DE4114304C1 - Area surveillance camera - has detector surfaces separated from imaging areas and in form of parallel strips of optoelectronic devices - Google Patents

Abstract

A camera (K) is used for the surveillance of a defined area or region of land (G). The camera has a lens (O) that focuses onto a number of detector surfaces (1, 2, 3). The detectors can be in the form of parallel strips, with each strip consisting of a matrix of opto-electrical elements. The centre element (3) can be in the form of a strip that provides overlapping of the other detector strips to ensure that there are no gaps in the field of vision. ADVANTAGE - Ensures that complete area is covered.

Description

The invention relates to a camera for recording remote sensing data according to the preamble of claim 1 and such according to the Preamble of claim 2.

Such a camera of the first-mentioned version is known from EP-A2-03 26 128. There is shown in Fig. 3 shows an arrangement of Chen FLAE and line detectors shown a camera for remote sensing of the earth in the image plane of a lens. The detectors consist of a single, opto-electronic detector element. Three line detectors arranged at a distance from one another determine the width of the usable image field by their length. They serve to take up the scanned terrain in the form of three picture line groups with different perspectives. Suitable mathematical evaluation methods can be used to obtain clear information about the three-dimensional structure of the terrain surface. Five area detectors arranged between the line detectors and preferably at the edges of the usable image field are exposed simultaneously at certain time intervals. Since the camera carrier may only move between two successive exposures so far that the area detectors still partially cover common areas. By selecting five pixels in these common sections of the terrain, it is possible with the help of known photogrammetric calculation methods to orientate the successive individual images to one another, ie to determine the position and angular position of the camera carrier at each point in time relative to the previous point in time.

The actual, complete image information of the scanned area can, however, only be obtained with the help of the line detectors, which are between two large individual shots, one large each, when used of CCD detector elements depending on the respective integration time deliver a dependent number of closely consecutive image lines. The too however, the camera carrier's next unknown movement of its own brings it with it be aware that the mutual orientation of the line images is not exactly  is known. Each individual image line must therefore precede that the orien through elaborate correlation and interpolation procedures be animals. This so-called following picture connection is for a pure one Three-line camera known for example from DE-C2-29 40 871. Here the photogrammetric evaluation without additional data is not so stable as in the case of area shots. This may result in less Accuracies and higher computing effort.

The invention is therefore based on the object of a camera at the beginning to provide the type mentioned, which uses as little computing effort as possible ne three-dimensional object reconstruction as accurate as possible.

This object is achieved according to the invention by the in the characterizing part of claim 1 mentioned features solved.

In claim 2, another solution is given that refers to the case that the greatest possible width of the usable Image field is sought. As you know, it is no longer possible to get by with closed line or area detectors, as these only with a limited number of individual detector elements are economically producible. It has therefore already been proposed To use multiple optics and in the image planes of the individual object tives, each covering the same area, individual times Len or surface detectors offset like a mosaic to arrange that the entire image field is recorded without gaps (see at for example EP-A2-03 26 128, column 2 below and column 3 above). Claim 2 relates to the expansion of the inventive concept to Ka meras with such multiple optics and nested like a mosaic Single detectors.

The subject matter of claim 1 differs from the detector arrangement known from FIG. 3 of EP-A2-03 26 128 in the image plane of a single objective in that instead of the three line detectors required there two surface detectors are now to be used which cover two strip-shaped image plane regions . These stripe-shaped image plane regions are of substantially the same length and are arranged approximately transversely to their longitudinal extension at a distance from one another, ie one behind the other in the direction of flight. The two area detectors consist of individual, mutually adjacent, parallel detector lines, which provide the actual image information. So here the principle of line-shaped individual images is departed from, which, as explained above, has the disadvantage that the mutual orientation of the individual image lines is initially unknown. In contrast to this, the orientation of all image lines, which are supplied by the detector lines of the two surface detectors mentioned, is the same in each case and therefore only needs to be determined once. The more detector lines the two area detectors therefore have, the less effort is required to calculate the mutual line orientation. Instead of a continuous sequence of line recordings, who - depending on the number of detector lines belonging to the two surface detectors - now generate quasi-flat recordings in relatively larger time intervals, and it is only necessary to orient them in relation to one another.

The two surface detectors, which are spaced apart from each other fen-shaped image plane areas in the image plane of a single lens cover, form the terrain from two ver in the course of overflying different perspectives, so that a stereo display is possible. The following picture connection and mutual orientation serve at least Mostly, another area detector in the image plane of the lens at a distance from the two aforementioned surface detectors or strips shaped image plane areas is arranged. This one more area Chendetektor can, as indicated in dependent claim 5, almost over extend the entire width of the usable image field, which by the Length of the first two surface detectors is determined. The Ver The use of a single such further area detector is egg ne possibility, but is not the cheapest embodiment, since the entire width is not required for the evaluation process. It is therefore proposed in sub-claim 6 as a cheaper option two further, spaced-apart area detectors  use, for example against each other according to subclaim 7 overlying edges of the usable image field can be arranged and overall require a much smaller detector area than in In the case of a single, spread over the entire width of the usable image field extending further area detector.

