EP1188035A1

EP1188035A1 - Method and device for scanning objects

Info

Publication number: EP1188035A1
Application number: EP00925063A
Authority: EP
Inventors: Klaus KÖRNER; Hans Tiziani
Original assignee: Universitaet Stuttgart
Current assignee: Universitaet Stuttgart
Priority date: 1999-04-29
Filing date: 2000-04-01
Publication date: 2002-03-20
Also published as: DE19919584A1; AU4391300A; WO2000066972A1; DE10081176D2

Abstract

The invention relates to a device for detecting objects in a three-dimensional manner. The inventive device comprises at least two illustration systems which are provided with projection lenses facing the object. At least one illustration system is configured as an observation system for observing objects and at least one illustration system is provided with an elementary means that can be moved in front of the projection lens. The elementary image of the elementary means moves through the object space on an image point line. According to the invention, the elementary means is configured as a moveable elementary means with a lateral component in relation to the optical axis of the projection lens. The observation system is arranged along the image point line for observation.

Description

Title: Object scanning method and apparatus

description

The present invention relates to the preambles of the independent claims. The present invention is concerned with the three-dimensional scanning of objects.

The 3D shape of surfaces is often measured using strip triangulation methods. As a rule, the object or the scene is illuminated from a large angle of incidence, for example between 30 ° and 60 °. However, this leads to disturbing shadowing of the object.

Discontinuous surfaces often pose a problem in optical 3D measurement. For example, larger paragraphs in the surface of the object can lead to a violation of the sampling theorem. This is remedied by the Gray code method, in which a sequence of binary images is projected on. One example is the COMET-500 system from Steinbichler Optotechnik GmbH. In order to achieve a large depth of field when measuring deep objects, all of the methods mentioned are usually dimmed down, both in terms of lighting and imaging the object surface. The company GF in D-14513 Teltow offers digital light projection on the basis of illuminated micro mirrors, digital micromirror devices. Grid images with a repetition frequency of approximately 10 Hz can be obtained generated and read. However, this frequency is not yet sufficient for high-speed imaging.

Patent specification WO 92/14118 describes a 3D measuring arrangement, which is referred to as "confocal", which evaluates the contrast of a striped pattern projected onto the object surface. Both the lighting and the imaging lens or a common lens are each focused on the same level in the space of the object. However, no possibility is given to achieve high accuracy for object distances in the decimeter and meter range.

Devices for 3D scanning of objects are known, for example, from DE 197 49 974, the article "Optical 3D measurement using structured light" by R. Kowarsid, J. Gerber, G. Notni, W. Schreiber and P. Kühmstedt in "Technisches Messen", Vol. 62, 1995, pp. 321 - 329. The essay "Phase-shifting grating projection moire topography" by Y.-B. Choi and S.-W. Kim in Opt. Eng. 37 (3) 1995-1010 with the scanning of three-dimensional objects using shifted shading grids. A signal is evaluated depending on the shifting of a shading grid. It should be noted that the term "shading grid" is used to clarify that diffraction properties of a grating are generally not important for the present invention. Unless otherwise stated, this also applies below.

While there are basically different ways to scan a three-dimensional object, problems arise when a large three-dimensional object is fast and precise to be scanned.

The aim of the present invention is to provide something new for commercial use.

The aim is achieved with the features of the independent claims.

In a first embodiment, the invention thus proposes a device for three-dimensional object detection with at least two imaging systems which have imaging optics facing the object, at least one being designed as an observation system for object observation and at least one having an elementary means which is movable in front of the imaging optics and whose elementary image moves on a pixel line through the object space, it being provided that the elementary means is designed as an elementary means which is also movable with a lateral component to the optical axis of the imaging optics, whereby the pixel line is inclined to the optical axis of the imaging optics, and the observation system for observation along the pixel line is arranged.

A first essential aspect of the invention is thus to be seen in the fact that two separate optical systems are provided with elementary means which are movable, the observation being carried out with one of the two systems along the pixel line. The elementary means can move against one another.

It is possible to set up a stereo camera in this way. In this case, the imaging systems represent systems. Detector arrays, in particular CCD arrays, can be used which are moved synchronously, for example with piezo actuators. The signals obtained from these detectors can preferably be selected by detecting when a specific object area results in a specific signal behavior. In the case of luminous objects, for example in the case of brightly illuminated or even luminous objects, this will be the case if a particularly large signal is detected by the observed object point in the first and second arrays.

However, it is not absolutely necessary to equip the device according to the invention as a 3D camera with a stereo objective. Alternatively, a scanner can be provided in which one of the imaging systems comprises an illumination system. This lighting system will preferably illuminate the object with a series of separate lighting elements which implement the movable elementary means. Alternatively, it is possible to simulate a series of light fields which simulate the elementary center movement as a virtual movement, that is to say are excited according to a predetermined pattern. As a rule, the observation system will comprise a multiplicity of observation elements, wherein an object area can be assigned to each observation element. An advantage of the arrangement according to the invention then consists in particular in that the evaluation of the signals can take place by the detection of the surface being accepted when a specific signal, such as a signal maximum, is recorded on a specific pixel of the array. The pupil of the observation system is preferably arranged in the focal plane of the imaging optics and / or at least essentially.

The invention thus solves the task of flat testing the 3D shape of technical and natural surfaces of objects in space and scenes, preferably with dimensions in the range above one millimeter. The invention thus enables the rapid acquisition and testing of the 3D shape of bodies in scenes with a large depth. Complete scenes can be recorded in real time.

Furthermore, the light output required for the illumination of object surfaces in a scene is sometimes greatly reduced. This is due to the fact that an object is recorded when a particularly large amount of light reaches the detector. A further improvement is the significant increase in the evaluation speed for 3D acquisition. There is the technical possibility of making the 3D point cloud of the object or scene available in video clock.

Protection is also sought for the processes according to the independent claims. In a first method for 3D recording of at least one object surface in at least one scene, at least one electromagnetic radiation source is arranged and this is formed by means of at least one structured array as at least one structured, luminous array with at least two surface elements. At least one surface element lights up. The structured, luminous array can also be understood as a transmitter array and the luminous surface elements can be understood as transmitter elements of this transmitter array. Furthermore, the structured luminous array can represent an array of controllable micro light sources, for example micro light emitting diodes. However, the radiation source can also act as an unstructured radiation source in a structured array, which can be a transmission or a reflection grating. be upstream. In any case, the luminous areas of the structured luminous array represent luminous flat elements in a luminance distribution in the structurally luminous array. Furthermore, at least one illuminating beam path is arranged with at least one illuminating lens, which is assigned to at least one structurally luminous array. This is how an image is realized and the object surfaces can be illuminated in a structured way. Furthermore, at least one imaging beam path for imaging elements of the object surface and at least one receiver array with at least two elements and at least one imaging lens assigned to the receiver array are arranged. Elements of the receiver array detect radiation from elements of the illuminated object surface in the recording process. Furthermore, elements of the receiver array are always formed by the imaging lens with a geometric-optical focus volume in the object space that corresponds to the scene space. At least one luminous flat element of the structured luminous array can experience a shift. By imaging the luminous flat element or elements with the illumination lens, an image of at least one luminous flat element in the object space is formed with a geometrical-optical sharp volume. The receiver array can be a target with a coating that is sensitized to X-ray, UV, VIS or IR radiation and is scanned out. Furthermore, in an arrangement for 3D recording, the receiver array can be designed as a CCD matrix camera. This enables optimal image acquisition for standard tasks with a good signal-to-noise ratio. Furthermore, the receiver array can be designed as a CMOS matrix camera. The random access to pixels enables tracking of moving elements of the object surface in space. The detection of radiation from the elements of the object surface by the elements of the receiver array takes place in a time range Δt _B in which the displacement of at least one luminous flat element of the structured luminous array is also carried out, at least one signal value being obtained in each case. It is within the

Time range Δt _B at least with a luminous flat element of the structured luminous array has been carried out an at least approximately predetermined displacement - including a predetermined optical displacement thereof as a result of a geometrical-optical path length change - and so transmits at least one luminous flat element to different ones Points in time at at least two different locations. The focus volume of at least one image of at least one luminous flat element of the at least one structured luminous array, this focus volume being formed in the object space, and the focus volume of at least one image of at least one element of the receiver array, this focus volume also in the object space is formed, and at least one element of the at least one object surface is coincident at least approximately once due to the implementation of the predetermined displacement of at least one luminous surface element of the structured luminous array with at least one displacement component, parallel to the optical axis of the illumination objective. Thus, the coincidence of the sharp volume of an image of a luminous flat element of the structured luminous array and the sharp volume of an image of an element becomes at least once and at least approximately in the object space of the receiver array and at least one element of the at least one object surface. When the coincidence occurs, at least the element of the receiver array involved in this coincidence experiences at least once a time-varying irradiation compared to the case of non-coincidence, and so this element of the receiver array detects a changed signal at least once.

A luminous flat element can be firmly bound to a structure of a body, for example to a maximum of transparency on a displaceable transmission grating in connection with a radiation source. For the shift, the positions of the luminous flat elements of the structured luminous array and the positions of the images of the luminous flat elements in the object space are determined and determined according to the Newtonian imaging equation from the position of the illumination lens in the 3D image arrangement and the focal length f _{B of} the illumination lens realized as far as the associated approximations are acceptable. The shift is preferably carried out at a constant speed.

In order to achieve a large depth of field for the imaging of object surfaces in a scene for the 3D recording method, the locations of certain relative luminance of the structured, luminous surface and thus also the locations of local extremes of the luminance in the array space can be shifted. The structured, luminous array can be an electronically controllable, structured, luminous array, for example an LCD with an upstream radiation source, which is shifted in a straight line by a movement system. These distances can be understood in the object space as traces of successively imaged light points are, for example, as the shifted extrema of the luminance in an illuminated line grating at the maximum transparency or the trace of an image of an illuminated slit. The traces of the light spots can be observed on an object surface if the image of a light spot and the observed point of the object surface at least approximate each other. Due to the triangulation effect, a lateral migration of the image of the light point can be observed when the image of a luminous flat element is shifted and the image of an element of the receiver array that coincides with it. The offset from the starting position increases with increasing deviation of the illuminated area of the object surface from the current point of coincidence of the two images, the element of the receiver array detecting an increasingly blurred image of the luminous flat element.

This method enables a predetermined displacement process of the structured, luminous array in a displacement direction with a component mz _A direction to make a clear statement about the presence of an element of the object surface at a predetermined location in the object space. The amount in z _A direction is selected such that the sharp surface gradually passes through the object space from a close range to a far range through a predetermined, controlled displacement of the luminous surface elements of the structured luminous array. This method is carried out with the entirety of the luminous surface elements of the structured luminous array and the entirety of the elements of the receiver array for the entirety of the elements of the object surfaces in the detection volume of the 3D recording arrangement. By multiple detection and removal read during the displacement process of at least one element of the receiver array, an interpolation can be carried out to improve the accuracy of the determination of the location of a detected element of the object surface. The aperture diaphragm of the imaging lens can in this case preferably be made small, for example the relative aperture can be 1:22, so that the sharp volume of the image of the elements of the receiver array has a great depth. In contrast, the lighting lens can have a comparatively large relative aperture. For example, the relative opening can be 1: 2. As a result, the sharp volume can have a small depth. In the case of the predetermined displacement of a luminous flat element in the array space, the focus volume in the case shown here moves in each case an image of a luminous flat element in the

Sharp volume of an image of an element of the receiver array. An element of the object surface can thus be permanently in the sharp volume of an image of a receiver element. However, it is only when the sharp volume of the image of a luminous surface element coincides with an element of the object surface that this element of the object surface is structured. Thus, by means of the element of a receiver array that is read out multiple times during the predetermined displacement of a luminous surface element, a signal curve with a relative maximum at the time of the coincidence can be detected. The predetermined displacement of the luminous surface elements can be controlled electronically. In addition, an electronically controlled change in the optical path length in the space in front of the receiver array can also be carried out.

Furthermore, a method for 3D recording of object surfaces areas proposed in a scene, in which the luminous surface elements are shifted relative to the illumination lens, each in a separate displacement distance, preferably in the time intervals Δt _{x of} the detection of light. In a luminance distribution, the luminous surface elements preferably have an at least approximately predetermined constant luminance at least at a point in time t _x within a time interval Δt _x . Furthermore, the luminous surface elements are each positioned on a B segment BS _A , the B segments BS _{AJ being} the target locations for the luminous surface elements at a point in time tj. represent within the time interval Δt _x . The images of these B sections BS _A3 are preferably formed in the object space by imaging with at least one illumination lens to form a section bundle SBi with a convergence point Ki. The convergence point Ki is at least at a distance d _K ι mm from the optical axis of the illumination lens from the last part of the distance d of the pupil center PZ _{0B of} the illumination _lens from the pupil center of the most distant imaging lens. Accordingly, the sensitivity to depth is low. The maximum distance dκι max is five times the same. The value dki = d is preferably realized. At least in a time range .DELTA.t _B during the shift operation of the illuminating surface elements are in each case exactly one image by a receiver element and in each case exactly one image of a luminous surface element in the object space at least at a single time t _± within each time interval Δti the detection of at least approximately together on the Image of a B section BS _A3 positioned. So at least at this point in time tj _. a pair is formed in the object space from the image of a receiver element and the image of a luminous surface element and so such pairs are generated in the object space and these are pushed through the object space. Sharpness volumes of images of the luminous surface elements coincide with surface elements of the object surface at least once in the displacement process. At the location of these pairs, a current point of coincidence is formed at this point in time t in the focus of the current cutting volume of the focus volume of the two images. The elements of the receiver array preferably detect a signal curve with at least one relative extremum of the signal size in the time interval Δt _{x of} the coincidence. During the shift, the positions of the luminous flat elements of the structured luminous array and the positions of the elements of the receiver array are always determined from the position of the illumination lens and the position of the imaging lens in the 3D image arrangement and

The focal length f _{B of} the illumination lens and the focal length f _{A of} the imaging lens are determined and implemented. In the object space, both the luminous surface elements of the structured luminous array and the elements of the receiver array are at least approximately mapped in the same plane in a part of the object space. The electronically controlled displacement of the luminous surface elements to another location can take place using micromechanical means. An electronically controlled shift is also possible. The average object point brightness and the color information can also be obtained from the signal curve by using a color camera.

