elements can be accomplished in particular by means of a recursive, arithmetic method for the calculation of deterministic random numbers (so-called pseudorandom numbers) - in that respect the distribution of the representation colors on the surface elements is not really random. The advantage of such a calculation method for pseudo random numbers is that it can easily be implemented on any computer. This enables a universal applicability of the present invention in a wide variety of systems, in which three-dimensional structures are to be spatially visualized.

A recursive arithmetic method for calculating pseudo-random numbers uses already known random numbers as starting values for the determination of a new pseudorandom number for each calculation cycle. With this mechanism, a sequence of pseudorandom numbers is generated in a recursive process. This sequence is deterministic, since the same sequence of pseudorandom numbers is always generated given the same starting values. A calculation method for generating pseudorandom numbers that can be used for the invention generally consists of a quadruple:

X (Z _m , S, f, g)

In this case, the set Z _1n denotes the so-called state space with the thickness m Z ₁₁₁ : = {0,1,..., W-1} with weN ⁺ and w »l,

the set S the starting values of the method

S: = {y _o , y _x , ..., y _k _,} ciZ _m ^k with keN ⁺ and *> l,

the mapping / is the arithmetic transfer function after Z _1n

where y _n denotes the «th state value and

Figure g is the normalization function to the right open unit interval

g: Z _m → [0, ϊ), y ^ giyd ^ - with i≤n. m

One possible concrete expression of this calculation rule may be e.g. be configured as follows:

With & = 1, the set of starting values is reduced to exactly one element: s = {y _o ) cz _m

For the arithmetic transfer function, one then selects for n ≥ 2 e.g.

f {y "- _ι ) = (ay _n ^ + b) moάm = y _n with a, be {l, ..., m - ϊ}

(Proposal by D.H. Lehmer, 1949).

However, the above-described concept for the recursive calculation of pseudorandom numbers is general and can thus be easily adapted to given requirements for the application of the invention.

To visualize the depth with the help of the pseudo-random, n (n> 3) different representation colors are defined first. The goal is to assign the different representation colors by pseudo-randomness to the surface elements into which the texturing of the three-dimensional structure is decomposed. An advantageous optical effect is achieved if the chosen different colors for the surface elements are not very different from each other (for example, one could choose three shades of gray - light, medium, dark).

For this purpose, in a first step, the selected display colors of the area elements are assigned via a suitable mapping / the pseudo-random numbers from the open-right interval [0, 1).

(1) /: [0, 1) → {1,2,3, ..., «} C N

Assuming the uniform distribution of the pseudorandom numbers z, over the interval [0, 1), for example, such an assignment to the nth representation color c "could be effected by the following mapping: Zi i → / (Z ₁ ) ^ Iz ₁ K ₊ Ii = C _n with n (n ≥ 3).

The operator [J is the ^Gauss' sche clip.

By means of such an or equivalent calculation rule, each pseudorandom number Z ₁ . exactly one representation color c _{n assigned} to the surface elements.

The escape and distance lines as real or virtual boundaries of the differently colored surface elements form an equidistant grid over the terrain to be displayed. Consequently, each individual area element can be clearly described by its location within this grid network arrangement. Incidentally, this uniqueness in the spatial assignment of the surface elements is maintained, regardless of whether the equidistant grid of the escape and distance lines is present relative to the location of the aircraft or absolutely, geo-referenced. By the way, this fact is also true analogously for non-square or non-rectangular grid arrangements, such as e.g. an equidistant grid composed of a hexagonal honeycomb structure or e.g. only consists of triangles. The lateral boundaries of this type of surface elements then form natural no escape and distance lines more. However, this circumstance is without relevance to the present invention and all statements made here can be applied equally equally to non-square or non-rectangular, equidistant grid arrangements of the surface elements.

If φ, λ e R denote the spatial coordinates of a surface element. Furthermore, let g be a suitable function that associates each location coordinate tuple with a "random" value from the interval [0, 1), thus:

(2) g: R ² → [0, 1) A possible concrete expression of this function could be with a pseudo-random number z. e [0, 1) eg be designed as follows:

With the help of such an equivalent or equivalent calculation rule, a "random" value from the interval [0, 1) is assigned to each location coordinate tuple.

Through the above introduced Figures (1) and (2), a composite map can now be defined which assigns to each location coordinate tuple a "randomly" selected color from the set of n (n ≥ 3) given representation colors for the area elements:

/ og: R ² → {l, 2,3, ..., «} c N, (φ, λ) i → f (g (φ, λ)) with φ, λ e R

The concrete implementations of Figures (1) and (2) should be interpreted as advantageously as possible for implementation in an operational system and may differ from the examples given here. However, the underlying concept always remains applicable, namely the "random" assignment of a color from the set of predefined representation colors to a specific area element of the texturing grid.

The "random" assignment of the representation colors to the individual surface elements of the texturing removes the strict and unnatural homogeneity in the visualization of terrain structures, as can be observed in the prior art (FIGS. 2, 3). Due to the "random" color assignment results in a very peculiar optical effect, which makes the terrain surface appear much more "natural" for the viewer. Due to the "random" distribution of the n (n ≥ 3) representation colors on the individual surface elements This leads to "random" optical clustering and patterning, which - in contrast to the chessboard representation - each have different shapes. This makes it possible for the observer to orient himself much more easily on the terrain, because he can orient himself in each case on certain, visually striking color clusters or inhomogeneous patterns as reference points. Also, and in particular, the depth of the terrain shown by the optically inhomogeneous structuring due to the "random" distribution of display colors clearly beneficial. In addition, the "random" color arrangement according to the invention for the surface elements brings about a much more natural overall visual impression of the area shown.

FIG. 4 illustrates the concept according to the invention for visualizing the depth of terrain structures in comparison to the checkerboard representation of the same terrain according to FIG. 3. A three-dimensional terrain profile is shown with one

Texturing, which is constructed from equal, square surface elements, wherein the surface elements are shown in the perspective of an observer. The display colors are randomly distributed to the individual surface elements of the terrain profile.

It should be pointed out that the surface structure consisting of square surface elements according to FIGS. 3 and 4 is overlaid in each case by a shadow (in the area of the steep slope) in order to further improve the depth representation.

Instead of the illustrated square surface elements, e.g. also rectangular, triangular or hexagonal (honeycomb) shapes can be used.

In the display area on which the three-dimensional is displayed, it may be a flat or curved surface in any arbitrary way.

Claims

claims

1. A method for visualizing a three-dimensional structure on a display surface, wherein the surface of the three-dimensional structure is decomposed into individual surface elements (FE), which are displayed on the display surface in central perspective, wherein the surface elements (FE) colored with at least three different colors be, characterized in that the distribution of the individual colors on the surface elements (FE) of the structure surface takes place randomly by means of pseudo-random numbers.

2. The method according to claim 1, characterized in that the pseudorandom numbers are equally distributed.

3. The method according to any one of the preceding claims, characterized in that the surface elements (FE) are square, rectangular, triangular or of the shape of a regular hexagon.

4. The method according to any one of the preceding claims, characterized marked, that the surface elements (FE) of distance lines and alignment lines are limited.