The further surface detector or detectors are expediently between between the two stripe-shaped image plane areas with which these covering two surface detectors arranged so that the latter to the with respect to the flight direction front and rear edges of the image plane come to rest, the length (in the direction of flight) of the usable image field is determined by their distance. Depending on the nature of the overflowing terrain, it can be unfavorable if the two fight fen-shaped image plane areas and the associated, essentially area detectors extending transversely to the flight direction, which the egg Provide image information that is too close to each other because the Convergence angle then becomes too small and thus becomes too imprecise speed when determining the height of the terrain points. Therefore it is useful to have the two surface detectors on the front and back lower picture plane edges and the width required for the subsequent picture connection to arrange their surface detectors in between, if possible on the lateral image plane edges.

According to the solution for cameras given in claim 2 with multiple optics is provided, the surface detectors in the resulting from the use of multiple optics different Arrange image planes so that when these image planes are superimposed, two strip-shaped image plane areas between which a distance is read hen is seamless and in an alternating manner by area detectors individual image layers are covered. In the individual picture planes there are gaps between the area detectors located there. Wei There is also another in at least two of the individual image levels Area detector at a distance from the two strip-shaped image plane levels richly arranged. Due to the length of these essentially the same length Image plane areas are also the width of the usable image field Right.

The following are embodiments of the invention based on the figures explained in more detail. They show in a schematic way:

Fig. 1, two each, in strip-shaped image plane regions in the image plane of a lens arranged surface detectors and a further intermediate area detector

Fig. 2 shows a modified version of Fig. 1 with two other chendetektoren FLAE,

Fig one in the case of two separate image planes at their superposition resulting uninterrupted succession of area detectors in two strip-shaped image plane regions chendetektoren. 3 with two other FLAE,

Fig. 4, the two separate image planes of Fig. 3 in separate Dar position with the area detectors within them,

Fig. 5 a, in the case of three separate image planes formed in the superposition complete array of individual surface detectors with two additional area detectors,

Fig. 6, the three separate image planes shown in FIG. 5 in a separated position with the Dar located in this area detectors,

Fig. 7 is a camera with a lens and an image plane via a terrain,

Fig. 8 is a camera with a lens, a beam splitter and two separate image planes.

FIG. 7 schematically shows a camera K over a terrain, which uses a lens O to image a terrain detail G in its image plane E. The camera K is carried by a camera carrier (not shown), for example an airplane or a satellite, in order to take pictures of the terrain. In the case of a horizontal straight flight, the optical axis OA of the camera K is directed vertically downwards. The camera's own coordinate system x, y, z may be oriented so that the x-axis as the roll axis points in the direction of flight, the z-axis as the yaw axis coincides with the optical axis OA and the y-axis as the pitch axis together with the x-axis lies in the image plane E. In Fig. 7 three beam paths are indicated, which lead to surface detectors 1 , 3 and 2 , the structure and arrangement of which is shown in more detail in the image plane E in Fig. 1.

Fig. 1 shows the image plane E seen from the lens O in the direction of the optical axis OA's. At the front and rear edges of the image plane E with respect to the direction of flight, which may coincide with the direction of the positive x-axis, two strip-shaped image plane regions E ₁ and E _{2 are} defined, which are completely covered with surface detectors 1 and 2 . These area detectors 1 and 2 are made up of op-electronic detector elements d _nm ( 1 ≦ n ≦ N, 1 ≦ m ≦ M), which are arranged in parallel, directly adjacent detector lines D _n , as in a corner area of the area detector 1 shown in more detail in Fig. 1. The two surface detectors 1 and 2 preferably run parallel to one another, a distance between them being necessary. This distance is given above all by the desired angle of convergence, since the two area detectors 1 and 2 see the terrain with different perspectives, namely looking back in the case of the area detector 1 and looking forward in the case of the area detector 2 . The angle of convergence should not be too small, since otherwise inaccuracies in the determination of the height coordinates of the terrain points may occur. By the mentioned arrangement of the two surface detectors 1 and 2, the usable image field F is given, the width f _x equal to the length of two successively arranged FLAE chendetektoren 1 and 2, and its length (in the direction of flight) f _y equal to the sum of the distance of the two surface detectors 1 and 2 and their widths.