Furthermore, a method for 3D recording is proposed, in which Δt preferably in the time intervals _. the detection of light, a luminous surface element is positioned on each B section BS _A -. The B routes BS _A -, will thereby directed at the pupil center PZ _{0A of} the imaging _lens in the array space, so that the convergence point Ki is at least approximately positioned in the pupil center of the imaging lens. Furthermore, the convergence point Ki is also positioned in the pupil plane of the illumination objective, and so during the displacement process an image of a receiver element and an image of a luminous flat element in the object space are at least approximately together on the image of a B section BS _A -, positioned. For example, a pair with a fixed assignment can be formed from the image of a receiver element and the image of a luminous flat element in the object space and, during the displacement process of the luminous flat elements, one image from a receiver element and one image from a luminous flat element in the object space, at least approximately once to be brought to coincidence. This includes the case with two central perspective lenses with competing main planes. In this case, the receiver array can be fixed and set such that the "continuous" focus area or the focus plane of the illumination lens coincides at least once with the focus plane of the imaging lens. It is advantageous if the "continuous" focus plane of the illumination lens is always in the comparatively large depth of field of the imaging lens remains. This approach can be implemented with electronic grids with a very high number of pixels. Electronic grids can be continuously stretched or compressed in the displacement process in order to meet the condition of the convergence of the track bundle.

Furthermore, a method for 3D recording of obect areas proposed in a scene, in which a luminous surface element is preferably positioned at least at one point in time ti within each time interval Δti with an at least approximately constant relative luminance in each case on a B path BS _Aj in the time intervals Δti of the detection of light. The convergence point Ki is positioned at least approximately in the focal plane of the illumination objective in the object space and additionally in the pupil center PZ _{0A of} the pupil of an imaging objective in the object space. During the displacement process, an image of a receiver element and an image of a luminous surface element in the object space are positioned at least approximately together on the image of a B section BS _Aj at least at a time t within each time interval Δt of the detection, and so at least At this point in time ti, a pair with a fixed assignment is formed in the object space from the image of a receiver element and the image of a luminous surface element, and pairs with a fixed assignment are thus generated in the object space. The B sections BS _Aj are positioned parallel to a straight line g _AP , the straight line g _AP intersecting the focal point F _{AB of} the illumination lens in the array space and the increase with the amount from the quotient “distance from the pupil center PZ _{0A of} the pupil of the Imaging lens in the object space from the axis of the lighting lens and the focal length f _{B of} the lighting lens ", this increase in the straight line g _{AP being} related to the axis of the lighting lens. Two central perspective lenses with mutually inclined axes can be used. The array can preferably be fixed and set in such a way that the “continuous” focus plane of the illumination objective is included at least once the focal plane of the imaging lens coincides.

At least approximately, a straight-line relative displacement of the receiver array relative to the imaging lens is carried out parallel to the optical axis of the imaging lens, and during the shift, signal values are read out successively from a receiver element in each case, and so a signal curve is formed by means of a receiver element and generated when imaging multiple times Displacement sections of elements of the receiver array with the imaging lens are at least approximately formed from their images in the object space at least approximately one section cluster SB ₂ with a convergence point K ₂ in the focal point F _{0A of} the imaging lens. The displacement of the receiver array is carried out in such a way that, during the displacement process, an image of a receiver element and an image of a luminous flat element in the object space are at least approximately together at least at a time t _x within each time interval Δt _x the image of a B segment BS, are brought to coincidence and shifted, and pairs of images are thus generated in the object space. Since each element of the receiver array enables the acquisition of a signal curve, parallel processing is possible. Furthermore, the imaging lens can also be moved to the receiver array.

Furthermore, a method for 3D recording of object surfaces in a scene is proposed, in which preferably the convergence point Ki of the route bush egg SBi together with the convergence point K ₂ of the route bush egg SB ₂ in the object space both with the focal point F _0A and with the pupil center PZ _0A Pupil of the imaging lens at least anna- are brought to coincidence, with the illumination lens and the imaging lens being at least approximately telecentric on the array side. The luminous surface elements are shifted at least approximately parallel to a straight line g _A on sections. The straight line g _A penetrates the focal point F _{AB of} the illumination lens in the array space. For the straight line g _A , the increase is realized with the amount from the quotient "focal length of the illumination lens and distance d of the focal point F _{ÄÄ of} the imaging _lens from the axis of the illumination _lens in the object space", this increase in the straight line g _A to an axis perpendicular to the axis of the illumination lens "and because of the telecentricity of the imaging lens in the array space the line g _A coincides with the line g _AP in this case. This method enables 3D recording in a very large depth measurement range, an arrangement with parallel and at least approximately identical lenses can advantageously be selected. The structured lighting is preferably carried out with an illuminated line grid. For the receiver array, a linear shift is carried out parallel to the optical axis of the imaging lens. In this way, a separate displacement path is generated for each element of the receiver array. When these displacement paths are imaged with the imaging lens, a second cluster of routes with a convergence point K of the imaged routes in the object space at the focal point F _{0A of} the imaging _lens is created from the images of these displacement routes. Furthermore, the convergence point Ki and the focal point F _{0A of} the imaging _lens in the object space are brought at least approximately to the coincidence. The point of convergence K _{x of} the distances in the object space is formed by the locations of certain relative luminance of the illuminated line grid, for example the maxima of the transmis- sion of the line grid, are displaced at least approximately parallel to a straight line g _A on displacement paths. With this method, periodic signals with a modulation maximum can be detected in the elements of the receiver array, from which the information about the absolute phase of an object point in connection with the arrangement can be obtained. If the illuminated line grid is moved at a constant speed, periodic signals with a constant frequency can be obtained in the elements of the structured illuminated array. This simplifies the signal evaluation and can therefore lead to a considerable reduction in the computing time. The sharpness areas are brought to coincidence by the synchronous positioning of the structured luminous array and the receiver array.

The position of the at least one luminous flat element can also be stationary and in this case at least components of the illumination lens move.

Furthermore, a method for 3D recording of object surfaces in a scene is proposed, in which preferably one luminous surface element each in the time intervals Δt _{x of} the detection of light in a time range Δt _B at least approximately at its own location 0 _AB3 in the structured light Array is arranged relative to the lighting lens and brought to light by control and is imaged by the lighting lens and this luminous _{surface element is always} mapped to a predetermined location in the object space 0 _0B3 at least at a point in time t _x within the time interval Δt _x . This image location 0 _0B3 of a luminous _{flat element} is shown in the object space a control is changed by actuating a different, predetermined flat element and lighting it up, so that the image of a luminous flat element on a controllable path curve, structured from distance increments AI _{0 of} the images of the distances AI _{A of} the luminous

Flat element in the array space, through which object space is pushed - in the sense of the controlled acceptance of predetermined, different positions. In each position after the shift - by an integer multiple n, including ln = 1, of the distance increment AI ₀ - at least one signal value is detected and read out from a receiver element, and so several processes of detecting and reading out elements of the receiver Arrays formed a waveform. The location of the detected and read element of the receiver array is continuously changed. The locations of the detected and read elements of the receiver array are located at locations 0 ^ in the receiver array and the image of this location 0 _M] , the image location 0 _0A: ,, is optical with the predetermined image location 0 _{0B] of} the luminous _{flat element} in the object space conjugated. Thus, an image of a detected and read element of the receiver array is coincident in the object space with the image of a luminous flat element at least at a time t _x within the time interval Δt _x , and thus a pair of images with changing images is generated, which takes on different positions in the object space. The object space is gradually penetrated in depth by such pairs. Here, sharp volumes of the image of a luminous flat element each coincide with a flat element of the object surface at least once in the time range Δt _B in a time interval and the detected and the read elements of the receiver array tinterval Δti of the coincidence on a signal curve with at least one relative extremum of the signal size, the time range Δt _{B being made} larger than the time interval Δti and thus at least one time interval Δti being temporally fitted into the time range Δt _B.

This method can be implemented without any movement of an array, ie completely electronically. The structured, luminous array and the receiver array can represent rigid, preferably three-dimensional structures; for example, the luminous array can have luminous diodes or vertically radiating laser diodes in a 3D arrangement. Little by little, individual surface elements are electronically controlled and illuminated. By controlling predetermined, luminous surface elements and reading out elements of a receiver array, whose images in the object space represent a pair of images, an extremum in the signal value of an element of the receiver array is obtained if and only if the image pair with an element of the ob - jektobe flat at least approximately coincides. A luminous surface element represents a small volume element due to its fixed position in the composite of the structured, luminous array and the parameters of its representation in the object space. By reading out exactly the element of a receiver array that has an image in the object space that is at least approximately optically conjugated with the image of the luminous surface element, the presence of an element of the object surface in this volume element is queried. Furthermore, the distance increments, which belong to the images of the luminous surface elements, preferably form sections on a straight line in the object space, which form a section bundle SBi with a con vergence center K _x belong. This ensures that the signal acquisition is unambiguous, since the sections of the SBi route bundle do not cross in the object space. With a pπori knowledge of the object surface, individual sections can be put together from distance increments. The control of various luminous surface elements simulates the real displacement of a luminous surface element. Basically, the mechanical array of the luminous array can represent a 3D model of the object surface to be examined and the luminous surface elements are simultaneously mapped onto the object surface. This also applies analogously to the structure of the receiver array. This can also have an object-adapted 3D structure. Both the luminous array and the receiver array can have several surfaces with luminous elements or receiving elements in the depth, so that the detection of three-dimensional object surfaces of great depth is possible without mechanical displacement. In this way, unknown object surfaces can be detected in a defined measuring volume.

Furthermore, a method for 3D recording of object surfaces in a scene is proposed, in which preferably one with object surfaces illuminated by a radiation source with a first and at least a second imaging beam path between the two axes of two imaging lenses for imaging the object surfaces Symmetry line is formed. At least one receiver array is assigned to each imaging beam path, and the two receiver arrays each have elements that detect light from the elements of the illuminated object surfaces in the object space in the acquisition process in the time range Δt _B the two receiver arrays are each moved to another location during the recording process. At least approximately at the same time, the light from the elements of the object surfaces is detected for a period of time Δt _x by the elements of the receiver array, and the elements of the receiver array are then read out, signal values being obtained in each case. There is preferably illumination of object surfaces in the foreground and, if appropriate, also of the more distant background of the scene. During the recording process, the two receiver arrays are shifted simultaneously on shift paths AS _A χ and AS _A2 . The images of the displacement sections AS _Ai and AS _A2 , the sections AS ₀ ι and AS ₀₂ , are positioned in the object space at least approximately on the line of symmetry between the two axes of the lenses. A convergence point K _{2i is} formed from the route _bundle SB _{2 and} the images of the displacement routes AS _{Alj of} the individual elements of the first _receiver array, the routes AS ₀ ij, and from the route bundle SB _{22 of} the images of the displacement routes AS _{2j of} the individual elements of the second receiver array, the lines AS ₀₂₃ , a convergence point K _{22 is} formed and the convergence point K _X2 and the convergence point K ₂₂ are brought to coincidence on the line of symmetry and form a convergence point K ₀ on the line of symmetry and the two receiver arrays shifted such that their images at least partially coincide in the object space, so that the images of the elements of the first receiver array and the images of the elements of the second receiver array in the object space are brought at least approximately in coincidence in pairs, the pair-forming elements of the represent corresponding elements in both receiver arrays. For example, a current point of coincidence is preferably formed from two images of elements, which is shifted through the object space. is pushed. This is preferably done with all elements of the receiver arrays. Signal profiles Si of the first receiver array are preferably formed by reading out the elements during the displacement of the first receiver array. The displacement of the first receiver array is carried out parallel to a straight line g _AlP and so the elements of the first _{receiver array are} shifted at least approximately parallel to a straight line g _AiP on displacement _paths AS _Aij . Furthermore, signal curves S _{2 of} the second receiver array are formed by reading out the elements during the displacement of the second receiver array, and the displacement of the second _receiver array is carried out parallel to a straight line g _A2P , and the elements of the second _{receiver array} are thus Arrays shifted at least approximately parallel to a straight line g _A2P on displacement _paths AS _A2j , the displacement of the second _{receiver array} taking place at least approximately simultaneously with that of the first receiver array. The line g _AiP is cut at a point P _Ai on the symmetry line in the main plane of the first imaging _lens in the array space and the straight line g _A2P is cut at a point P _A2 on the symmetry line in the main plane of the second imaging _lens , the straight line g _AiP additionally contains the focal point F _{Ai of} the first imaging _lens and the straight line g _{A2P contains} the focal point F _{A2 of} the imaging _lens in the array space. Due to the natural structuring of the illuminated or even self-illuminating object surface, the signal profiles S ₃ and S ₂ j recorded in each element of the receiver array are more or less modulated. The z ₀ position of the respective associated element of the object surface is to be determined from the evaluation of this modulation, which occurs particularly on the sharply depicted elements of the object surface. The two signal profiles Sij and S _2j of two corresponding elements lj and 2j Receiver arrays are stored in the memory of a computer by moving the two receiver arrays. The elements of two receiver arrays represent corresponding elements, the images of which coincide in the object space in a focus volume at least at one point in time. Thus, an element of the first and an element of the second receiver array in a common focus volume form at least one Time of a pair of corresponding elements. Now each of the two signal profiles Si- and S ₂ - by means of a window function with at least one single window, with a minimum window length corresponding to two signal values and a maximum window length that is at least approximately the length of the signal profiles Si-, and S ₂ - , corresponds to overlapping signal pieces Si part ₃ and S _{2 te} ιi ₃ in each of the two signal profiles Si and S _{2] formed} from the windows. Window lengths with a length of, for example, 8 or 16 signal values are advantageous. This window function is synchronously shifted by at least one signal value, which corresponds to an increment of the shifting of the receiver arrays, over each of these two signal profiles S ₃ and S _2] and from each current window in the position k, with 1 <k < m, a signal piece Si part position ₃ and S _{2 t} eιi position k _{3 is} formed. In this case, these one after the other signal pieces Si _te i _l _D Posiion k] and S ₂ formed covering eil ₃ position ₃ n _j each of the two waveforms Si and S ₂ -, in a part region, wherein in each case at the same end of the two waveforms Si, and S _{2] is started} with the shifting of the window function in both signal pieces. Subsequently, starting from two signal pieces in the position 1 Si part Po _s i _t i _o n _{I 3} and

S ₂ part position ι _j i calculates the cross-correlation function, but first inverting one of the two signal pieces, that is to say mirroring all the values of the same, and so from an original signal piece S _{x Teιl pos} i _t i _{on 1 3} and an inverted signal piece S _{2 Teιl} P _os i _t i _{on 1 INV 3,} the maximum of the cross correlation function MCC i ₂ - is calculated _pos i _t i _{on 1} and stored . The inversion is necessary in order to obtain correctable signals, since the imaging beams of the elements of a corresponding pair move in opposite directions along a track in the object space during the displacement in an at least approximately identical section of the scene, that is, for example, towards one another. This track lies parallel to the main section of the 3D recording arrangement. After calculating the maximum of the cross-correlation function MCC i 2 j position ι i of position 1, the window function is shifted to position 2, so that for the next two signal pieces the maximum of the cross-correlation function MCC i _{2 3} position 2 in the manner described to the window function at the other end of the two waveforms Si, and S has arrived m in the _2-position], and again the maximum MCC i ₂ is calculated, -, _pos i _t i _on m of the cross-correlation function MCC i _{2 3} position is determined m . A maximum value curve is formed from the m calculated maxima MCC _m , the resulting maximum M _{m 3} again being determined in this maximum value curve and the location of the maximum M _m - the maximum value curve of the two original signal curves and thus the path of the displacement of the two receiver arrays is assigned. This maximum value curve calculated in this way can have the course of a Gaussian function. In order to prevent errors, an intensity threshold can be used, whereby signal pieces with a very low average intensity are excluded from further processing. The location of the respective maximum M - is defined as the location of the image of the respective element of the object surface associated with the two corresponding elements lj and 2j in the array space. The location of this maximum M - in the array space becomes z ₀ coordinate of the respective element of the object surface in the z ₀ direction and also the x ₀ and y ₀ position of the respective element of an object surface, since the geometry of the 3D recording arrangement is known. In this way, the positions of the elements of an object surface from which signal profiles are recorded can be calculated, the geometry of the 3D recording arrangement being known and the displacements, including the step size of the displacement, of the two receiver arrays being predetermined.

