EP1000376A1

EP1000376A1 - Method for improving optic perceptive faculty by modifying the retinal image

Info

Publication number: EP1000376A1
Application number: EP98942492A
Authority: EP
Inventors: Heinrich Alexander Eberl; Günter ABERSFELDER; Helmut Grantz; Thorsteinn Halldorsson; Horst Schmidt-Bischoffshausen; Stefan Uhl
Original assignee: DaimlerChrysler AG
Current assignee: EBERL, HEINRICH A.
Priority date: 1997-07-07
Filing date: 1998-07-03
Publication date: 2000-05-17
Also published as: JP2002510407A; WO1999003013A1; DE19728890A1

Abstract

The invention relates to an image improvement system in accordance with application DE19631414. Said system is characterized in that the reflex image in the interior of the eye is scanned on the retina (NH) and, after modification, is projected back into the eye by the same path. The invention proposes using an elliptical scan.

Description

Process to improve optical perception by modifying the retinal image

The invention relates to glasses with the aid of which a retinal reflection on the inside of the retinal reflex image of the eye is recorded electronically with different brightness of the surroundings, modified with a computer and superimposed on the original image physiologically without delay via a lighting device and a back reflection via the same glasses is that there is an improved visual impression.

The use of optoelectronic glasses to mirror computer-generated images with the name "cyberspace" or "virtual reality" is increasing rapidly today. This technology is of broad use both for use in the entertainment industry and in the various fields of industry, transport and medicine and will continue to increase in popularity and importance with the availability of ever faster image processing computers.

The most widespread application is with closed, non-transparent glasses, in which images of miniaturized cathode ray tubes or liquid crystal matrices are presented to the eye via mirror or glass fiber systems. The particular attractiveness of this technique is to couple with moving three-dimensional image representation, the image sequence or the action with different movements of the glasses wearer. A change in the direction of view is simulated by moving the head or changing the perspective as the movement progresses. The movements of the arms and fingers of the wearer of the glasses can be brought into the image with the aid of sensors in order to enable him to intervene directly in the action.

In newer systems with the name "augmented reality", the wearer of glasses can use partially transparent glasses to view both the surroundings and an image of cameras from the same scene or of other image contents reflected in the glasses via a miniaturized monitor Look at the helmet. A well-known variant of this method has already been introduced for the control of combat aircraft with the name Helmet-mounted-display (HMD).

With these techniques, however, several problems are known which are due to the way the face works and which are waiting for improved technical solutions. With closed glasses and a rigidly coupled monitor and monitor image, when the glasses are moved, the scene moves in the same direction, which contradicts his viewing habits in an unnatural way. By imaging the eye, he is used to the fact that the scene runs straight in the opposite direction. This problem has so far only been solved incompletely by the cumbersome measurement of the head movement and the eyeball with external rotation angle sensors, corresponding image processing and tracking of the generated image.

The eye itself is able to roughly stabilize the retinal image by adjusting movements of the eye apple, which originate from so-called vestibular ocular reflexes (VOR) of the ear-canal system and serves to hold the fixation point during head movements. The fine adjustment is done with the image as a reference. This image tracking is also used by the eye to adapt the VORs to dynamic eye alignment.

This means that a superimposition of external images can only give a realistic image impression when they are coupled to the real retinal image.

When the glasses are closed, an attempt is made to use the image of the blood vessels (fundus) as a reference (retina tracking). However, this only provides an insufficient resolution and is only suitable for monocular considerations (see, for example, E.Peli, "Visual issues in the use of a head-mounted monocular display", Optical Engineering, vol. 29, No.δ, p.883 (1990) A simultaneous stabilization in both eyes of images with these is practically impossible because of the different orientation of the eyes Conflict between the vestibular and visual information often on movement disorders up to Seeta ankness. These problems of the existing technology are described, for example, in the review article by E. Peli, "Real Vision & Virtual Reality" in Optics & Photonics News, July 1995, pp. 28-34.

The object of the invention is to solve the problems of image stabilization when superimposing external images with the real image.

The invention is based on the older German patent application 19631414 with the designation: "Device for recording the retinal reflex and superimposition of additional images in the eye". This describes a device with which the retinal reflex image using a confocal imaging, two-axis scanning system via the reflection of the The inside of partially transparent and appropriately curved glasses is recorded in series with a highly sensitive photodetector.