Within the stripe-shaped image plane areas E ₁ and E ₂ , corresponding terrain strips are simultaneously imaged on the surface detectors 1 and 2 covering them. The images of these areas are registered on the basis of the detector structure in the form of a multiplicity of parallel line images recorded at the same time, which in turn consist of a multiplicity of adjacent image points assigned to the individual optoelectronic detector elements. It is thus sufficient to carry out areal recordings at certain time intervals that are dimensioned such that the terrain, for example, recorded with the area detector 1 at two successive recording times, is completely contiguous or slightly overlap (in the direction of flight).

The number N of detector lines D _n within an area detector 1 or 2 must in any case be N ≧ 2 , because this leaves the principle prevailing in the three-line camera, according to which in each recording time of each of the three viewing directions with different perspectives only ever one image line assigned. In the present case, the transition to areal recording per recording time is in principle already made for N = 2 , since then two parallel, adjacent image lines are already available at the same recording time for each perspective viewing direction. In practice, however, an area detector 1 or 2 will contain far more than two detector lines D _n , since the advantage of the new principle can then be better exploited.

In the embodiment according to FIG. 1, a further area detector 3 is arranged between the two area detectors 1 and 2 , which extends over the entire width of the image field F. This also consists of several parallel detector lines and serves to enable the subsequent image connection.

Referring to FIG. 2 are provided between the two surface detectors 1 and 2, two additional area detectors 4 and 5 which respectively occupy only a fraction of the width of the image field F and are useful on its lateral edges. This embodiment is preferred in practice. However, it is not absolutely necessary for the two further surface detectors 4 and 5 to be located exactly at the edges of the image field F or to be at the same height with respect to the x coordinate. It is even possible to arrange one or both of these further surface detectors 4 and 5 not between the two surface detectors 1 and 2 , although it is a prerequisite that at least one of the latter two surface detectors mentioned is not arranged at the front or rear edge of the image field F. .

Five or six pixels are to be selected for the subsequent image connection using known correlation and photo grammetric methods. This can e.g. B. happen as indicated in Fig. 2 by crosses.

So far it has been assumed that the surface detectors 1 and 2 in ge closed form, ie can be manufactured in one piece. However, this only applies to certain limits, ie currently up to approximately 4096 detector elements per detector line. If a wider field of view is to be covered, this could be achieved by lining up separate surface detectors, but there would always be gaps where the latter meet. It has therefore already been proposed to use multiple optics and to adjust so that in their image planes with respect to the optical axes or the respective image coordinates, systems in an identical position, ie congruent, the same terrain cut comes to the picture. Then separate surface detectors can be arranged at a distance from one another in the different image planes in such a way that the desired image plane region is covered by them in an alternating manner without gaps. The multiple optics can be realized either by parallel individual lenses or by switching on beam splitter systems, for example semi-transparent mirrors.

A possible version of the use of double optics is shown schematically in FIG. 8. Behind the single lens O, a semi-transparent mirror S is arranged at 45 ° with respect to the optical axis OA, which results in two image planes rotated by 90 ° relative to one another. In both image planes two stripe-shaped image plane regions E ₃ and E _{4 are} defined, which visually overlap each other congruently with regard to the object or terrain strips depicted on them. As in the total of four stripe-shaped image plane areas (two E ₃ and two E ₄ ) separate surface detectors can be arranged in order to ensure complete coverage when superimposed, FIGS . 3 and 4 can be seen.

Accordingly, square image detectors 11 , 12 , 13 , etc., for example, are arranged in the one image plane designated EA in FIG. 8, as shown in FIG. 4. The same applies to the surface detectors 21 , 22 , 23 arranged in the second image plane denoted by EB. In each of the two image planes, E ₃ and E ₄ are located in the middle between the two respective strip-shaped image plane regions E ₃ and E ₄ on one of the two lateral edges of the usable image field F a wider area detector 10 or 20 is arranged. The square surface detectors in the exemplary embodiment each have the width b, their mutual distance Δa also being equal to b due to the square dimensioning. Basically, the distances between the separate area detectors in an image plane in the arrangement of Fig. 3 inversely equal to the width of the separate area detectors in the image plane and walls ren.

Fig. 5 shows a possible arrangement in se parater illustration shows separate area detectors when using a triple optics in optical superposition, Fig. 6, the individual image planes EA, EB and EC. In the superimposition, there are two rows R ₁ and R ₂ of surface detectors in two strip-shaped image plane regions E ₅ and E ₆ , which are each completely covered by the surface detectors. From Fig. 5 it is clear that the arranged in the image plane EA surface detectors ren 31 , 32 , 33 and 34 each have a distance Δa from each other, wel cher by the sum of the widths of the surface detectors arranged in the image planes EB and EC each at a distance from each other , for example 41 and 51 , is given. The same applies to the distances and widths of the other area detectors. In the present case, square area detectors are chosen, the widths a, b and c of which are the same. However, rectangular surface detectors whose widths are different can also be selected without further notice. As can be seen from FIG. 6, a further surface detector 30 is arranged in the image plane EA on the right edge of the usable image field F and in the image plane EB another surface detector 40 is arranged on the left edge of the usable image field F.