Furthermore, the axis of a first imaging lens for imaging the object surfaces can be aligned parallel to the axis of a second imaging lens for imaging the object surfaces. It is possible that the main plane of the first imaging lens in the array space and the main plane of the second imaging lens coincide at least approximately in a common plane and the receiver arrays are at least approximately jointly in one plane. Thus it is given that the point P _Ai lies on the line of symmetry and the point P _A2 lies on the line of symmetry and so the two points P _A1 and P _A2 are brought to coincidence at least approximately in one point P _A. In this way, the 3D point cloud can also be obtained from free-space scenes in the background of the scene.

Furthermore, a method for 3D recording of object surfaces in a scene is proposed, in which preferably illuminated object surfaces are imaged with a first and at least one second imaging beam path. In the acquisition process, the two receiver arrays are simultaneously and parallel to the respective optical axes of the parallel, at least approximately identical imaging beams. corridors whose main planes coincide are shifted, with the object surfaces in the scene being illuminated. The signal course Siz is formed by reading out laterally adjacent elements of the first receiver array during the displacement of the first receiver array so that exactly those elements of the receiver array are used for signal formation that lie on lines that are parallel to one another Straight lines g _{AiP are} aligned, which intersects the point P _A in the common main plane of the imaging lenses. Thus, the signal curve formed corresponds at least approximately to the signal _curve Si which arises during a real shift parallel to a straight line g _AiP , and the signal _curve S _2z is formed by reading out elements of the second receiver array lying laterally next to one another during the shift of the second receiver array, that exactly those elements of the receiver array are used for signal formation which lie on lines which are aligned parallel to a straight line g _A2P which intersects the point P _A in the common main plane of the imaging objectives. The signal _curve S _2z formed _{thus corresponds} at least approximately to the signal S ₂ _which arises during a real shift parallel to a straight line g _A2P . Thus, at least at a point in time ti in a time interval Δti, a current point of coincidence of elements of the two receiver arrays is formed, which is formed in succession at different predetermined locations in the object space in the time range Δt _B.

The two signal courses Si _j , S _2j of two elements of the receiver arrays, which correspond at least at one time in each case, become, by means of the correlation method already described above, with two windowed signal courses with the piecewise inversion of signal pieces for determining the z ₀ position of an element of the object surface, the z ₀ coordinate of the respective element of the object surface is calculated, and thus also its x ₀ and y ₀ position, and so the entire 3D point cloud calculated from object surfaces in a scene, the geometry of the 3D recording arrangement being known and the displacements of the receiver arrays being predetermined.

Furthermore, the imaging lens can also be moved to the receiver array.

Furthermore, a method for 3D recording of object surfaces in a scene with at least one electromagnetic radiation source is proposed, which is designed as a structured, luminous array with regions of different luminance. Furthermore, at least one radiation source is preferably designed as at least one structured array as a structured, luminous array with luminous surface elements. A structured array in the manner of a line grating with an upstream radiation source can preferably also be used. Furthermore, an electronically controllable line grid can be implemented in the array space. The radiation source and the structured array together form the structured, luminous array. The locations of certain relative luminance of the structured luminous array and also the local extremes of the luminance of this structured luminous array can be made electronically displaceable. The radiation source can be designed for radiation in the visible and in the invisible spectral range, for example in the spectral range from 750 nm to 900 nm. at least one illumination beam path is arranged with at least one illumination objective. The structured, luminous array is assigned to the lighting objective. However, an image of the structured, luminous array can also be assigned to the lighting objective for imaging. The lighting objective has an effective aperture diaphragm with an extension D _B and a diaphragm center BZ _B. The structured, luminous array and the lighting lens are used for structured lighting of the object surfaces in the scene. Furthermore, at least one imaging beam path with at least one imaging stage is arranged for imaging the elements of the object surfaces in the scene. At least one receiver array is assigned to this imaging lens. The imaging lens has an effective aperture diaphragm with an aperture center BZ _A for imaging the elements of the object surfaces. This imaging lens is assigned at least one receiver array with elements that detect light from the elements of the structured illuminated object surfaces in the object space during the recording process. The distance d of the pupil center PZ ₀ B of the illumination lens, as an image of the aperture center BZ _B in the object space, from the pupil center PZ _{0A of} the imaging _lens , as an image of the aperture center BZ _A in the object space, is at least one eighth of the extent D _{B of} the aperture diaphragm of the illumination lens. In this case, an image of a luminous surface element in the object space is formed from a luminous surface element in a luminance distribution with a preferred, at least approximately predetermined constant relative luminance by imaging with the illumination lens. Furthermore, a movement system with preferably at least one movable component is arranged, which is assigned to the structured, luminous array. The displacement distances of the luminous surfaces Chen elements in the array space are preferably formed from the mechanical movement of the structured, luminous array. However, it is also possible for the luminous surface elements to be moved electronically at the same time, for example in the lateral direction, and for the movement system to use at least one movable component to move the structurally luminous array parallel to the optical axis of the illumination objective. After mapping these displacement paths through the illumination lens into the object space, their image is at least approximately formed as a path cluster SBi with a convergence point K _x .

Furthermore, in the case of an arrangement for 3D recording of object surfaces in a scene, the displacement distances of the luminous surface elements can be arranged at least approximately in parallel, and thus the convergence point Ki can be positioned at least approximately in the focal plane of the illumination lens in the object space and in the pupil center of the imaging lens in the object space.

Furthermore, with an arrangement for 3D recording of object surfaces in a scene, the luminous array can be designed as an electronically controllable line grid with controllability of the location of the lines and the line width. The lines can be arranged perpendicular to the main section and the displacement distances of the luminous flat elements and thus also the luminous surface elements with local extrema of the luminance in the array space - as a result of the mechanical movement of the structured luminous array and the electronic control of the structured luminous

Arrays can be formed in the array space. From these displacement distances in the array space in the main section and in each to Main section parallel section plane at least approximately at least one cluster of lines with a convergence point K _{x be} formed. The convergence point K _{x of} the _cluster of lines can be arranged in the pupil center PZ ^ of the imaging objective in the array space.

Furthermore, in an arrangement for 3D recording of object surfaces in a scene, the displacement distances of the luminous surface elements can be arranged at least approximately parallel to a defined straight line g _AP . The luminous surface elements in a luminance distribution preferably have an at least approximately predetermined predetermined constant luminance. The straight line g _AP intersects the focal point F _{AB of} the illumination lens in the array space and has the increase with the amount from the quotient “distance of the pupil center PZ _{0A of} the pupil of the imaging lens in the object space from the axis of the illumination lens and focal length f _{B of} the illumination lens” , this increase in the straight line g _{AP being} related to the axis of the illumination lens.

Furthermore, in the case of an arrangement for 3D recording of object surfaces in a scene, a component of the movement system can be assigned to the receiver array, and thus, during the mechanical movement of the receiver array on a displacement path, its elements displacement paths AS _A;] on parallel straight lines may be assigned, preferably from the images AS _{θD of} these lines AS _A -, when imaging by the imaging _lens at least approximately a line bundle SB ₂ with a convergence point K ₂ in the object space. The convergence point K _x and the convergence point K ₂ with the focal point F _0A and the pupil center can PZ _{0A of} the pupil of the imaging lens in the object space can be brought at least approximately to the coincidence. Furthermore, the imaging lens can be implemented telecentrically on the side of the space of the arrays.

Furthermore, in the case of an arrangement for 3D recording of electronic surfaces in a scene, a component of the movement system can be assigned to the receiver array, and thus the elements of displacement elements AS _A -, on a displacement path during the mechanical movement of the receiver array be assigned to parallel straight lines, at least approximately at least one line tuft SB ₂ with a convergence point K ₂ in the object space being formed from the images of these lines when imaged by the imaging lens. The convergence point K _x and the convergence point K ₂ can be brought at least approximately to the coincidence with the focal point F _0A and the pupil center PZ _{0A of} the pupil of the imaging lens in the object space and the illumination lens and the imaging lens can each be designed telecentrically on the side of the space of the arrays. Thus, the axes of the illumination lens and the imaging lens can be arranged parallel to one another and the focal planes of the same can be brought to coincide in the object space. Furthermore, in an arrangement for 3D recording of object surfaces in a scene, the components of the movement system can be arranged such that in the array space with the focal point F _{AB of} the illumination lens as a reference point for the luminous array, a total direction of movement is at least approximately parallel to a straight line g _A is realized in the array space, so that the elements of the structured, luminous array move on parallel straight lines to the straight line g _A and this straight line g _A with the focal point F _{AB of} the lighting object in the array space is brought to the section and has the increase with the amount from the quotient "focal length f _{B of} the illumination lens and distance d of the focal point F _{ÄÄ of} the imaging _lens in the object space from the axis of the illumination _lens ", this increase in the straight line g _A to a perpendicular straight line Axis of the lighting lens is related.

Furthermore, in the case of an arrangement for 3D recording of object surfaces in a scene, the structured array can be formed at least on a partial area of a disk, which is preferably associated with a rotary precision bearing with a shaft with a rotary motor, so that a rotating disk is formed.

Furthermore, in the case of an arrangement for 3D recording of object surfaces in a scene, the rotating disk can be formed with transparent plate sectors of different geometrical-optical thickness.

Furthermore, in the arrangement for 3D recording of at least one object surface in at least one scene, the receiver array can represent a color camera.

In addition, the use of a special receiver array with RGB channels and a fourth channel, the NIR channel, for example with a wavelength interval of 750 nm to 900 nm, is possible for obtaining the information for the 3D point cloud.

Furthermore, in an arrangement for 3D recording of at least one object surface in at least one scene at least one electromagnetic radiation source, the radiation source is designed as a structured, luminous array with luminous flat elements by means of at least one structured array. At least one illumination beam path with at least one illumination lens, which has an effective aperture diaphragm with an extent D _B and a diaphragm center BZ _B , is arranged for structured illumination of the object surfaces in the object space. The lighting lens is assigned to the structured, luminous array, including an image thereof. In addition, the at least one illumination beam path is assigned an imaging beam path with at least one imaging stage for the at least one object surface with at least one imaging lens assigned to the receiver array or an image of the same for imaging the elements of the object surface, which has an effective aperture diaphragm with a Aperture center BZ _A. A receiver array is used to detect electromagnetic radiation from the elements of the illuminated object surfaces in the object space during the recording process. The distance d from the pupil center PZ _{0B of} the illumination _lens , as an image of the aperture center BZ _B in the object space, from the pupil center PZ _{0A of} the imaging lens, as an image of the aperture center BZ _A in the object space, is at least one eighth of the extent D _{B of} the aperture diaphragm of the illumination lens. The luminous flat elements have an at least approximately predetermined luminance in a luminance distribution, so that at least one image of a luminous flat element is formed in the object space by imaging with the illumination objective. According to the invention, the sharp volume of at least one image of a luminous flat element in a structured luminous array is in the object space - by the predetermined assignment of the luminous flat element to Illumination lens and the assignment of the elements of the receiver array to the imaging lens and the assignment of lighting lens to the imaging lens in the SD recording arrangement using Newton's imaging equation - permanently fitted into the focus volume, which is defined by the totality of the images of the elements of the receiver Arrays is shown in the object space.

The sharpness volume, which is given by the totality of the images of the elements of the receiver array in the direction of beam propagation, has at least as great a depth as the sharpness volume of an individual image of a luminous surface element. In the object space, an image of a luminous surface element of a structured array is permanently assigned to at least one image of an element of the receiver array.

A structured, luminous array with a plurality of permanently arranged luminous surface elements can be designed in a spatial structure from the data record from a known target object surface. In the object space, after the surface elements are depicted, the lighting lens creates images of the same at different locations. In the 3D recording arrangement, at least one element of a receiver array is arranged in the optically conjugated locations in the array space of the imaging objective. When an object is positioned exactly in the object space, which has elements of the object surface where there is an image of the luminous surface elements, the elements of the receiver array each detect a signal value above one

Threshold. In this case, a parallel detection is carried out by the elements of the receiver array. This can be done with a at high speed.

Furthermore, in an arrangement for 3D recording of at least one object surface, the structured array can be designed as a transparent microlens array and the focal length and the axial position of the microlenses can be designed such that their foci are arranged in a 3D area, which at least approximately represents an optically conjugated surface to the target surface. The foci of the microlenses represent at least approximately some optically conjugated locations on the target surface of a test specimen. The deviation from a target position can be determined by determining the focus position in the image.

Furthermore, in an arrangement for 3D recording of at least one object surface on the structured array, at least one relief with a spatial structure with at least one period in the form of at least one ramp with at least one inclined ramp surface in the compensation surface can be formed. On the inclined ramp surface, luminous surface elements are preferably arranged as a binary code pattern. These flat elements are formed by window areas which are illuminated by the radiation source. The ramp surfaces are preferably inclined in such a way that the compensation line AG _A - due to the oblique ramp surface in the main section as shown by the illumination lens in the object space as a picture, provides a straight line AG _0] which at least approximately targets the pupil center PZ of the imaging lens. As a rule, there are several ramps, so that the different compensating lines AG _0] of several different ramps after their imaging by the illumination lens from their images form a bundle of straight lines with one Convergence point K _{x is} formed. The lighting lens is preferably opened high. The convergence point K _x is brought to coincidence, at least approximately, in the pupil center PZ of the imaging objective. This means that when taking pictures of the surface of the object, a ramp can be clearly traced at all depths without problems with lateral misalignments. The imaging lens can be comparatively short in focal length, shorter than the lighting lens, and is so far dimmed that there is a large depth of field. The depth of focus range of the imaging lens thus determines the depth range for the 3D recording arrangement. The main section of the images of the ramps form a cluster with the point of origin in the PZ pupil center. The ramp images intersect the object surface to be detected. At the intersection of a ramp image with the object surface, a sharp image of the mask is created on the ramp surface.

