GB2247090A - Flat-field scanner - Google Patents

Abstract

A scanner 20 for use in a thermal imaging system comprises an oscillating mirror 12 and a powered optical system 21 used in a double-pass mode to transmit radiation between a detector 10 and a focal plane 22. During the first pass rays normal to the detector 10 are brought to a focus on the surface of the mirror 12 and after reflection are once more wandered parallel by the second pass. A real image of the detector 10 is formed which on movement of the oscillating mirror 12 scans a flat field with a telecentric pupil. The mirror 12 is mounted at a small offset angle and is itself the aperture stop. For specific applications the field of view is provided by auxiliary optics in the form of objective lenses rather than telescopes. Mirror 12 need not be planar and the scanner 20 may operate at any magnification at a near unity. Various specific forms of the optical system 21 are disclosed. <IMAGE>

Description

FLAT-FIELD SCANNER This invention relates to flat-field scanners and particularly, but not exclusively, to thermal imaging systems incorporating flat-field scanners.

In scanning systems there is frequently a requirement to cause an image of an external scene to be swept across a detector formed by an extended array of detector elements. It is frequently also desirable that the exit pupil of the system should be telecentric in order to limit the acceptance angle required of the detector elements, and that radiation from one or more bodies of controlled or known emission is caused to fall periodically on the detector elements, in order to provide the necessary information to allow electronic compensation of the gain and offset errors which result due to imperfections in the detector array.

It is also a frequent requirement, for reasons of cost and commonality, that a single scanner should be capable of use over a variety of fields of view and resolutions by the use of different auxiliary optics, and that the interface with these auxiliary optics should be such as to minimise their cost and complexity.

A farther desirable feature in many cases is the provision of a well-corrected focal plane at which a graticule or other device may be placed. If the detector array is cooled, it is desirable that the magnitude of the radiation reflected from any refractive device placed at this focal plane should be substantially constant, since otherwise an undesirable shading of the image due to Narcissus effect may occur, so that if a reflective Narcissus graticule (as described in U.K. Patent No.

2074754 B) is used, the intensity of the signal from the graticule would vary with position in the scanned field.

If the device placed at the focal plane has plane surfaces (which is desirable for reasons of cost and ease of alignment) the implication is that the focal plane should be substantially flat and that the pupil at the focal plane should be substantially telecentric.

The present invention provides a flat-field telecentric scanner for use with a detector, said scanner comprising an oscillating mirror and a powered optical system used in double-pass mode so arranged that during the first pass of the optical system rays normal to said detector are brought to a focus substantially on the surface of said oscillating mirror, and so that after reflection from said oscillating mirror said rays are once more rendered parallel to each other by a second pass of said optical system, said optical system also being such that a real image of said detector is formed by rays making a first pass of said optical system, being reflected from said oscillating mirror and thereafter making a second pass of said optical system, said real image being caused to scan a substantially flat field with a substantially telecentric pupil by rotation of said oscillating mirror about an axis substantially coincident with the surface of said mirror.

The present invention enables use of an oscillating mirror at a sufficiently small offset angle with respect to incident radiation that the mirror itself can be used as the aperture stop of the scanner without a significant change in pupil area throughout the scanned field, the optics being such that entrance and exit pupils of the scanner are telecentric, a flat scanned field is provided, and the field of view for specific applications is selected using auxiliary optics in the form of objective lenses rather than optical telescopes, thereby saving the cost and transmission losses due to the eyepiece components of such telescopes.

The principle of operation is the use of a flat-field optical system in a double-pass mode in such a way that the principal rays of telecentric cones of radiation falling on the detector are brought to a focus at the oscillating mirror, after which they are rendered once more telecentric by a return pass through the optical system. The beam at the oscillating mirror may or may not be collimated, and in either case the oscillating mirror may or may not be plane. The scanner may operate at any magnification at or near unity.

There are several types of optical system which are suitable for use with the present invention including both reflector and refractor systems Embodiments of the present invention will now be described by way of example with reference to the accompanying drawings, in which: Figs. 1 and 2 illustrate known scanning systems; Fig. 3 illustrates a first embodiment of the present invention; Fig. 4 illustrates a second embodiment of the present invention; Fig. 5 illustrates a modification of the Fig. 4 scanner; Fig. 6 illustrates a third embodiment of the present invention; Fig. 7 illustrates a modification of the Fig. 6 scanner; and Fig. 8 illustrates a thermal imaging system incorporating the Fig. 7 scanner.

The operation of the scanners to be described is most easily understood if the scanner is regarded as causing a projected image of the detector to sweep across the scene. Since the radiation path is bidirectional the detector may itself be a radiation source.

In the Fig. 1 scanner the beam from the detector (10) is collimated by a lens assembly (11), and is caused to fall on a mirror (12) which is moved angularly with a "saw-tooth" waveform to generate the scanning action.