If, for example, a total of 23 square area detectors, each with 1000 × 1000 detector elements, are arranged in the overlapping image plane areas E _3, for example in FIG. 4, a usable image field can be covered which corresponds to the image format of a conventional aerial camera of approximately 23 × 23 cm ² corresponds. This can be achieved with a reasonable number of individual detectors as well as a reasonable, consequent adjustment and calibration effort.

Claims (10)

1. Camera for recording remote sensing data, which is intended to be carried in a camera carrier flying over a terrain, with a lens (O) imaging a terrain area (G) in its image plane and at least three surface detectors ( 1 , 2 , 3 ) located in the image plane , which are made up of individual opto-electronic detector elements (d _mn ) arranged in parallel detector lines (D _m ), characterized in that two stripe-shaped, essentially the same length, the width (f _x ) of the usable image field (F) determining and approximately transversely to their longitudinal extension at a distance from one another, the image plane regions (E ₁ , E ₂ ) are covered by one of the surface detectors ( 1 , 2 ), and that at least one further surface detector ( 3 ; 4 , 5 ) is arranged at a distance from the two strip-shaped image plane regions (E ₁ , E ₂ ). 2.Camera for recording remote sensing data, which is intended to be carried in a camera carrier flying over a terrain, with at least two, after installation and adjustment in the camera carrier, each of the same terrain area in their image plane lenses and arranged in the individual image planes, made of opto -Electronic detector elements constructed surface detectors, characterized in that the surface detectors ( 11 , 21 , etc.) are each arranged so that when the individual image planes (EA, EB) are superimposed, two strip-shaped image plane regions (E ₃ , E ₄ ) gaps and are alternately covered by area detectors ( 11 , 21 , etc.) of the individual image planes, and that in at least two of each of the individual image planes (EA, EB) there is a further area detector ( 10 , 20 ) at a distance to the two strip-shaped image plane areas (E ₃ , E ₄ ) is arranged, the latter being essentially eq eich are long and this length determines the width (f _x ) of the usable image field (F) common to the image planes. 3. Camera according to claim 2, characterized in that two lenses are present and in the two strip-shaped image plane areas (E ₃ , E ₄ ) each of the two associated image planes (EA; EB) surface detectors ( 11 , 12 , etc.) at regular intervals (Δa) given by the width (b) of the area detectors ( 21 , 22 , etc.) in the strip-shaped image plane areas of the other image plane (EB). 4. Camera according to claim 2, characterized in that three lenses are present, in the two strip-shaped image plane areas (E ₅ , E ₆ ) of each of the three associated image planes (EA, EB, EC) surface detectors ( 31 , 41 , 51 , etc.) in a row or in at least two parallel, immediately adjacent rows (R ₁ , R ₂ ) at regular intervals, given by the sum (b + c) of the widths of the area detectors in the stripe-shaped image plane areas of the other two image planes (Δa) are arranged and the area end detectors ( 61 ) of a row (R ₂ ) compared to those ( 31, 32 ) of a directly adjacent row (R ₁ ) by more than their own width (a) are such that they are do not touch each other. 5. Camera according to claim 1, characterized in that only one further area detector ( 3 ) is present, which extends almost over the entire ge width (f _x ) of the usable image field (F). 6. Camera according to one of claims 1 to 4, characterized in that two spaced-apart further surface detectors ( 4 , 5 ) are present. 7. Camera according to claim 6, characterized in that the two further surface detectors ( 4 , 5 ) are arranged on opposite edges of the usable image field (F). 8. Camera according to one of the preceding claims, characterized in that the one or more surface detectors ( 3 ; 4 , 5 ; 10 , 20 ; 30 , 40 ) between the two strip-shaped image plane regions (E ₁ , E ₂ ; E ₃ , E ₄ ; E ₅ , E ₆ ) are arranged and thus the length (f _y ) of the usable image field (F) is determined by the distance between the two stripe-shaped image plane regions. 9. Camera according to one of claims 2 to 4 and 6 to 8, characterized characterized in that at least one lens (O) is available and min at least one of the other lenses by a beam splitter stem (S) is replaced in the beam path of an existing lens. 10. Camera according to claim 9, characterized in that the beam divider systems are semi-transparent mirrors (S).