Furthermore, in the case of an arrangement for 3D recording of object surfaces in a scene with two imaging beams with two at least approximately identical, parallel arranged imaging lenses, a first imaging lens and a second imaging lens, the main planes of the two imaging lenses can be brought to coincidence and each of them can be assigned a receiver array with detecting elements, so that a first and a second receiver array are arranged, to which at least one movement system is assigned. The resulting movement of the first receiver array can take place on a segment AS _Ax on the first upper branch of a letter Y and the segment AS _Ax can be parallel to a straight line g _A i _P lie on the one hand intersecting the focal point of the first imaging lens in the array space and on the other hand intersecting the intersection point P _{A of} the line of symmetry between the two optical axes of the two imaging lenses through the coinciding main planes, so that the detecting ones intersect Move elements of the first receiver array on the lines AS _ÄI -, with part of the line of symmetry forming the lower part of the letter Y. The resulting movement of the second receiver array can take place on a segment AS _A2 on the second upper branch of the letter Y, and the _segment AS _Ax can lie parallel to a line g _A2P , which on the one hand is the focal point of the second imaging _lens in the array space intersects and on the other hand intersects the intersection point P _{A of} the line of symmetry between the two optical axes of the two imaging lenses through the coinciding main planes. The detecting elements of the second receiver array can thus move on the lines AS _Ax -. The scene can be an open space scene.

Furthermore, the resulting movement of the first receiver array can take place on a path parallel to the optical axis of the first imaging lens, and precisely the elements of the first receiver array can be read out and a signal curve can be formed from them, which are located on paths AS _Ax - which lie parallel to a straight line g _AxP , which on the one hand intersect the focal point of the first imaging _lens in the array space and on the other hand intersect the intersection point P _{A of} the line of symmetry between the two optical axes of the two imaging lenses through the coinciding main planes. The elements of the first receiver array used for signal formation correspond to those which are located on lines AS _A -, with part of the symmetry Trielinie forms the lower part of a letter Y and the resulting direction of movement of the second receiver array can take place over a distance parallel to the optical axis of the second imaging lens, exactly the elements of the second receiver array are read out and a signal curve is formed from these, which are located on lines AS _A2j , which are parallel to a straight line g _A2P , which on the one hand intersects the focal point of the second imaging _lens in the array space and on the other hand intersects the intersection point P _{A of} the line of symmetry between the two optical axes of the two imaging _{lenses through the coinciding} main planes . The elements of the second receiver array used for signal formation correspond to those which are located on lines AS _A2j .

Furthermore, in an arrangement for 3D recording of at least one object surface with two imaging beam paths with two at least approximately identical, parallel arranged imaging lenses, a first imaging lens and a second imaging lens, the main planes of the two imaging lenses can be brought at least approximately to the coincidence and each each of which can be assigned a receiver array with detecting elements, so that a first and a second receiver array with elements are arranged, and the first and the second receiver array each have at least one receiver surface, each of which is perpendicular to the main section . The receiver _{surface of} the first receiver array preferably contains a _segment AS _Ai that _lies parallel to a line g _AxP on the first upper branch of a letter Y, which on the one hand intersects the focal point of the first imaging _lens in the array space and on the other hand that Penetration point P _{A of} the symmetry line SL intersects between the two optical axes of the two imaging lenses through the coinciding main planes, so that the detecting elements of the first receiver array are arranged in the main section on the line AS _Ax . Part of the line of symmetry SL preferably forms the lower part of the letter Y. At least one receiver surface of the second receiver array is preferably on a path AS ₂ on the second upper branch of the letter Y parallel to a straight line g _A2P , which on the one hand is the focal point of the second imaging lens in the array space and on the other hand intersects the intersection point P _{A of} the symmetry line SL between the two optical axes of the two imaging lenses through the coinciding main planes, so that the detecting elements of the second receiver array are arranged in the main section on the path AS _Ax . This arrangement enables the detection of illuminated elements of the object surface in the object space on a plane perpendicular to the main section. The structure of the receiver matrices is the same and is arranged in a position symmetrical to the symmetry line SL and at the same height.

The procedure is preferably as follows: The signals from the two receiver areas are preferably read out line by line, so that the receiver area of the first receiver array supplies the signal profiles Si and the receiver area of the second receiver array provides the signal profiles S ₂ . These signal profiles are evaluated line by line, the lines containing the corresponding elements at the same distance from the main cut. In order to find the location of an element of the object surface in the object space, the evaluation is carried out in accordance with the correlation described above. ons procedure with two windowed signal curves for use. Here, too, signal pieces are generated by a window function. A signal piece is inverted by mirroring the signal values. The cross-correlation is carried out in each case by an original signal piece and in each case by an inverted signal piece, the signal pieces each representing symmetrically arranged line sections in the 3D arrangement, and a correlation coefficient is obtained and stored in each case. The window of the window function, which can have a length of 64 pixels, for example, is shifted, for example, in increments of one, which here corresponds to a pixel in the respectively evaluated line. For overview measurements, the window can also be shifted more than one pixel. The length of the window is chosen depending on the relative opening. The window length can also be designed variably. The z ₀ position of the elements of the object surface is thus determined in a plane perpendicular to the main section in the symmetry line SL.

Furthermore, in the case of an arrangement for 3D recording of at least one object surface, a lighting objective can be assigned a first imaging objective with a receiver array and a second imaging objective with a receiver array, the pupil center PZ _{0A of} the first imaging _objective at a distance d from the pupil center PZ _0B of the lighting lens is arranged. A spatially structured receiver array is assigned to each of the two imaging lenses, so that a first and a second receiver array are arranged in the array space. The first and the second spatially structured receiver array at least two receiver areas each on spatially separate areas and the receiver areas of the first and the receiver areas of the second receiver array are each arranged such that at least approximately pairs of optically conjugated images of at least parts of receiver areas of the first receiver array and of parts of the receiver areas of the second receiver array are formed in the object space. Here, too, the evaluation is carried out using the correlation method with two windowed signal profiles, as already shown above.

The invention is described below by way of example only with reference to the drawing. In this show:

Figures 1 to 10 different representations of the invention. 11 to 15 for understanding the invention and its

Explanation of important details.

The basics of the invention are first discussed with reference to FIGS. 11 to 15.

11 shows the illumination of an object with an object surface 200 by means of a lens 201 by means of a light source realized by a pinhole 202. The opening of the pinhole 202 is arranged on the dashed optical axis 203 of the lens 201. 11a, 11b and 11c differ in the distance of the pinhole from the lens, as indicated by the distances A, B, C.

In Fig. 11a, the distance A between pinhole 202 and

Main plane of the lens 201 selected so that the pixel of the

Pinhole 202 is inside the object. Accordingly, the light cone projected into the object space from the pinhole 202 is comparatively extended on the object surface, as indicated by the area A ^λ .

In Fig. 11b the distance between the pinhole and the lens is chosen so that the pinhole is imaged exactly on the object surface. The illuminated area on the object surface is very small. In Fig. 11c, the distance between the pinhole and the lens is so large that the light cone has its smallest extent in front of the object surface and diverges again towards the object surface. Depending on

Distance further differs the light spot size. It should first be noted that the pixels for the situations in FIGS. 11a, 11b, 11c lie on a straight line, as indicated in FIG. 11D. It should also be noted that the different dimensions of the light cones in the situations of FIGS. 11a to 11c lead to a differently high light intensity on the surface and accordingly the illuminated spot appears more or less bright. This is symbolized in FIGS. 11a to 11c by the arrows, the length of which represents the intensity of the light scattered on the surface in any direction. The highest intensity of the scattered light occurs in FIG. 11b. FIG. 12 once again illustrates the relationship between pinhole spacing, light spot expansion and intensity of the scattered light.

The light from the surface of the object 200 is diffusely scattered on typical surfaces, so that it can be observed from different viewing directions. This is shown in Fig. 13.

In principle, with a single light source such as a pinhole, it is now not necessary to determine where a particularly advantageous observation position is present. As a rule, this takes place Illumination of an object but simultaneously with a large number of illumination points, for example with a grid or stripe mask for shading a light source. This makes signal evaluation difficult. While with a simple pinhole diaphragm it is easily possible to determine the surface point by simply observing the signal strength, further difficulties arise with grille diaphragms or the like, since it still has to be identified where the strongly illuminated object area of a particular th illumination point lies on the object.

The spatial assignment is now typically carried out using a detector array by evaluating the signals thus obtained. In order to facilitate the assignment and evaluation, it is desirable if a certain area of the object surface always falls on the same element of the detector array. For this purpose, another effect known from optics is used in the invention, which is illustrated with reference to FIG. 14.

14a shows the illumination of the object surface (which is located in the image plane in the case shown) with a pinhole whose opening lies exactly on the optical axis 203. According to the arrangement of the pinhole on the optical axis, the pinhole image point projected onto the object also lies on the optical axis. If, as shown in FIG. 14b, the pinhole is laterally displaced to the optical axis and was at a distance A, the image point moves laterally on the surface by a distance AA. This is well known from geometric optics.

It can be seen that an overlay of the displacement of

Fig. 11 in the direction of the optical axis and the movement of 14 laterally to the optical axis will result in a movement of the image point on a straight line which has a transverse component to the optical axis, that is to say is inclined against it. The vectorial motion superposition is illustrated in Figure 15.

In order to facilitate the identification of the object area, the invention now proposes to observe an object point by means of an observation system which lies exactly on the inclined pixel straight line, which is described by the movement of the pixels through the object space. When transitioning from diaphragms to gratings, it is now important that luminous dots located at different distances from the optical axis, such as openings of a shading diaphragm provided at different distances from the optical axis result in straight lines of pixels with different inclinations in the object space with the same movement of a shading grating . This is illustrated with reference to Figure 15a. It can be shown that all these pixel straight lines converge in a single point, namely the confocal point, which lies on the focal plane of the imaging optics. The invention makes use of this knowledge by placing the pupil of the observation system in this point; observation of the object surface areas is only possible along the straight line of pixels, as illustrated by FIG. 15b, which simplifies the evaluation in the desired manner.

15c shows signal behavior in an arrangement in which a single opening is used for illumination, compare FIG. 11a. The observed light intensity I is plotted over the grating displacement path X. The strong one

Signal increase that approximates a delta function speaks, is given when the pixel falls on the object surface and this thus scatters light with a particularly high luminance. By contrast, when a plurality of lighting elements are superimposed, such as in the case of a grating diaphragm, there is a situation as in FIG. 15d, where the intensity curve I is shown on a specific pixel with the displacement X of the shading grating if the displacement X is both a component in the direction of the optical axis of the associated imaging system as well as a component lateral to it. The graph shows that although the delta function of FIG. 15c undergoes changes in intensity, there is still a significant signal increase when light is focused on the predetermined area of the surface. A simple evaluation of the object shape, taking into account the signal strength on a single array element, that is to say pixels, is still possible.

The further discussion of embodiments relates to FIGS. 1 to 10. The arrangement and the method are shown in FIG. A distinction is made between the array space and the object space. The following notation is used: The sizes and points of the array space are indicated with the letter A in the first place and the sizes and points of the object space with the letter 0. In the second place in the index, the associated lens is identified, namely in the case belonging to the illumination lens 1 with the letter B and in the case of belonging to the imaging lens 2 with the letter A. In the array space there are a line grating 3 with a grating constant p and an upstream radiation source with visible light, ie a light source 4. This light source 4 can be computer-controlled, so that the average illuminance of the distance of the respective focus plane after photometric law is adapted. The line grating 3 is assigned to the illumination lens 1 with a telecentric beam path in the array space perpendicular and extrafocal. The illumination objective 1 maps the line grating 3 into the object space, as a result of which structured illumination of the object surface 5 occurs at least at one point in time. For simplification, the two main planes of the illumination _lens 1, H _AB and H _0B , are combined in FIG. 1. In lenses of this class, the two main levels are far apart. In the array space, a receiver matrix 6 is assigned to the imaging objective 2, which also has a telecentric beam path in the array space, perpendicular to the axis and extrafocal. The imaging objective 2 images the object surface 5 in the array space. A single imaging beam A ₀ ι is shown. To simplify the two main levels of the imaging objective 2, _HÄÄ and H _0A , in Figure 1 are also merged. The optical lens 1 and the imaging lens 2 are arranged with their optical axes parallel to one another with the axis spacing d. The illumination _lens 1 and the imaging _lens 2 have the focal points F _AB and F _{ÄÄ on} the array side and the focal points F _0B and F _0A in the object space. Because of the telecentricity, the focal points F _0B and F _0A coincide with the exit _pupils PZ _0B and PZ _0A in the object space. Two illumination beams BLSoi and BLS ₀₂ and an imaging beam ABS ₀ are shown. The first linear guide of the movement system, not shown here, is rigidly connected to the receiver matrix 6 and carries a second, smaller linear guide, not shown here, which in turn carries the line grating 3. The first linear guide is connected to a high-precision length measuring system which has a highly stable zero point. The axis of movement of the first linear guide is parallel to the

Objective axes and the measuring axis of the length measuring system lies parallel to the two lens axes. The direction of movement of the second linear guide is perpendicular to the objective axes. The line grating 3 on the second linear guide is assigned a counter-grating which is firmly connected to the first linear guide and has an illumination and receiver optics in the manner of an incremental length measuring system. The evaluation electronics have an electronic interface to the computer in order to have the calculated displacement of the line grid 3 as real-time information available in the computer as phase information. At the same time, a first reference structure is applied to the line grating 3 in the part outside the image field used, which is optically scanned by a second reference structure, which is also applied to the counter grating. Both guides of the movement system start from the zero position. The direction of movement of the first linear guide is aligned parallel to the optical axis of the imaging lens. The movement takes place to the focal points hm. A position control system is assigned to the smaller, second linear guide, which carries the line grating 3, in order to be able to realize a movement of the line grating with a speed that is as constant as possible and thus also with a constant phase speed. The target values for the position of the first linear guide are calculated from the current, absolute actual phase φ _{d ter of} the Lmi- grating 3, which is derived from a zero point. This is done in such a way that the locations of the same phase or the same relative luminance on the line grating 3 move parallel to a straight line g _A , for example on the B section BS _A2 . The straight line g _A is defined such that it intersects the focal point F _{AB of} the lighting object 1 and also the main point H _{AÄ of} the imaging _lens 2. The luminous flat elements move in the array space on the B sections BS _A -,. The pictures of these B routes BS _A3 , e - finally, the B sections BS _Ax and BS _A shown in FIG. 1 are mapped into the object space. For example, the B routes BS _Ax and BS _{A2 become} images BS ₀ ι and BS ₀₂ . The images BS ₀ ι and BS ₀₂ form a cluster of lines SB _X with the convergence point K _x , which coincides with the focal point F _{0A of} the imaging _lens 2. Furthermore, the elements of the receiver array are moved on AS _Aj lines. The routes AS _Ax and AS _{A2 are shown} . Their images represent the route bundle SB ₂ with the routes AS ₀ ι and AS ₀₂ with the convergence point K ₂ in the object space, which coincides in the focus F _{0A of} the imaging _lens 2 with the convergence point K _x , the point of convergence of the convergence point Ki and of the convergence point K ₂ is generally the coincidence point K ₀ . By means of this movement regime, the planes of the object space perpendicular to the axis are “traversed” one after the other by the focus area in that, in the presence of an object surface, a stripe pattern can be observed in sharp focus from the illumination objective 1, which can be imaged onto the receiver matrix 6 by the imaging objective 2. is formed. This leads to the fact that every sufficiently small object detail, in the object space, when it is "grasped" by the focus area, generates a modulated periodic signal in the associated pixel ij on the receiver matrix 6, in which the information about the absolute phase φ _0bj i _{j of} the object point is contained.