The beam is then increased in diameter (and the scanning angle is reduced to the value desired for the specific application) by means of a telescope (13) comprising two lens groups - an "eyepiece" group (13A) (so called by analogy with visual telescopes) and an objective group (13B). Referencing is accomplished by means of radiation emitting bodies placed at the edges of the field at the intermediate image (14) of the detector formed by the telescope (13). Mirror (12) oscillates by only a few degrees about a nominal position of about 450 to the optical axis of the telescope 13 and to the optical axis of the lens assembly (11)..

If the front objective (13B) of the telescope (13) is made the aperture stop of the system and the oscillating mirror (12) is placed at the exit pupil of the telescope (13), the cone of radiation received by any element of the detector (10) from the external scene remains static on the detector, and if the oscillating mirror (12) is placed at the focus of the detector lens assembly (11) the exit pupil is telecentric as desired.The main disadvantage of this arrangment is that control of the pupil of the system is relatively poor because if the telescope (13) suffers from distortion the semi-angle of the cone received by each element of the detector (10) will vary with position in the scanned field, giving rise to shading of the picture and if the telescope (13) is a multi-field or zoom device (as is frequently required) it is not generally practical to make the telescope entrance pupil the aperture stop of the system.

An alternative approach with the Fig. 1 scanner is to place an aperture stop as close as possible to the oscillating mirror (12). For mechanical reasons, this distance cannot be zero, and is usually at least half the diameter of the beam at the oscillating mirror. If the detector (10) is a linear array (or an array in which the length perpendicular to the sweep direction is much greater than that parallel to the sweep direction) the scan pupil lies substantially on the oscillating mirror (12), while the pupil transverse to the scan lies at the aperture stop (which is spaced from the oscillating mirror). This gives rise to astigmatism of the pupil which necessitates a considerable increase in the size of the objective group (13B) of the telescope (13) if vignetting is to be avoided.

A farther possibility with the Fig. 1 scanner is to use the oscillating mirror (12) itself as the aperture stop of the system but one difficulty with this approach is that of designing a detector lens assembly (11) which gives good performance when its entrance pupil is projected forward on to the oscillating mirror (12). A more fundamental problem is that as the angle of the mirror changes, so does the area of the beam which it intercepts. For example, if the mirror is nominally at 45 degrees to the optical axis at the centre of the scan, and angular movement of the mirror is + or - 5 degrees, the effective pupil area changes by 19%. Such a variation would lead to unacceptable shading of the image in many cases.The approach would be satisfactory if the offset angle on the mirror with respect to the optical axis could be greatly reduced, but if this is done in the Fig. 1 arrangement the detector lens assembly (11) and telescope (13) become impractical due to the large separation between the oscillating mirror (12) and the lenses (11), (13A).

A farther method of overcoming the problem of pupil control according to the prior art is shown in the scanner of Fig.2. In this case a concave mirror (15) is used to image an aperture stop (16) located adjacent the lens (11) onto the remotely located oscillating mirror (12). The detector lens (11) is now a relay lens instead of a collimator(2) but the mirror (12) remains with an offset angle of about 45 degrees with respect to incident radiation. The arrangement is perfectly satisfactory from the point of view of performance, but is complex and not particularly compact.

The prior art contains many other examples of scanners suitable for linear detector arrays, but the examples described are typical and sufficient to illustrate the advantages of the present invention.

Fig. 3 illustrates a first scanner (20) in accordance with the present invention incorporating a Petzval lens (21) which gives a telecentric exit pupil with an entrance pupil sufficiently remote to allow the oscillating mirror (12) to be placed there. Lens (21) could be the 3-element type illustrated at (11) in Fig. 1 but could be 2-element provided with aspheric correction for spherical aberration. The mirror (12) is located on the optical axis of the lens (21) whilst the detector (10) is offset as is the scanned focal plane (22). Radiation from the field of view and provided by auxiliary optics (not shown) traverses plane (22) and makes a first pass of the optical system formed by the lens (21) to be reflected by mirror (12) back through the lens (21) towards the detector (10).The mirror 12 is located at a focal point for the detector (10) to provide a telecentric exit pupil despite the offset axes of the detector (10) and plane (22) with respect to the optical axis of the lens (21) and mirror (12), the mirror (12) has a very small angular offset with respect to incident radiation. Mirror (12) is planar and operates in a collimated light region. The Fig. 3 arrangement is quite practical, but has a rather restricted back focal length which may preclude its use with detectors (10) in large encapsulations.

A second embodiment of scanner in accordance with the present invention is shown in Fig. 4. This is based on the well-known Offner two-mirror relay, the oscillating mirror (12) forming the secondary mirror of the relay.

The arrangement is flat-field and telecentric, and since it comprises only reflecting elements is fully achromatic. The arrangement is suitable for modest f/numbers and scanned fields, but optical aberrations restrict performance. These aberrations can be reduced by the use of a concentric refractive corrector (25) as shown in Fig. 5, but at the cost of increased complexity and reduced transmission.

In Figs. 4 and 5 the mirror (12) is convex having a radius of curvature half that of the concave mirror (26) of the relay, both mirrors having a common centre of curvature. The axes of the detector (10) and the focal plane (22) are offset with respect to that of the relay.