It is proposed to use lasers and a beam splitter to image the improved image serially on the retina over the same light path in the opposite direction as the recorded image.

It also opens up the possibility of overlaying other images on the retina.

The above-mentioned problems are basically solved with this technique, but concrete implementations and applications have not been specified. The basic idea of the new invention to use this method to improve the perception of the eye. The physical-technical problems that have to be solved for this result from the physiological properties of the eye and the constantly changing lighting conditions in the environment. Because of the variable lighting conditions and the different optical tasks in its basic functions, the eye is a very dynamic sense organ. It adapts to varying the intensity of the backlight over 12 decades. It switches from color vision in daylight to pure black and white vision at night. Light in the wavelength range 400-1500 nm is from transmitted to the eye and imaged on the retina. Only light in the range of 400 nm to 750 nm is perceived, ie the infrared light in the range of 750-1500 nm, which is very bright with both exterior and interior lighting, remains unused for visual perception.

The eye detects an angular range of approximately 100 ° horizontally and vertically. However, the image resolution decreases very quickly with the angular distance from the visual axis. Attentive, instantaneous vision is limited to a central angular range of only +/- 5 ° and "sharp" vision, for example when reading or driving, is limited to the very small central angular range of +/- 0.5 ° Different movements of the eyes This leads to the following consequences, which under certain circumstances impair the perception of the eye and which are to be improved in the context of the new invention:

Adaptation

Sharpening performance

Vision defects

Age - related underperformance and

Movement dynamics

The object of the present invention is now to propose an arrangement which, like the eye, is designed to be very variable in its basic functions and adapted to the requirements of the visual process, but at the same time the special physiology and dynamics of the eye and the varying lighting conditions of the environment and the invisible IR range taken into account and exploited. This can only be achieved inadequately with the scanning and scanning variants (serial raster scan, serial spiral scan) specified in the earlier invention report. This concerns both the scanning pattern of the image recording of the retinal reflex and the rear projection of the laser image into the eye.

A fundamental problem of serial versus parallel image scanning is the short dwell time of the scanner in every image pixel. A uniform scan, for example of 0.5 million pixels in a scan time of 40 ms, means an integration time of only 0.08 μs, ie 80 ns, in each pixel. In comparison, the parallel time integration of all pixels of the eye itself is 10-20 ms.

As is known from the use of lasers to record the retinal structure of the eye in the so-called laser scanning ophthalmoscopes, a laser power of approximately 40 μW is necessary in order to achieve a signal-to-noise ratio of 17 from an image pixel during the raster scan (see, for example, A Plesch, U. Klingbeil, and J. Bille, "Digital laser scanning fundus camera", Applied Optics, Vol. 26, No. 8. p.1480-1486 (1987)). Converted to the larger area, this would result in an irradiance in correspond to an image of an extensive source on the retina of 40 W / cm ² , which corresponds to the irradiance of bright spotlights or sun on the retina, ie relatively bright sources with a good signal-to-noise ratio can only be recorded on the retina with the raster scan. In order to detect the imaging of weaker sources on the retina, the sensitivity must be increased significantly.

However, serial image scanning has the crucial advantage of better suppression of stray light, easier optics and the possibility of exact reversal of the beam path when image back projection with a laser for the recording of the retinal reflex and for this reason should be retained in this invention report. An extension of the dwell time can be achieved by changing the scan pattern.

Because of the uneven distribution of the photoreceptors, with the highest density of the cones for sharp vision in the center of the retina and the opposite course of the rods for the blurred but light-sensitive night vision, the raster scan is by no means the optimal scan pattern. A scan pattern adapted to the vision process should become increasingly slower and denser for day vision towards the center, and vice versa for adaptation to night vision. In addition to the dwell time, the recorded signal can be influenced by changing the spot size of the scan and thus also the image resolution.

The number of signal photons N _s that are recorded by the retina per image pixel by a scanning recording device can be calculated using the following formula:

N _S = (BT Δλ τ) (A ₀ R) (S / 2π) (A _p / D ² ) (1 / ε)

in which

B = the spectral irradiance on the retina

T = the optical transmission from the retina to the photodetector τ = the integration time in an image pixel on the retina

A ₀ = the area of the image pixel

R = reflectivity of the image pixel

Δλ = spectral width of the received signal

A _p = pupil area

D = distance of the pupil to the retina

S / 2π = the angle distribution factor of the optical backscattering of the retina ε = energy of a photon at the absorption wavelength

describe.