FIG. 2 shows, for example, the signal profiles S ₀ and S _R in a pixel of the receiver matrix 6 in relation to the signal profile S _G , which can be detected on the line grating 3 with the aid of a counter grating when the grating 3 is moved. The signal curve in the pixel S _{0 of} an object point and the signal curve S _R in the pixel of a reference point are shown. Here the reference plate is closer at the focal point F _0B as the object surface. At sampling point A _PR in the area of the maximum modulation of the signal in the pixel of a reference point, the relative phase φ _{RR is} calculated and at sampling point A _P0 in the area of maximum modulation of the signal in the pixel of an object point the relative phase φ _RO bj is calculated using equation (3) the absolute phase difference Δφcitter is calculated and with equation (4) the absolute object phase φ ₀ bj, from which the Z _0B coordinate of each object point, namely Z _0bjr , is determined with equation (5). The highly stable zero point N serves as the starting point.

FIG. 3 shows a 3D recording arrangement with an illumination lens 1 and an imaging lens 2, both lenses in the array space having a beam path with central perspective on both sides and thus a small structural volume. The axes of the two lenses are inclined to each other. The latter enables a particularly high depth sensitivity to be achieved in a comparatively small measurement volume, for example for the acquisition of teeth in orthodontics. Accordingly, the entire arrangement is miniaturized. The light coming from the light source 4 illuminates a line grating 3. This is moved along a line parallel to the straight line g _AP by means of a computer-controlled slide of a linear guide (not shown here) and is projected onto the object surface 5 by the illumination _objective 1 with the pupil center PZ _0B . The straight line g _AP intersects the focal point F _{AB of} the lighting lens 1 in the array space and the main plane H _{AB of} the lighting lens 1 in the array space at the point H _ABG. The image of the straight line g _AP , the straight line g _0P , lies parallel to the axis of the Illumination lens 1 and intersects the main plane in the extension

Point H _ABG . The illumination lens 1 has the focal length f _B. The object surface 5 is imaged on a receiver matrix 6 by means of an imaging objective 2 is connected to a computer with a frame grabber, an evaluation of the recorded images being carried out by means of a computer. The imaging objective 2 is arranged in relation to the illumination objective 1 in such a way that the pupil center PZ _{0A is} located on the straight line g _0P . The pupil opening of the illumination objective 1 is made as large as possible, for example the aperture ratio is 1: 2. The depth of field in the object space is thus very limited. In contrast, the pupil opening of the imaging objective 2 is made as small as possible, for example the aperture ratio is 1: 22. The depth of field in the object space is thus comparatively large. This is necessary because the receiver matrix 6 is fixed. Regarding the measuring sequence: The line grating 3 is moved and 6 images are taken with the receiver matπx, for example 32. The movement of the grating 3 with the grating constant p is measured with high precision to about 1% of the grating constant p. Images are recorded, the phase change between two images generally being less than 2π, for example 3 / 2π. The position of the receiver matrix 6, whereby a CMOS camera can be used due to the large dynamic range and the possibility of evaluating individual pixels, is selected so that the entire object area of interest can be grasped sharply. The receiver matrix 6 can also be rotated according to the Scheimpflug condition so that it contains the two points A _Ax and A _A2 . Taking into account the geometry of the SD recording arrangement and the imaging conditions, the 3D point cloud of the 3D measurement object is determined from the 32 images recorded. For this purpose, the absolute object phase φ _{0b3 is} determined for each object point. From this the z ₀ b ₃ ~ is calculated

Coordinate for every single point. The x ₀ b ₃ ^_ are determined via the imaging _scale as a function of the z _0b] ^_ coordinate and the y ₀ b ₃ coordinates are calculated from the known pixel pitch of the receiver matrix 6.

FIG. 4 shows a 3D recording arrangement with an illumination objective 1 with a comparatively large pupil opening, for example with an aperture ratio of 1: 2.8 and a central perspective beam path optimized for the oblique image. The illuminated field is asymmetrical to the axis of the illumination lens 1. Lenses in the middle focal length range are preferably used here, for example with focal lengths around 25 mm. The imaging lens 2, for example with an aperture ratio of 1: 2.8, has a telecentric beam path in the array space. The axes of the two lenses are arranged parallel to each other. The main planes of the illumination lens 1 and the imaging lens 2 coincide at least approximately in the array space. The light emanating from the light source 4 illuminates a reflection grating, for example an electronically controllable digital micromirror device 61. Furthermore, the electronically controllable digital micromirror device 61 and the receiver matrix 6 are rigidly connected to one another in the at least approximately the same level in the array space. Both components are located here on a computer-controlled slide of a linear guide. The direction of displacement of the slide of this linear guide is aligned parallel to the axes of the two objectives, so that both components move synchronously and parallel to the optical axis of the illumination objective 1. Due to the electronic control of the Digital Micromirror Device 61, it is not only possible to move a glowing lamp laterally

Flat element 3A are made, but also the displacement of a light curtain can be realized according to a program specification. So for a single Luminous flat element 3A on the direct mirror device 61, as well as for the luminescent elements with precise coordination with the displacement by the computer-controlled slide, a movement can be realized on a path BS _A3 parallel to the straight line g _AP . The line g _Ap intersects the focal point F _{AB of} the illumination lens 1 in the array space and the main plane of the imaging lens 2 in the array space in the main point HΛΛ. The image of the straight line g _AP in the object space, the straight line g _0P , lies in the axis of the imaging objective 2. For example, the luminous flat element 3A is shifted as a luminous flat element of constant relative luminance, from point B _Ax to point B ₂ . The luminous flat element 3A can represent a maximum in an at least approximate cos ² light distribution, that is to say have a phase value of 0 when a signal is detected. The luminance distribution on the digital micromirror device 61 is projected onto the object surface 5 by the illumination lens 1. The straight lines and lines parallel to the straight line g _{AP are} imaged by the illumination lens 1, the images of the displacement lines BS _{A3 of} the luminous flat elements, that is to say here the luminous micromirror of the digital micromirror device 61, forming a line cluster in the object space. For example, a luminous flat element 3A moves on the displacement path BS _A3 . The image of this displacement route BS _A3 , route BS ₀₃ , is aimed at the convergence point Ki in the object space. The object surface 5 is imaged on a receiver matrix 6 by means of an imaging lens 2. The computer-controlled slide of a linear guide, not shown here, also carries the receiver matrix 6 and so the pixels of the receiver matrix 6 each experience a shift on a distance AS _A3 parallel to the optical axis of the imaging objective 2. The pictures of the distances AS _Ä3 , the routes AS _θ3 , form a _{path bundle} in the object space with a convergence point K ₂ in the focal point of the imaging _lens , so that in a telecentric imaging lens the convergence points Ki and K ₂ , the focal point F _0A , and the pupil center PZ _0A coincide. Thus, an object point 0 in the object space is tracked by an image of a luminous surface element 3A in the depth of the object space during the displacement. FIG. 4 shows in the object space the current coincidence point K _θ3 i in position A ₀ ιB ₀ ι of the image of the luminous surface element 3A and the image of the element of the receiver array at point A _A ι, the image A ₀ ι, at the time t _x . This element of the receiver array is moved on the AS _A3 route. A deposit of an element 0 of the object surface 5 from the current coincidence point A ₀ ιB ₀ ι, in the direction of the imaging beam belonging to the element of the receiver array to the time range of the recording Δt _x leads to a signal with the phase _r s φ _rS = 2π x Δx _A / p, where p is the period of an at least approximate cos ² distribution and Δx _{A is} the image of Δx ₀ according to FIG. 4. If object point 0 coincides with the current coincidence point Koj ι + ι _<, for example in position A ₀₂ B _o2 at time t _{x +} ι, a signal with phase φ _r s = 0 is registered. The current point of coincidence is shifted further through the object space. In this way, a periodic signal can be registered, which is dampened by the increasing blurring of the image. The width of the modulation curve of the periodic signal results from the ratio of the distance d of the optical axes to the diameter of the aperture diaphragm D _{B of} the illumination lens. In order to be able to uniquely identify a strip of the zero order, this ratio must generally not exceed the value 10, the meaningful ratio being significantly determined by the evaluation algorithm used for the signal course. FIG. 5 shows a 3D recording arrangement for objects that are illuminated with structured light. The aim here is to determine the point cloud of the latter particularly quickly and precisely by determining the absolute phase of elements of the object surface. For this purpose, a light source 4 illuminates a concentrically designed grating 81, as a result of which a structured, luminous array is formed, which is located in a sector of a pane 83 on a flat surface which belongs to a transparent plane-parallel plate 84. On the disk 83 there are several transparent segments with transparent plane parallel plates 84 in the entire area. The concentrically formed, illuminated grating 81 is imaged with the illumination objective 1. The disk 83 is driven by a motor 85. The speed is, for example, 24 rpm. There are also reference marks for recognizing the center of a segment, which are used for reading out the receiver matrix 6. Furthermore, the information about the axial position of the disk 83 can be obtained in a flat area on the upper side of the disk 83 in the immediate vicinity of the concentrically formed grid 81. The information about the radial position of the disk 83 obtained in the optical sensor head 87 is fed to a piezo actuator 92 in a control circuit (not shown here) with an electronically controllable voltage source for the piezo actuator 92 as a control voltage. The piezo actuator 92 is connected to the receiver matrix 6, so that here the radial position, rather than the distance between the concentrically formed grating 81 and the receiver matrix 6, is not really kept constant. In this way, a defined position between a luminous surface element and an element of the receiver matrix 6 is maintained. The axial position of the associated structured illuminated area is determined by a second piezo actuator 93 in a control loop is kept in a constant position by holding the disk 83 in the desired position as it rotates. With precision bearings, the elimination of the axial runout can usually be dispensed with. Furthermore, a plane-parallel transparent compensation plate 94 for axially compensating the position of the receiver matrix 6 is arranged in the imaging beam path after the imaging objective 2. In this way it is achieved that at each angular position of the disk 83, that is to say in each segment assigned to the light source and the receiver matrix 6, the optical paths are the same, ie the object distance for the structured illuminated array and the image width for the receiver matrix 6 are the same. In this way, a sharply imaged strip pattern can be observed in the same plane in the object space on the object surface and this can also be imaged sharply on the receiver matrix 6 via the imaging objective 1 and a transparent plane-parallel plate 95 of a certain optical thickness, since the sharp plane of the Imaging beam path through the adaptation of the transparent plane parallel plate 95 lies in the same plane. As a result of the rotation of the disk 83, transparent segments with different geometrical-optical thicknesses enter both the illumination and the imaging beam path. The geometrical-optical thicknesses are always coordinated so that the sharp planes of the two beam paths coincide in the object space. The position of the focus planes in the object space changes from segment to segment, so that the entire object space is gradually "deeply focused" in depth. The position of the concentric grating changes in the radial direction from segment to segment. In this way, the phase is changed step by step every time a segment is changed, for example in steps of 90 °. The evaluation of a synchronization pulse in the edge area ensures that only then an image of the receiver matrix is detected when the plane-parallel plate of defined thickness with the concentrically formed grating 81 is located in full area in front of the receiver area. A periodic signal curve can thus be obtained in each element of the receiver matrix 6, each signal value of the recorded signal curve being generated in another sector by means of another plane parallel plate. The phase difference from signal value to signal value is known with high precision due to the optically read reference grid.

FIG. 6 shows a 3D recording arrangement which is particularly suitable for mobile use, for example for the computer-aided 3D orientation of robots in the vicinity for gripping tasks. Furthermore, 3D hand-held devices for detecting objects in the close range up to 1 m can be based on this arrangement. This 3D recording arrangement consists of an illumination lens 1 and an imaging lens 2, both lenses having a central perspective beam path in the array space, and a transparent profile grating 53, which can be viewed as a structured 3D array and by a light source 4 is illuminated. This can represent a flash light source with a flash frequency in video clock. The transparent profile grid 53 has a plurality of ramps 54 and 56. As shown in FIG. 8, there is a Bmar code pattern, a bitmap, on the inclined ramp surfaces 55 of the ramp 54, the directions of the ramp surfaces or at least the associated best-fit lines in the main section always pointing to the pupil center PZ of the imaging lens 2 . The binary code pattern is mapped into the object space at different depths. Another possibility for the formation of a ramp area provides

Figure 7 represents. The ramp surface 57 has several stages. There is a different binary code pattern on each level. The designations Ai, A ₂ , A ₃ and A _{4 are} assigned to the individual patterns. The compensating line through the ramp area, the straight line AG _A] points to the pupil center PZ of the imaging lens 2. The ramp area 55 on the transparent profile grating 53 on the ramp 54, the main section of which can be represented by a compensation line AG _A3 , is represented by the lighting lens 1 mapped in the object space. The illumination objective 1 should have a large relative aperture, for example 1: 1.2. Furthermore, flat areas of the object surface are shown simultaneously in the pictures Aχo, A ₂₀ , A3 ₀ and A ₄₀ for illustration in possible positions. A luminous flat element 3A is also depicted and represents the image B _FEL . The sharp volume SV _{FEL of} the image B _FEL is shown. It is not very deep due to the large relative opening ratio. The imaging lens, on the other hand, is heavily dimmed. In this way, the elements of the receiver array are mapped with a much larger depth range. The image of the entire ramp area is thus in the sharp volume SV _{EE of} the associated element of the receiver matrix 6. With a large relative opening of the illumination lens 1, the image of a pattern disappears very quickly when the object moves in the depth. In several planes parallel to the main section, ramp surfaces 54 can be arranged on the transparent profile grid 53, each with a different mean object width, so that elements of an object surface 5 can be detected at different depths in the object space and the depth detection area is particularly large.