As this formula shows, stronger signals, i.e. A larger number of signal photons can be obtained on the recording device by the following measures:

Extension of the dwell time τ of the scan in the individual pixels,

Enlargement of the scanning spot A ₀ on the retina Enlargement of the spectral bandwidth Δλ.

The invention proposes scanning the retina in a sequence of concentric circles (circle center is equal to fovea centralis), whose radius gradually increases or decreases. This type of scanning is known as a circular scan. The circular scan is optimal due to the rotational symmetry of the eye lens and the pupil around the visual axis and the rotationally symmetrical distribution of the photoreceptors in the retina.

The invention further proposes that an identical circular scan be used for the recording of the retinal reflex from the surroundings and the image projection with the laser. Since with the circular scan from the outside to the center, after reaching the center, the scan axis runs the same way backwards, you can either take the image during the scan to the center and project it from the center to the outside, or take the image over the entire scanning process and projection only in a second can be used.

With a constant deflection of scanning mirrors in two directions (Lissajou figure), the circular scan inevitably slows down the dwell time towards the center. However, the invention provides that, for day vision, the scanning time of neighboring circles, depending on the exposure conditions, can be additionally slowed down and even accelerated for night vision.

Because of the uneven distribution of the cones over the retina with a density in the center that is more than two decades higher, the sampling rate (residence time per pixel) in this area can be increased by this factor, 100.

For night vision with the higher distribution of the rods with increasing radius, it makes sense that the dwell time decreases to the same extent in the opposite direction.

As is known to the person skilled in the art, a circular scan can be carried out in analog control with periodically oscillating orthogonal scanning mirrors, or in digital control by approaching the circular track with a large number of straight sections. The third alternative is the use of programmable algorithms for analog control signals, which are digital can be called and are best suited for these variable conditions.

So that the received signal can also be increased in proportion to its area by enlarging the scanned image spot, the invention further provides that the current image pixel size on the retina can be variably set in addition to the scanning speed.

With the change of the image spot size, the image resolution is also adapted to the situation. In addition to changing the scanning area, the resolution can be set by changing the radius of the scan radii.

With an enlargement of the scanning image pixel of e.g. 10 μm to 100 μm is e.g. the image resolution of about 2 to 20 arc minutes (resolution range of reading and viewing) is reduced by a factor of 10, at the same time the received signal is increased by a factor of 100.

As is known to the person skilled in the art, in confocal scanning the image resolution is determined by the aperture diameter in the intermediate focus in front of the photodetector and can be adjusted by changing it. The invention provides that liquid crystal diaphragms or electro-optical diaphragms are used for this so that this setting is as fast as possible, i.e. can be performed within one sampling cycle.

Since the timing of the scanning and the size of the image pixel during recording and projection should be as identical as possible, the invention proposes that the change in the scanning process and aperture control in the projection channel is the same as in the recording channel. The variation in the optical integration time and image pixel area can then be compensated for in the projection channel by correspondingly varying the transmission power of the laser. The level of the received signal is also dependent on the spectral bandwidth of the receiver and can be increased by broadening it. The invention provides that in the area of bright day vision (photopic vision) a splitting of the beam path into the color channels red-green-blue corresponding to the eye color sensitivity can be carried out with a spectral width of about 100 nm each. This enables color-true image recording and, with appropriate three-color lasers, a color back projection into the eye.

In the case of weak ambient lighting, in which colors are no longer perceived by the eye (scotopic vision), the invention provides for the merging of all channels into a single (black and white) receiving channel without color resolution. Furthermore, the invention provides that this receiving channel not only encompasses the visible range of 400-700 nm, but also the near infrared range of 700-1000.

This leads to the following advantages for increasing the reception signal with weak backlight:

the eye has full transparency between 400 - 1000 nm and displays a comparable image between 700 - 1000 nm and between 400 - 700 nm.

the degree of reflection of the retina between 700 - 1000 nm is R = 10 - 20% compared to R = 3 - 5% between 400 - 700 nm

Photo receivers with high quantum efficiency such as photomultipliers and silicon avalanche diodes are available over the entire spectral range from 400 to 1000 nm

Incandescent lamps, which are used for indoor lighting of buildings, or outdoors for street lighting and vehicles, emit 10 times more light between 700 and 1000 nm than between 400 and 700 nm. the reflectivity of the vegetation of nature is higher by a factor of 5-10 between 700 - 1000 nm than between 400 - 700 nm.