FIG. 9 shows an application for the extraction of digital point clouds, which are used for 3D television. this can be the case, for example, at sporting events where there is good lighting. The arrangement according to FIG. 9 can also be used for 3D recordings in the amateur video field and in photography. A sequence of images is recorded with the two receiver matrices 6 and 14 in order to generate a single 3D image. The displacements of the receiver matrix 6 and the receiver matrix 14 are carried out in such a way that the focus area changes from image to image until the object space is completely covered in depth. The lighting is done using natural light or artificial light, which usually has no spatial structure. A first imaging lens 33 and a second imaging lens 2 are assigned to each of the two receiver matrices 6 and 14. The first imaging lens 33 and the second imaging lens 2 are telecentric in the array space, as a result of which the focal point F ₀ ι and the pupil center PZ ₀ ι coincide in the imaging lens 33. In the imaging lens 2, the focus F ₀₂ and the pupil center PZ ₀₂ also coincide due to the tele-center. The images of the straight lines parallel to the straight line g _A ι _P and thus also the displacement distances of the elements of the receiver matrix 6 on these straight lines are imaged by the imaging lens 33 in the object space and there set up a cluster of lines with the convergence point K _{2i of} the lines, respectively Tufts of straight lines with the cut in K _2i , which also contains the straight line g ₀ ιp. In addition, the images of the straight lines parallel to the straight line g _A2P and thus also the displacement _{distances of} the elements of the receiver matrix 14 on these straight lines are imaged by the imaging _objective 2, as a result of which, in the object space, a tufts of lines with the convergence point K _{22 of} the lines and a tufts of lines with the section arise in K ₂₂ . This _tufts of straight lines contains the straight line g _02P , which coincides with the straight line g ₀ ι _P. The displacements of the receiver matrix 6 and the receiver matrix 14 are carried out in such a way that the stretching tufts with the convergence point K _2i and the stretching _tufts with the convergence point K ₂₂ coincide at the point of coincidence K ₀ . As a result, there is exactly one corresponding element on the receiver matrix 14 for each element of the receiver matrix 6. The two receiver matrices 6 and 14 are each equipped with at least one powerful computer, which in turn is integrated in a high-performance communication network, via an interface connected. The receiver matrices 6 and 14 are connected to a precise linear guide 20, which in turn is connected to a computer-controlled, highly dynamic linear motor 21. The two receiver arrays 6 and 14 are moved in the z _A -Rιchtung in the direction of the two foci F _A and F ι _A2, the position of the common plane Sharp SCH i-1 whereby changes in the object space. Objects 5, 18 and 19 are adequately illuminated by a light source 15. First, the object 5 at point P _{3 x} , which is located on the straight line g ₀₃ ι z _P , which represents the location of the competing straight lines g ₀₃ ι _P and g _{0 3 2Pp} , from the common plane of focus

SCH l detected. Then, for example just a few milliseconds later, the object surface is recorded

18, for example through the point P _{k x +} ι through the common plane of sharpness SCH 1 + 1, with an object point of the object surface 18 at this point in time on the straight line g ₀ k I 2 p which defines the location of the compensating straight line g _{0 k 1 P} and g represents ₀ 2 _P. A few milliseconds later, the first points on the object surface are recorded

19. The stroke frequency of the movement of the two receiver matrices 6 and 14 takes place, for example, at 24 Hz. The evaluation of the images taken by the two receiver matrices 6 and 14 takes place in such a way that signals are formed from elements which are parallel to each other the straight line g _A ι _P and g _A2P located straight lines, for example on the straight lines gAjiPr gAkip, gAj ₂ p or g _A k ₂ p at the same time in at least approximately the same location, that is to say represent corresponding elements. As already described, the z ₀ coordinates are calculated from the detected signals S _Xj and S _2j of a pair of corresponding elements using correlation methods, and the entire point cloud of the object surface or the scene is derived from these by knowing the geometry of the arrangement and the imaging scales the z ₀ coordinate. In the case of moving objects, point clouds are provided in the video cycle. The information about the color of the respective object point results, for example, from the receiver element which is closest to the correlation maximum. The surface models of the objects and the scene are calculated from the calculated point clouds. This requires a very high computing speed, which can, however, be achieved on the basis of special processors. It is possible to assign a separate processor to each element of the receiver matrices 6 and 14. This is very advantageous for the execution of the cross-correlation already described. Data reduction techniques can be used to transmit the calculated data for 3D playback.

FIG. 10 shows a 3D recording arrangement that can be used, for example, as a multi-level obstacle sensor in daylight. Two imaging beam paths with two imaging lenses 33 and 2 are arranged, the imaging lenses 33 and 2 each being designed to be telecentric in the array space. This is not absolutely necessary, but it has constructive advantages. Each imaging objective 33 and 2 is a spatially structured receiver array

106 and 114 assigned. These each have two receiver areas

107 and 108 as well as 109 and 110 and are perpendicular to Main cut arranged. In addition, these receiver surfaces 107 and 108 and 109 and 110 are parallel to a straight line g _A ι _P or parallel to a straight line g _A2P . These straight lines g _AiP and g _A2P each intersect the focal point F _ÄΛI and F _{AA2 of} the associated lens and the point P _A in the common main plane of the two imaging lenses 33 and 2. The images of the receiver surfaces 107 and 108 and 109 and 110 become after the Scheimpflug condition mapped in the object space, the coincidence of the two images of the receiver surfaces 107 and 109 and 108 and 110 of the spatially structured receiver array 106 and 114 being given. From an illuminated object surface 5, elements are imaged on the four receiver surfaces 107 and 109 as well as 108 and 110. For example, the points Oi and 0 _{2 are} imaged on the object surface. The correlation method already described in detail can be used for the determination of the z ₀ position of the points O _x and 0 ₂ using the gray value or color distributions from the environment when using color cameras as receiver matrices. This results in four signal curves Si and S ₂ as well as S ₃ and S ₄ each in the line of the associated receiver matrix in the main section of the line-by-line reading, for example in the A _A or B _A position is started. Here, for a point in the object space on the two images of the receiver surfaces 107 and 108 and 109 and 110 which coincide in the object space, the associated, fixed elements of two receiver surfaces 107 and 108 or 108 and 110 each form a corresponding pair.

Claims

claims

1. Device for three-dimensional object detection with at least two imaging systems which have imaging optics facing the object, at least one being designed as an observation system for object observation and at least one having an elementary means which is movable in front of the imaging optics and whose elementary image moves through the object space on a pixel line , characterized in that the elementary means is designed as an elementary means movable with a lateral component to the optical axis of the imaging optics and the observation system is arranged for observation along the pixel line.

2. Device according to the preceding claim, characterized in that two imaging systems represent both observation systems.

3. Device according to the preceding claim, characterized in that the observation systems comprise detection arrays, in particular CCD arrays.

4. Device according to the preceding claim, characterized in that the detector arrays are moved synchronously.

5. Device according to the preceding claim, characterized in that the detector arrays are arranged on piezo setting means, in particular for synchronous movement with a change in distance.

6. Device according to the preceding claim, characterized in that an evaluation unit is provided which is connected to the two detector arrays for signal evaluation and is arranged for evaluating the signal behavior to signal an object surface point detection when on an array area of the first imaging system a predetermined signal behavior and also a predetermined signal behavior is detected on an assigned array area of the second imaging system.

7. Device according to one of the preceding claims, characterized in that one of the imaging systems is a lighting system.

8. Device according to the preceding claim, characterized in that the lighting system for illuminating the object comprises a plurality of lighting elements which realize the movable elementary means.

9. Device according to one of the preceding claims, characterized in that an illuminant excitation circuit is provided in order to simulate the elementary means movement at least partially as a virtual movement by successively a predetermined excitation pattern of the light elements.

10. Device according to one of the preceding claims, characterized in that the lighting system comprises a movable grid, in particular geometric shading grid.

11. Device according to one of claims 7 to 10, characterized in that the observation system comprises a plurality of observation elements.

12. The device according to the preceding claim, characterized in that the observation elements are realized by areas, in particular elements of a detector array, in particular a CCD array.

13. Device according to the preceding claim, characterized in that the detector output is assigned an evaluation unit which is designed to signal a surface detection when a certain signal pattern, in particular a signal maximum, in particular an absolute signal maximum, is detected on an observation element of the array. is taken.

14. Device according to the preceding claim, characterized in that the evaluation unit is designed to determine the coordinates of the detected object surface area as a function of the position of the movable element means.

15. Device according to one of the preceding claims, characterized in that the pupil of the observation system is arranged at least approximately m of the focal plane of the imaging optics of the illumination system.

16.Method for 3D recording of an object surface in a scene with an electromagnetic radiation source, this being designed by means of at least one structured array as an em structured luminous array and emitting a flat element and an illumination beam path with an illumination objective assigned to the structured illuminating array, which is a Realization of the illuminating flat element, and an imaging beam path for the imaging of elements of the object surface and receiver array with at least two elements and with an imaging object assigned to the receiver array and elements of the receiver array in the recording process radiation from elements of the illuminated object surface Detect in the object space and images of elements of the receiver array are always also formed in the object space with a geometrical-optical focus volume by the imaging lens and by imaging the luminous area element with the illumination lens em image of the same in the object space with a geometrically optical focus volume and the detection of radiation from the elements of the object surface by elements of the receiver array takes place in a time range (Δt _B ) in which at least one luminous surface element of the structured illuminated arrays, a predetermined shift is performed, including a predetermined optical shift as a result of a geometric-optical path change tion, so as to emit a luminous surface element at different points in time at different locations, characterized in that the sharpness volume of an image of the luminous surface element is formed in the object space, and the sharpness volume of an image of the receiver array in the object space and an element of the object surface are brought at least approximately to a coincidence by a predetermined displacement of the luminous surface element with a displacement component parallel to the optical axis of the illumination objective, wherein when the coincidence occurs, the element of the receiver array involved in this coincidence is exposed to a radiation that changes over time.

17. A method for 3D recording of at least one object surface in at least one scene according to claim 16, characterized in that in the time intervals (Δti) of the detection of radiation, the luminous surface elements are shifted relative to the illumination lens on their own displacement path and luminous surface elements at least at one time

(t _x ) within a time interval (Δti) in a luminance distribution each have an at least approximately predetermined constant luminance and the luminous surface elements are always positioned on a B-section (BS _Aj ), whereby the B-sections ( BS _Aj ) represent the target locations for the luminous surface elements at least at one point in time (ti) within the time interval (Δti), and the images of these B sections (BS _Aj ) in the object space always by imaging with an illumination lens (1) a cluster of lines (SB _X ) with a convergence point (Ki) are formed, the convergence point (Ki) at least at a distance from the optical axis of the illumination objective (1) from the last part and at a maximum distance from the distance of the pupil center ( PZ _0B ) of the illumination _lens (1) from the pupil center (PZ _0A ) of the imaging _lens (2) is positioned in the object space and at least in one time range (Δt _B ) during the displacement process of the luminous surface elements, exactly one image of a receiver element and exactly one image from a luminous surface element in the object space at least approximately at a single point in time (ti) within each time interval (Δti) of the detection, at least approximately jointly positioned on the image of a B section (BS _A3 ) and thus at least at this point in time (ti) from the image of a receiver element and the image of a luminous surface element in each case If a pair with a fixed assignment is formed and such pairs are generated in the object space and these pairs are pushed through the object space, the sharpness volumes of images of the luminous surface elements coincide with surface elements of the object surface (5) at least once in the displacement process within a time interval (Δti) and the elements of the receiver array in the time interval (Δti) of the coincidence detect a signal curve with at least one relative extremum of the signal size, the time range (Δt _B ) being made larger than the time interval (Δti) and thus at least one time interval (Δti) i Time range (Δt _B ) is adjusted in time and the positions of the luminous surface elements during the shift of the structured luminous array and the positions of the elements of the receiver array always from the position of the illumination lens (1) and the position of the imaging lens (2) in the 3D recording arrangement and the focal length (f _B ) of the illumination lens (1 ) and the focal length (f _A ) of the imaging lens (2) can be determined and implemented in such a way that in the object space both the luminous surface elements of the structured luminous array and the elements of the receiver array are at least approximately in one part of the object space in the same Level.

18. A method for 3D recording of at least one object surface (5) in at least one scene according to at least one of claims 16 and 17, characterized in that in the time intervals (Δti) of the detection of radiation in each case a luminous surface element on a B- Distance is positioned and the B-sections are directed to the pupil center of the imaging lens in the array space, so that the convergence point is positioned at least approximately in the pupil center of the imaging lens, and during the displacement process an image of a receiver element and an image of one luminous surface element in the object space are positioned at least approximately jointly on the image of a B section and a pair with a fixed assignment is formed from the image of a receiver element and the image of a luminous surface element in the object space and pairs of images are thus generated in the object space and true at the end of the shifting process of the luminous surface elements, an image of a receiver element and in each case an image of a luminous surface element in the object space is brought at least approximately to the coincidence.

19. A method for 3D recording of at least one object surface (5) in at least one scene according to at least one of claims 16 and 17, characterized in that the convergence point (Ki) at least approximately in the focal plane of the illumination lens (1) in the object space and additionally in the pupil center (PZ _0A ) the pupil of an imaging lens (2) is positioned in the object space and during the shifting process an image of a receiver element and an image of a luminous surface element (3A) in the object space at least at one point in time (t _x ) within each

The time interval (Δt _x ) of the detection can be positioned at least approximately jointly on the image of a B section (BS _A3 ) and thus at least at this time (t _x ) from the image of a receiver element and the image of a luminous surface element (3A) in each case A pair with a fixed assignment is formed in the object space and such pairs with a fixed assignment are generated in the object space and the B segments (BS _A3 ) are positioned parallel to a straight line (g _AP ), the straight line (g _AP ) being the focal point F _AB of the illumination lens (1) in the array space and the increase with the amount from the quotient “distance of the pupil center (PZ _0A ) of the pupil of the imaging lens (2) in the object space from the axis of the illumination lens (1) and the focal length (f _B ) of the illumination lens (1) ", this rise in the straight line (g _AP ) being related to the axis of the illumination lens (1).