As these examples show, with poor lighting (night vision), the received signal can be increased again by a factor of 100 by expanding the spectral range.

The expansion of the spectral range can either be permanently installed in each device or can be made variable by changing spectral filters. If a color display is not required, it makes sense to use green laser light for the rear projection into the eye because of the highest sensitivity and contrast perception of the eye with this color.

Additional methods for signal improvement that can be used here are the integration of several successive images and the image correlation, e.g. Images of the two eyes.

Overall, by varying the two parameters, the dwell time of the scan in the image pixels and the size of the image spot, with the addition of the infrared range and the use of image correlation, an entire dynamic range of the received signals can be recorded over seven decades.

With a total optical transmission of the receiving channel of T = 0.2 (see formula above), the receiving area of this dynamic recording system comprises irradiations on the retina between 10 ^" W / cm and 100 W / cm, which includes the area of the typical inside and outside brightness .

Because of the slow and fast eye movements, it is necessary to design the scanning system so that it constantly follows the change of the visual axis through the glasses, i.e. that the axis of symmetry of the image scan, both in the recording and in the projection, is identical to the visual axis.

To achieve this object, the invention provides that before and after the scanning of the network treflexes or the image projection into the eye, a Centering the circular scan on the eye pupil is performed. The largest scanning angle of the circular scan is selected so that when the scan symmetry axis is deposited from the visual axis, the outer surface of the eyeball, sclera with rainbow skin and pupil opening is detected by the circular scan. Since these parts of the eye, which are well illuminated by the outside light, are not imaged sharply, but diffusely in the intermediate image plane of the photodetector, the received signal does not provide any image information here, but rather an integral indication of the optical backscattering capacity of the original.

If the received signals over sections of the same length, e.g. Quadrants, which are compared with each other from each circle, are only of the same height if the axis of the circular scan is identical to the eye axis (visual axis). Signal differences, due to the different backscatter from sclera, iris and pupil opening, are then a measure of the axis offset and their direction. After standardization with the entire received signal over each circle, these storage signals can be used to set the zero position of a next circle scan (bias). This means that the original placement of the axes can be reduced with each circle scan until it becomes negligible when the circle scan is immersed through the pupil opening (pupil tracking). Fig. 1 shows schematically the concentric scanning process with adjusted system, Fig. 2 shows the search mode for centering the scan through the eye pupil.

As an alternative to the use of ambient light, the invention provides that pupil tracking in the outer regions of the circular scan, with simultaneous signal evaluation in the recording channel as described above, is also carried out with the active illumination of the laser projection into the eye.

The invention also provides that the light scattered back by both the surroundings and the laser is recorded and evaluated even during the laser image projection. This simultaneous recording of the retinal reflex from the environment and the post-processing laser image projection opens up the possibility of the degree of overlap and constantly checking the temporal synchronization of the two images, recognizing any differences as image interference (moiré pattern), in order to then compensate for them subsequently using correction signals.

The recording and projection technique in the sense of the invention can be carried out either independently on one eye of a viewer or on both eyes at the same time. Because of the steroscopic vision of both eyes, a three-dimensional image recording and image reproduction is realized in the latter case.

It is not easy to understand that the recording of an error-free and distortion-free reflection image of the surroundings from the retina via glasses that are not individually adapted to each viewer in terms of their optical properties, nor can they sit completely stable on the viewer's head. The solution to this in the sense of the invention consists first of all in the relatively low optical requirements for the serial confocal point scan compared to e.g. an areal image from the eye, secondly in the complete dynamic adjustment of the optical beam path of the scanner via the glasses into the eye, which always takes into account the intrinsic movements of the eye and the glasses themselves, thirdly in the exact return of the beam path between recording and projection and the short time between these processes. Two scanning elements and a correction mirror, which can also be adjustable, are used to adjust the scan by the eye, even with the various eye movements. 3 shows a schematic overview of the entire system. The retina of the NH eye is scanned with the focused beam. Here AA represents the eyeball and AP the eye pupil. The partially permeable glasses are referred to here as BG.

The rays passing through from the surroundings are focused on the retina, at the same time the retina is scanned at points, the scanning beam always looking into a radiation sink through the glasses. The circular scan is carried out with the two-axis scanning elements HSS and VSS. With the auxiliary mirror HS, which can be actively adjustable, the direction of incidence and position of the beam on the inner surface of the glasses BG are set. With the SUS jet switch Either the illuminating laser beam can be passed through a central bore and the reception beam, which usually has a much larger diameter, can be reflected into the reception unit in separate directions, or an actively switching mirror element can be used that switches between reception and transmission.

The receiving unit can e.g. from three separate reception channels for the primary colors red, green and blue or other wavelength ranges e.g. exist in the near infrared range. The beam path of all spectral channels is brought onto an axis using dichroic mirrors DS. An actively adjustable field-of-view aperture GFB is used to adjust the spot size of the scanning beam on the retina and to possibly fine-tune the optical axis.

The sending unit can e.g. be created from three lasers with the basic colors red LR, green LG and blue LB. Before the beam combination on one axis with dichroic mirrors DS, the individual beams are either modulated externally with image modulators MR, MG and MB or, more simply, directly via the excitation current of the laser emission. The size and position of the laser scanning spot on the retina is controlled by an actively controllable aperture LAA which is set in the intermediate focus of two lenses in the beam path. The receivers for scanning the retinal reflex image are e.g. Suitable for photomultipliers, which alternately automatically switch to photon-counting operation when the optical signals are very weak and to a measuring operation when the signals are strong. Avalanche photodiodes can also be used as receivers.

Semiconductor lasers or miniaturized solid-state lasers with a low continuous wave power (<300 μW) are provided as light sources for the rear projection of the images into the eye, which can cause no danger to the eye. With the use of semiconductor lasers, the image modulation could be carried out directly via their power supply. So that all colors are generated, the use of three lasers with the basic colors red, green and blue is recommended. As the well-known color triangle of the human face shows, everyone else can Colors as well as the non-colors gray and white are formed by color summation of monochromatic laser lines of these colors. The invention also includes the possibility of using individual colors as a monochromatic solution.

The invention provides, as shown in Fig. 4, a signal processor SP, which electronically processes the direct image of the retina and synchronizes all functions of the device and that of scanners VSS / HSS, auxiliary mirror HS and laser spot setting LAA and size of the field of view aperture GFB coordinates. The image processing computer BVC then takes over the image perceived by the eye or images from other technical sensors which are fed to the computer via an external connection EA and processes them according to a predetermined software SW before they are modulated onto the laser beams as an image signal with the aid of the signal processor. In Fig. 4 the flow of the optical and electrical as well as the software signals are shown separately. The complete laser unit is denoted by DE, ME as the modulation unit and PME the complete receiver unit and SUS as the beam switch between the transmitter and receiver unit.

In addition to processing the current image processed by the computer, laser projection makes it possible to project into the eye and merge it with the original image, and to synchronously superimpose external images that are fed to the computer from the outside into the external image in the eye. If the time span between image acquisition and projection is correspondingly short in comparison to the rapid eye movements, the eye will no longer perceive an image interruption, as when viewing a television screen.

The separate but simultaneous image scanning on both eyes also captures the perspective differences between the two images. Since these are retained in both eyes during laser back projection, a restoration of spatial vision is guaranteed.

The components used in the invention are today largely miniaturized and available inexpensively. For scanning the circular figures miniaturized tilting mirrors can be used. The second option for producing the circular figures is to use wedge plate scanners that are designed for a beam path in transmission. The continuous beam is refracted by a fixed angle through each of the plates, the entire deflection angle can then be continuously adjusted to zero by a fixed rotation of the wedge plates against each other. When the wedge plates rotate together at a fixed rotational frequency, the deflected beam then describes a circular track. The third possibility is the use of acousto-optical deflection units, which offer the advantage of low inertia and fast deflection. The variably adjustable auxiliary mirror HS will preferably be a mirror that can be adjusted in two axes with microactuators.

For the adjustment of the laser spot size and the reception field of view, micromechanical actuators such as e.g. can also be used in widely used laser printers and CD record players.