0. A method for 3D recording of at least one object surface (5) in at least one scene according to one of claims 16, 17 and 19, characterized in that at least approximately a straight relative displacement of the receiver array to the imaging lens (2) in parallel is carried out to the optical axis of the imaging lens (2) and signal values are read out from one receiver element several times in succession during the shift, and a signal curve is thus formed by means of a receiver element and when imaging displacement paths of elements of the receiver array generated multiple times with the imaging lens (2) from whose images in the object space at least approximately at least a cluster of lines (SB ₂ ) with a convergence point (K ₂ ) in the focal point (F _0A ) of the imaging _lens (2) is formed.

21. A method for 3D recording of at least one object surface (5) in at least one scene according to at least one of claims 16, 17, 19 and 20, characterized in that the convergence point (Ki) of the cluster of lines (SBi) together with the The point of convergence (K ₂ ) of the cluster of lines (SB ₂ ) in the object space with the focal point (F _0A ) as well as with the pupil center (PZ _0A ) of the pupil of the imaging objective (2) is brought at least approximately to the coincidence, the illumination objective (1) and the imaging objective (2) is at least approximately telecentric on the array side, and luminous surface elements (3A) are shifted at least approximately parallel to a straight line (g _A ) on lines and the straight line (g _A ) the focal point (F _AB ) of the illumination lens (1) penetrates in the array space and for the straight line (g _A ) the increase with the amount from the quotient “focal length of the illumination lens (1) and distance (d) from the focal point (F _M ) of the illustration - Realization objective (2) from the axis of the illumination lens (1) in the object space "is realized, this increase in the straight line (g _A ) being based on a straight line perpendicular to the axis of the illumination lens (1).

22. A method for 3D recording of at least one object surface (5) in at least one scene according to at least one of claims 16 to 21, characterized in that the position of at least one luminous surface element is made stationary and at least components of the illumination lens are moved.

23. Method for 3D recording of at least one object surface in at least one scene with at least one electromagnetic radiation source, this being designed by means of at least one structured array as an at least em structured luminous array with at least two flat elements and at least one flat element shining, each luminous surface element is defined by an at least approximately predetermined luminance and by a predetermined location, and with at least one illumination beam path with at least one illumination lens assigned to the structured, luminous array, which realizes an imaging of the luminous surface elements and with at least one imaging beam path for the imaging of elements the at least one object surface and at least one receiver Array with at least two elements and an imaging lens assigned to the receiver array and elements of the receiver array in the recording process, detecting radiation from elements of the at least one illuminated object surface in the object space and from elements of the receiver array through the imaging lens always also images in the object space, which corresponds to the scene space, is formed with a geometrical-optical focus volume and elements of the at least one object surface are imaged with at least one imaging lens and, by imaging the at least one luminous surface element with the illumination objective, an image of a luminous surface element in the object space with a geometrically optical focus volume is formed, characterized in that luminous surface elements in the time intervals (Δt _x ) of the detection of radiation in a time range (Δt _B ) at least approximately at a separate location in the structure The luminous array is arranged relative to the lighting lens and can be illuminated by actuation and imaged by the lighting lens and the at least one luminous surface element is always mapped to a predetermined location in the object space at least at one point in time (t _x ) within the time interval (Δt _x ) and this image location of the at least one luminous surface element in the object space is changed by a control by actuating and illuminating a different, predetermined surface element, so that the image of each luminous surface element on a controllable trajectory, structured from spacing increments of the spacing images of the luminous surface elements in the array Space through which object space is pushed - in the sense of the controlled assumption of predetermined, different positions - and in each position after the displacement by an integer multiple of the distance increment, at least one signal value is detected with a receiver element and thus from several processes of detection and reading a signal curve is formed from elements of the receiver array and the location of the detected and read element of the receiver array is continuously changed and the locations of the detected and read elements of the receiver array are located at locations in the receiver array and the image thereof Location with which the predetermined image location of the luminous surface element in the object space is at least approximately optically conjugated, and thus an image of a detected and read element of the receiver array with the image of a luminous surface element at least at one point in time (t _x ) lb of the time interval (Δt _x ) in the object space is brought to coincidence and so a pair of images with alternating images are generated, and these pairs gradually assume different positions in the object space and thus gradually through such pairs the object space in the depth is penetrated, the sharpness volume of the image of one luminous surface element with one surface element of the object surface coinciding at least once in the time range (Δt _B ) in a time interval (Δt _B ) and the detected and read elements of the receiver array in the time interval ( Δt _x ) of the coincidence have a signal curve with at least one relative extremum of the signal size, the time range (Δt _B ) being greater than the time interval (Δt _x ) is made and so at least one time interval (Δti) is temporally fitted into the time range (Δt _B ).

24. Method for 3D recording with at least one object surface (5) in at least one scene with a first and at least one second imaging beam path, two imaging objectives (2, 33) between the two axes for imaging the object surfaces (5, 18, 19) a line of symmetry is formed, and with at least one receiver array assigned to the imaging lens in each imaging beam path, the two receiver arrays each having elements that emit radiation from the elements of the illuminated object surfaces in the time domain (Δt _B ) during the recording process Detect in the object space, characterized in that the two receiver arrays each experience a shift to another location in the acquisition process in the time domain (Δt _B ), and at least approximately the detection of radiation from the elements of the object surfaces by the elements of the receiver array takes place simultaneously and the elements of the receiver array ans are then read out and signal values are obtained in each case and during the recording process the two receiver arrays are simultaneously shifted on displacement paths and the images of the displacement paths in the object space are at least approximately positioned on the line of symmetry between the two axes of the objectives (2, 33) and out of the cluster of paths a convergence point (K _2i ) is formed from the images of the distances of the individual elements of the first receiver array and from the bundle of lines of images of the distances of the individual elements of the second receiver Arrays a convergence point (K ₂₂ ) is formed and the convergence point (K _2l ) and the convergence point (K ₂₂ ) on the line of symmetry are brought at least approximately to the coincidence and form a convergence point (K _o ) and thus shift the two receiver arrays are that their images at least partially coincide in the object space, so that the images of the elements of the first receiver array and the images of the elements of the second receiver array in the object space are brought at least approximately in coincidence in pairs, the pair-forming elements of the two Receiver arrays each represent corresponding elements and signal curves (Si) of the first receiver array are formed by reading out the elements during the displacement of the first receiver array and the displacement of the first receiver array parallel to a straight line (g _A ip) is carried out and so the elements of the first receiver array at least approximately parallel to a straight line (g _A ip) are shifted and signal profiles (S ₂ ) of the second receiver array by reading out the

Elements are formed during the displacement of the second receiver array and the displacement of the second receiver array is carried out parallel to a straight line (g _A2P ) and so the elements of the second receiver array are at least approximately parallel to a straight line

(gA ₂ p) and the straight line (g _A ip) at least approximately at one point (P _A ι) on the symmetry line in the main plane of the first imaging lens (2) in the array space and the straight line (g _A2 p) at least approximately at a point (P _A2 ) on the symmetry line in the main plane of the second imaging lens (33) are cut, the straight line (g _A ip) additionally being the focal point (F _A ι) of the first imaging lens (33) and the straight line (g _A2P ) _contain the focal point (F _A2 ) of the imaging _lens (2) in the array space, and from the two signal profiles (Sι ₃ ) and (S ₂₃ ) of two corresponding elements and the receiver arrays , which are stored in the memory of a computer by way of the displacement of the two receiver arrays, by the synchronous displacement of a window function, this window function having at least one window with a minimum window length corresponding to two signal values and a maximum window length which is at least approximately the same Long of the waveforms (Si-,) and (S ₃ ) corresponds to at least one signal value, which corresponds to an increment of the displacement of the receiver arrays, over each of these two waveforms (Sι ₃ ) and (S ₂₃ ) and from each current window in the position k, with 1 <k <m, a signal piece (Si part position k ₃ ) and (S ₂ part position k ₃ ) is formed, these successively formed S Signal pieces (Si part] position k ₃ ) and (S ₂ part) position k ₃ ) overlap in each of the two signal curves (S _i3 ) and (S ₂₃ ) in a partial area, with each of the two signal curves (Sι ₃ ) and (S ₂ -,) is started with the shifting of the window function in both signal pieces, and each starting from two signal pieces in position 1 (S _x part position ι ₃ ) and

(S ₂ part position ι ₃ ) the cross-correlation function is calculated, whereby however one of the two signal pieces is inverted beforehand, i.e. all values of the same are mirrored, and thus from an original signal piece (Si τeιi position 1 ₃ ) and from an inverted one Signal piece (S _{2 T} e _X ι position 1 _{INV 3} ) the maximum of the cross-correlation function (MCC i2 ₃ position 1) between 0 and 1 is calculated and saved and then after the Shifting the window function, the maximum of the cross correlation function is calculated in the manner described (MCC i _{2 3 pos} i _t i _{on 2)} in the 2-position for the next two signal components up to the window function at the other end of the two waveforms (Si _j, S ₂ j ) has arrived at position m and the maximum (MCC 12 ₃ position m) of the cross-correlation function (MCC 1 ₂ position m) is again determined and a maximum value curve is formed from m calculated maxima (MCC _m ), in which Maximum value curve, in turn, the maximum (M _{m 3} ) is determined and the location of the maximum (M _m -) of the maximum value curve is assigned to the two original signal curves and thus the path of the displacement of the two receiver arrays and, if one is used Window function with only one window in only one position, the location of the maximum (M _{1 3} ) of the calculated cross-correlation function (MCC 1 _{2 3} position 1), the two original signal profiles and thus d The path of the displacement of the two receiver arrays is assigned and the location of the maximum determined in this way (M ₃ ) as the location of the corresponding element and associated image of the respective element of the object surface (5, 18, 5) in the array space is defined and from the location of this maximum (M ₃ ) in the array space the z ₀ coordinate of the respective element of the object surface is calculated and also the x ₀ and y ₀ position of the respective element of an object surface (5, 18, 5 ) is calculated from the known geometry of the 3D recording arrangement, and thus the positions of the elements of an object surface (5, 18, 19), to which signal profiles can be assigned, are calculated, the geometry of the 3D recording arrangement being known and the displacements of the Receiver arrays are predetermined.

5. A method for 3D recording of at least one object surface (5) in at least one scene according to claim 24, characterized in that the axis of a first imaging lens (2) for imaging the object surfaces (5, 18, 19) parallel to the axis of a second imaging lens (33) for imaging the object surfaces (5, 18, 19) and the imaging lenses (2, 33) are made at least approximately identical and the main plane of the first imaging lens (33) in the array space and the main plane of the second imaging lens (2) coincide at least approximately in a common plane and the receiver arrays are at least approximately jointly in one plane and the two points P _Ai and P _A2 are brought to coincidence at least approximately in one point P _A.

26. A method for 3D recording of at least one object surface (5) illuminated with a radiation source (15) in at least one scene with a first and at least one second imaging beam path, the axis of a first imaging objective (2) in the imaging beam path for imaging the object surface - Chen (5, 18, 19) parallel to the axis of a second imaging lens (33) in the imaging beam path for imaging the object surfaces (5, 18, 19) is aligned, and so between the two axes of the two imaging lenses (2, 33) Symmetry line is formed, and with at least one receiver array assigned to the imaging lens in each imaging beam path and the two receiver arrays each have elements. sen, which detect electromagnetic radiation from the elements of the illuminated object surfaces in the object space in the recording process in the time domain (Δt _B ), characterized in that the two receiver arrays in the recording process in the time domain (Δt _B ) each have an electronically controlled, mechanical shift to another Experience location and at least approximately the detection of electromagnetic radiation from the elements of the object surfaces for the duration of a time interval (Δt _x ) is carried out simultaneously by the elements of the receiver array and the elements of the receiver array are subsequently read out and signal values are obtained in each case and during the recording process, the two receiver arrays are displaced simultaneously and parallel to the respective optical axes of the parallel, at least approximately identical imaging beam paths whose main planes coincide, the object surfaces in the scene being illuminated, and de r signal curve (Sι _z ) is formed by reading out laterally adjacent elements of the first receiver array during the displacement of the first receiver array in the direction of the optical axes in such a way that exactly the elements of the receiver array are used for signal formation which lie on lines which are aligned parallel to a straight line (g _A ip) and which at least approximately intersect the point (P _A ) in the common main plane of the imaging lenses (2, 33), and so the signal curve formed thereby at least approximately corresponds to a real shift parallel to a straight line (g _A ι _P ), and thus a current one at least at a time t _x in a time interval Δt _x Coincidence point is formed, which is formed successively at different predetermined locations in the object space in the time range Δt _B , and the signal curve (S _2z ) by reading out elements of the second receiver array lying laterally next to one another during the displacement of the second receiver array in The direction of the optical axes is formed in such a way that exactly those elements of the receiver array are used for signal formation that lie on lines that are parallel to a straight line (g _A2P ) and that point (P _A ) in intersects at least approximately the common main plane of the imaging lenses (2, 33), and so the signal _{curve formed} at least approximately corresponds to the signal _curve (S ₂ ) that _arises during a real shift parallel to a straight line (g _A2P ) and the straight line (g _A ι _P ) and the straight line (g _A2P ) are brought to the intersection at one point (P _A ) and in addition the straight line (g _AlP ) the focal point (F _Al) point of the first imaging lens (33) and the straight line (g _A2P) the fuel (F _A2) containing the imaging lens (2) in the array region and from the two signal courses (S _z ι ₃₎ and (S _z23) of two changing corresponding elements and the receiver arrays, which correspond only at a point in time t _x , the signal _profiles (S _z ι ₃ ) and (S _z23 ) being stored in the memory of a computer by way of the displacement of the two receiver arrays, and in this case the evaluation is carried out using the correlation method with two windowed signal profiles with the piecewise inversion of signal pieces in each case.

27. Method for 3D recording of at least one object surface (5) illuminated with a radiation source (4) in at least one scene according to at least one of claims 25 and 26, characterized in that at least one imaging lens is moved to the receiver array.

28. Arrangement for 3D recording of at least one object surface (5) in at least one scene with at least one radiation source (4), which is designed as at least one structured, luminous array by means of at least one structured array and at least one luminous surface element is formed with at least one an illumination beam path with at least one illumination objective (1) assigned to the structured illuminating array, including an image thereof, which has an effective aperture diaphragm with a

Has aperture center, for structured illumination of the object surfaces (5) in the object space and with at least one imaging beam path associated with the illuminating beam path, with at least one imaging stage for the at least one object surface (5) with at least one imaging objective (2) assigned to the receiver array or an image thereof , 33) for imaging the elements of the object surfaces, which has an effective aperture diaphragm with a diaphragm center, elements of the at least one receiver array detecting radiation from the elements of the illuminated object surfaces in the object space during the recording process, and with a distance (d) of Pupil center (PZ _0B ) of the illumination _lens (1) in the object space from the pupil center (PZ _0A ) of the imaging _lens (2) in the

Object space, characterized in that the structured, luminous array has a movement system with at least one assigned to a movable component and at least from the mechanical movement of the structured, luminous array, the displacement distances of the luminous surface elements in the array space are formed and after imaging these displacement distances by the illumination lens (1) in the object space, the image of which is at least approximately as at least one Route bundle (SBi) is formed with a convergence point (Ki) and the convergence point (Ki) is at a minimum distance from the axis of the illumination lens (1) from the last part of the distance (d) and a maximum distance from the axis of the illumination lens (1) 16 times the distance (d).

29. Arrangement for 3D recording of at least one object surface (5) in at least one scene according to 28, characterized in that the displacement distances of the luminous surface elements in the array space are arranged at least approximately in parallel and so the convergence point (Ki) at least approximately is positioned in the focal plane of the illumination lens (1) in the object space and in the pupil center (PZ _0A ) of the imaging _lens (2) in the object space.

30. Arrangement for 3D recording of at least one object surface (5) in at least one scene according to at least one of claims 28 and 29, characterized in that the luminous array is designed as an electronically controllable line grid with controllability of the location of the lines.

31. Arrangement for 3D recording of at least one object Surface (5) in at least one scene according to claim 30, characterized in that the lines are arranged perpendicular to the main section, and the displacement distances of the luminous surface elements and thus also of the luminous surface elements with local extremes

Luminance are formed in the array space as a result of the mechanical movement of the structured luminous array and the electronic control of the structured luminous array in the array space and at least approximately at least one cluster of lines with a convergence point (Ki) is formed from these displacement lines in the array space is and the convergence point (Ki) in the pupil center (PZ _ÄÄ ) of the imaging _lens (2) is arranged in the array space.

32. Arrangement for 3D recording of at least one object surface (5) in at least one scene according to at least one of claims 28 to 31, characterized in that the displacement distances of the luminous surface elements are arranged at least approximately parallel to a defined straight line (g _AP ) are and the straight line (g _AP ) intersects the focal point (F _AB ) of the illumination lens (1) in the array space and the increase with the amount from the quotient “distance of the pupil center (PZ _0A ) of the pupil of the imaging lens (2) in the object space of the axis of the illumination objective (1) and focal length (f _B ) of the illumination objective (1) ", this increase in the straight line (g _AP ) being related to the axis of the illumination objective (1).

33. Arrangement for 3D recording of at least one object surface (5) in at least one scene after at least one of claims 28 to 32, characterized in that a component of the movement system is assigned to the receiver array and thus, during the mechanical movement of the receiver array on a displacement path, its elements are assigned displacement paths (AS _A3 ) on parallel straight lines, from which Images (AS ₀₃ ) of these routes (AS _A3 ), when imaged by the imaging lens (2), at least approximately at least one route cluster (SB ₂ ) with a convergence point (K ₂ ) is formed in the object space, and the convergence point (K _x ) and the convergence point (K ₂ ) with the focal point (F _0A ) and the pupil center (PZ _0A ) of the pupil of the imaging lens (2) in the object space are at least approximately brought to coincidence and the imaging lens (2) is telecentrically executed on the side of the space of the arrays

34. Arrangement for 3D recording of at least one object surface (5) in at least one scene according to at least one of claims 28 to 33, characterized in that a component of the movement system is assigned to the receiver array and thus during the mechanical movement of the receiver. Arrays on a displacement path, the elements of which are assigned displacement paths (AS _A] ) on parallel lines, with at least approximately at least one path cluster (SB ₂ ) with a convergence point (K.) From the images of these paths when imaged by the imaging lens ( ₂ ) ₂ ) is formed in the object space, and the convergence point (K _x ) and the convergence point (K ₂ ) with the focal point (F _0A ) and the pupil center (PZ _0A ) of the pupil of the imaging lens (2) in the object space at least approximately to the are brought and the lighting lens (1) and the imaging lens (2) are each telecentric on the side of the space of the arrays and the axes of the lighting lens (1) and the imaging lens (2) are arranged parallel to each other and the focal planes of the same are brought to coincidence in the object space and that the components of the movement system are arranged in such a way that an overall movement direction for the luminous array is at least approximately parallel to a straight line (g _A ) in the array space, so that the elements of the structured luminous array move on straight lines parallel to the straight line (g _A ) and this straight line (g _A ) with the focal point (F _AB ) of the illumination lens (1) is brought to the intersection in the array space and the increase with the amount from the quotient "focal length ( f _B ) of the illumination _lens (1) and distance (d) of the focal point (F _ÄÄ ) of the imaging _lens (2) in the object space from the axis of the Beleu directional lens (1) ", this rise in the straight line (g _A ) being related to a straight line perpendicular to the axis of the illumination lens (1).

35. Arrangement for 3D recording of at least one object surface (5) in at least one scene according to at least one of claims 28 to 34, characterized in that the structured array is formed at least on a partial area of a pane (83) which has a rotatable toric precision bearing with a shaft with a rotary motor is assigned so that a rotating disc (83) is formed.

36. Arrangement for 3D recording of at least one object Surface (5) in at least one scene according to claim 35, characterized in that the rotating disc (83) is formed with transparent plate sectors of different geometrical-optical thickness.

37. Arrangement for 3D recording of at least one object surface (5) in at least one scene of at least one of claims 28 to 36, characterized in that the receiver array is designed as a color camera.

38. Arrangement for 3D recording of at least one object surface (5) in at least one scene according to at least one of claims 28 to 37, characterized in that a special receiver array with RGB channels is used and this a fourth channel, an NIR channel for which the information for the 3D point cloud is assigned.

39. Arrangement for 3D recording of at least one object surface (5) in at least one scene with at least one radiation source (4), this being designed by means of at least one structured array as at least one structured, luminous array with at least two surface elements and at least one luminous surface elements is formed, and with at least one illumination beam path with at least one illumination objective (1) assigned to the structured illuminating array, which has an effective aperture diaphragm with an extension and an aperture center, for structured illumination of the object surfaces (5) in the object space and with at least an imaging beam path assigned to the illumination beam path for the Object surfaces (5) with at least one imaging lens (2) assigned to the receiver array or an image of the same for imaging the elements of the object surfaces (5), which has an effective aperture diaphragm, elements of the at least one receiver array emitting radiation from the elements during the recording process of the illuminated object surfaces in the object space, and with a distance of the pupil center (PZ _0B ) of the illumination _lens (1) from the pupil center (PZOA) of the imaging lens (2), the distance being at least one eighth of the extent of the aperture of the illumination lens (1) , where luminous surface elements in a luminance distribution have an at least approximately predetermined luminance, so that at least one image of a luminous surface element is formed in the object space by imaging with the illumination objective (1), characterized in that the sharpness volume m at least one image of a luminous surface element in a structured luminous array - through the predetermined geometrical-optical assignment of the luminous surface element to the illumination objective and the geometrical-optical assignment of the elements of the receiver array to the imaging objective and the geometrical assignment of the illumination objective to the imaging objective - is permanently fitted into the focus volume, which is represented by the images of the elements of the receiver array in the object space,

40. Arrangement for 3D recording of at least one object surface (5) in at least one scene according to 39, characterized in that the focus volume, which by the images of the elements of the receiver array are given in the direction of beam propagation, have at least as great a depth as the sharpness volume of an individual image of a luminous surface element and in the object space an image of at least one luminous surface element of a structured array each has an image of an element of the receiver - Arrays are permanently assigned.

41. Arrangement for 3D recording of at least one object surface (5) in at least one scene according to at least one of claims 39 and 40, characterized in that the structured array is constructed with a plurality of luminous surface elements which are arranged fixedly in a spatial structure.

42. Arrangement for 3D recording of at least one object surface (5) in at least one scene according to at least one of claims 39 to 40, characterized in that the structured array is designed as a transparent microlens array and the focal length and the axial Position of the microlenses are designed such that their foci are arranged in a 3D surface which at least approximately represents an optically conjugated surface to the target surface and the foci of the microlenses represent at least approximately optically conjugated locations of the target surface of a test specimen.

43. Arrangement for 3D recording of at least one object surface (5) in at least one scene according to at least one of claims 39 to 41, characterized in that on the structured array (53) at least one image ran with a spatial structure with at least one period in the form of at least one ramp (54, 56) with at least one inclined ramp surface (55, 57) in the compensating surface and on the inclined ramp surface (55, 57) luminous surface elements as from that Radiation source (4) illuminated window surfaces are arranged and the ramp surfaces (55, 57) are inclined so that the best-fit line (AG _A3 ) through the inclined ramp surface (55, 57) in the main section as shown by the illumination lens (1) in the object space as an image provides a straight line (AG _θ3 ) which at least approximately _targets the pupil center (PZ _0A ) of the imaging _objective (2), with several different ramps (54, 56) for several different compensation lines (AG _0] ) according to their mapping the illumination lens (1) is formed from the images of a line bundle (Bi) with a convergence point (Ki) and the convergence point (Ki) at least approximately in P upillenzentrum (PZ _0A ) of the imaging lens (2) is brought to coincidence.

44. Arrangement for 3D recording of at least one object surface (5) in at least one scene according to claim 43, characterized in that the luminous surface elements are arranged as a binary code pattern.

45. Arrangement for 3D recording of at least one object surface (5) in at least one scene with two imaging beam paths with two at least approximately identical, parallel imaging lenses (2, 33), a first imaging lens (2) and a second imaging lens (33), being the main levels the two imaging lenses (2, 33) are brought to coincidence and each of them is assigned a receiver array, so that a first and a second receiver array are arranged, characterized in that the first and the second receiver array are each at least a movement system is assigned and the resulting movement of the first receiver array takes place on a path on the first upper branch of a letter Y, and the path lies parallel to a straight line (g _A ip) which, on the one hand, is the focal point of the first imaging lens (2 ) intersects in the array space and, on the other hand, intersects the intersection point (P _A ) of the axis of symmetry between the two optical axes of the two imaging lenses (2, 33) through the coinciding main planes, part of the line of symmetry forming the lower part of the letter Y and the resulting movement of the second receiver array takes place on a path on the second upper branch of the letter Y, u nd the line is parallel to a straight line (g _A2P ), which on the one hand intersects the focal point of the second imaging _lens (33) in the array space and on the other hand intersects the intersection point (P _A ) of the line of symmetry through the coinciding main planes.

46. Arrangement for 3D recording of at least one object surface (5) in at least one scene with two imaging beam paths with two at least approximately identical, parallel imaging lenses (2,33), a first imaging lens (2) and a second imaging lens (33), the main planes of the two imaging lenses (2, 33) being brought to coincidence and each of them a receiver array is associated with elements so that a first and a second receiver array are arranged, characterized in that at least one movement system is assigned to each of the first and the second receiver array and the resulting movement of the first receiver array on a path parallel to the optical axis of the first imaging objective (2), and exactly the elements of the first receiver array are read out and from these a signal curve is formed, which are located on lines (AS _Al j) that are parallel to a straight line (g _A ip ) which on the one hand intersect the focal point of the first imaging lens (2) in the array space and on the other hand intersect the intersection point (P _A ) of the axis of symmetry between the two optical axes of the two imaging lenses (2, 33) through the coinciding main planes, wherein part of the line of symmetry forms the lower part of a letter Y and the resulting direction of movement of the second receiver er array takes place on a path parallel to the optical axis of the second imaging objective (33), and exactly the elements of the second receiver array are read out and a signal curve is formed from them, which are located on paths that are parallel to a straight line (gA ₂ p) lie on the one hand intersecting the focal point of the second imaging lens (2) in the array space and on the other hand intersecting the intersection point (P _A ) of the axis of symmetry through the coinciding main planes.

47. Arrangement for 3D recording of at least one object surface (5) in at least one illuminated scene with two imaging beam paths with two at least approximately identical, at least approximately parallel arranged imaging lenses a first imaging lens and a second imaging lens, the main planes of the two imaging lenses being brought at least approximately to coincidence, the pupil center of the first imaging lens being arranged at a distance (d) from the pupil center of the second imaging lens, and one each of these Receiver array is assigned with detecting elements, so that a first and a second receiver array are arranged, characterized in that the first and the second receiver array are each arranged at least approximately perpendicular to the main section, and the receiving surface of the first receiver Arrays is at least approximately arranged so that it contains the distance that is on the first upper branch of a

Letter Y is arranged, and the path is arranged at least approximately parallel to a straight line which on the one hand intersects the focal point of the first imaging lens in the array space and on the other hand the intersection of the line of symmetry between the two optical axes of the two imaging lenses through the coinciding main planes intersects at least approximately, with part of the line of symmetry forming the lower part of the letter Y and the receiver area of the second receiver array being arranged at least approximately in such a way that it contains the distance which is arranged on the second upper branch of the letter Y, and the Line is arranged parallel to a straight line, which on the one hand intersects the focal point of the second imaging lens in the array space and on the other hand at least approximately intersects the point of intersection of the line of symmetry through the coinciding main planes and thus at least from images of parts of the at least one pair of at least approximately optically conjugated images is formed in the two receiver surfaces in the object space.

48. Arrangement for 3D recording of at least one object surface (5) in at least one illuminated scene with at least two imaging beam paths with two imaging objectives, a first (33) and a second imaging objective (2), the pupil center of the first imaging objective (33 ) at a distance (d) from the pupil center of the second imaging lens (2), and each of which is assigned a receiver array, so that a first (106) and a second receiver array (114) are arranged, characterized there - By that the first and the second receiver array are spatially structured and each have at least two receiver areas on spatially separate areas and the receiver areas (107, 108) of the first receiver array (106) and the receiver areas (109, 110) of the second receiver arrays each (114) are arranged so that at least approximately pairs of optically conjugated images of at least parts of Em receiver areas of the first receiver array (106) and parts of the receiver areas of the second receiver array (114) are formed in the object space.

49. Arrangement for 3D recording of at least one object surface (5) in at least one scene according to at least one of claims 28 to 48, characterized in that the receiver array is designed as a CCD matrix camera (6).

0. Arrangement for 3D recording of at least one object surface (5) in at least one scene according to at least one of claims 28 to 49, characterized in that the receiver array is designed as a CMOS matrix camera.