The beam deflection unit and scanner can be accommodated in a simple glasses frame. With the help of fiber optic cables, laser projection units in a small housing e.g. can be accommodated in the size of a paperback book with food supply. The data exchange with an external permanently installed image processing computer can take place either via radio waves or infrared rays. According to the current state of the art, all elements of the device of the invention could thus be carried effortlessly by a human being and the wireless image data exchange with the external computer would enable its unlimited freedom of movement.

As in the earlier application of the inventor No. 19631414.3, this type of opto-electronic glasses can be used in a wide variety of applications:

Taking pictures of the outside world, processing them, projecting them back and merging them with the original picture in the eye, such as for Improved vision when driving a vehicle or as a visual aid for the visually impaired

Overlaying images from other recording systems, e.g. from the same scene in other spectral ranges to the direct image in the same applications as today or in the future the helmet-mounted display will be used

Superimposition of virtual images that are produced by the computer alone in the same or future applications of "virtual reality" or "cyberspace" image projection.

Claims

Patent claims

1. Image enhancement system in which the reflex image is scanned inside the eye and, after modification, is projected back into the eye in the same way according to application no. 19631414.3, characterized by the use of an ellipse scan.

2. Image enhancement system according to claim 1, characterized in that the ellipse scan is used by determining the outer edges of the pupil for adjusting and centering the scanner system without further external sensors.

3. Image enhancement system according to claim 1, characterized in that the recorded image is synchronized with the back-projected image temporally and locally.

4. Image enhancement system according to claim 2, characterized in that the scan duration is dynamically adapted to the requirement of resolution, acquisition time requirement and exposure time.

5. Image enhancement system according to claim 4, characterized in that the size of the scanning spot is dynamically adjusted according to the requirement of the environmental conditions.

6. Image enhancement system according to one of claims 4 or 5, characterized in that the track spacing of the scan tracks is adapted dynamically in accordance with the requirement of the ambient conditions.

7. Image enhancement system according to claims 1 to 6, characterized in that the size of the scanned area is adapted according to the requirements of the application.

8. image enhancement system according to claim 1, characterized in that the recorded image is brightened by a downstream image processing system and is then projected back.

9. Image enhancement system according to claim 1 and 8, characterized in that the recorded image by a projection of the image at a different wavelength than it was recorded and is thereby transformed into a different light wavelength range.

10. Image enhancement system according to claim 1 and 9, characterized in that the wave range of the recorded image is evaluated outside the perceptual range of the eye and is transformed into the visible range.

11. Image enhancement system according to claim 1 and 8, characterized in that the recorded image is so strongly brightened that the black-and-white (stick-eye) originally recognizable by the eye is transformed into color information (suppository vision).

12. Image enhancement system according to claim 1, characterized in that the recorded image is sharpened by calculating and modulating the projection using a suitable algorithm (Fourier transformation) in such a way that visual defects in the eye are compensated for.

13. Image enhancement system according to claim 1, characterized in that an external sensor for determining the pupil position is used for the adjustment of the scanner system.

14. Image enhancement system according to claim 1 to 13, characterized in that the recorded image is evaluated with regard to the image content in order to activate external reactions and control functions.

15. Image enhancement system according to claim 14, characterized in that the image contents of both eyes are compared.

16. Image enhancement system according to claim 14, characterized in that the position of the pupils of both eyes is compared.

17. Image enhancement system according to claim 15, characterized in that the image content of the forea centralis of both eyes is compared.

18. Image enhancement system according to claim 16 and 17, characterized in that the position of the pupils and image contents of the forea centralis of both eyes are used to determine the visual axis for triangulation (distance determination)

19. Image enhancement system according to claim 1 to 18, characterized in that the image information of the eye is used to determine the absolute brightness of the environment.

20. Image enhancement system according to claim 19, characterized in that the image information of the eye is used to determine the absolute color temperature of the light.

21. Image enhancement system according to claim 1 and 2, characterized in that the system is used to determine the pupil size.

22. Image enhancement system according to claim 8, characterized in that the recorded image is brightened by a downstream image processing system so that the physiological apparent sensitivity is shifted to another less sensitive area.

23. Image enhancement system according to claim 1, characterized in that the ellipse scan runs from the outside inwards.

24. Image enhancement system according to claim 1, characterized in that the ellipse scan runs from the inside to the outside.

25. Image enhancement system according to claim 1, characterized in that the ellipse scan is changed by a coincidence of the two focal points to form a circular scan.