A further embodiment of scanner in accordance with the present invention is illustrated in Fig. 6 and incorporates a Dyson lens (30). The lens (30) is a unit-magnification relay being a thick plano-convex lens and the mirror (12) is a concave mirror, the thickness and curvatures being such that the curved surfaces are concentric with their common centre of curvature lying on the plane surface (27) of the lens (30). The convex surface (28) of the lens (30) brings parallel bundles of rays entering through the plane surface (27) to a focus on the surface of the concave mirror (12). The concave mirror (12) is oscillated about an axis perpendicular to the optical axis of the lens (30), and the detector (10) and the scanned image-plane (22) lie substantially on the plane surface (27) of the lens (30) but offset with respect to its optical axis.

The Fig. 6 configuration is limited to types of detector (10) which may be placed substantially in contact with the lens (30) and which have a sufficiently small encapsulation to allow unobstructed access to the scanned focal plane (22). With-infra-red detectors of the type used for thermal imaging, cooling is generally required,so that the elements of the detector must lie a significant distance from optical surfaces, and due to the need for a Dewar, the detector encapsulations are generally too large to allow use of the scanner shown in Fig. 6. Moreover, the optical transmission of the thick lens (30) in the thermal imaging wavebands is relative poor, and the materials are costly.If the lens (30) is simply thinned, severe spherical aberration occurs, but this may be corrected by aspherising the concave mirror (12), and results in the scanner illustrated in Fig. 7, while Fig. 8 shows a combination of the Fig. 7 scanner and an objective lens which allows imaging of distant objects or scenes.

The lens (32) in Figs. 7 and 8 is substantially plano-convex and is made preferably from a material of high refractive index and low dispersion. For thermal imaging in the 8-12 um band, germanium is the most suitable material, while for 3-5 um operation silicon is to be preferred. Gallium arsenide is a possible compromise if operation in both wavebands is desired.

The aspheric concave mirror (12) is substantially concentric with the spherical surface (32A) of the lens (32) when rotated to the position corresponding to the centre of the scanned field, and is placed at the focal surface formed by rays incident normally to the plane surface (32B) of the lens (32). The diameter of the concave mirror (12) is chosen so that it itself acts as the aperture stop of the system, and the concave mirror (12) has its principal normal lying along the optical axis of the system. The concave mirror (12) preferably has its surface aspherised as required to compensate for the spherical aberration caused by the lens (32), although it is possible in a complete imaging system such as that shown in Fig. 8 to use a spherical mirror (12) and to compensate for the spherical or chromatic aberration in the attached objective lens (40) or vice versa, or to make the mirror (12) a second surface mirror, the curvatures of which are chosen to provide the desired aberration correction.

The detector (10) is offset from the optical axis of the lens (32) by an amount sufficient to allow access to the scanned focal plane (22) and since aberrations increase with offset distance, it is desirable to minimise the offset distance. For this reason a plane fold mirror (41) is introduced between the detector (10) and the lens (32) to allow the detector (10) to be positioned clear of the scanned focal plane (22).

Scanning is achieved by rotation of the concave mirror (12) about an axis perpendicular to the optical axis of the lens (32) and when this mirror is rotated, the image of the detector (10) moves in a plane locus with a telecentric pupil. There is some degradation of image on in:eneral, but by suitable optimisation of parameters, adequate quality for many applications can be achieved.

Temperature references (42) may be placed at the edges of the scanned field (22) as shown in Fig. 8, and, if desired, a transmitting plate located in field (22) could, for example, carry a Narcissus graticule (as described in U.K. Patent 2074754B) which would (due to the telecentric scan pupil) be uniformly "illuminated" by the detector (10).

A difficulty with extended detector arrays is that of providing efficient cold shielding. A fixed concave mirror (44) can be placed behind the oscillating mirror (12) in such a position that radiation from the cooled detector (10) is reflected back on itself, thereby reducing the level of stray radiation.

The Fig. 7 scanner may be used on its own to examine objects placed at the scanned focal plane (22) but for the examination of distant objects or scenes it is necessary to attach objective lenses (40) of a focal length selected to suit the desired application. In most prior-art scanners a telescope rather than an objective lens must be used, so the present invention removes the need for a telescope eyepiece thereby reducing costs and improving optical transmission. The auxiliary objective lenses may be single-field, multi-field or zoom types as desired.

Claims (8)

1. A flat-field telecentric scanner for use with a detector, said scanner comprising an oscillating mirror and a powered optical system used in double-pass mode so arranged that during the first pass of the optical system rays normal to said detector are brought to a focus substantially on the surface of said oscillating mirror, and so that after reflection from said oscillating mirror said rays are once more rendered parallel to each other by a second pass of said optical system, said optical system also being such that a real image of said detector is formed by rays making a first pass of said optical system, being reflected from said oscillating mirror and thereafter making a second pass of said optical system, said real image being caused to scan a substantially flat field with a substantially telecentric pupil by rotation of said oscillating mirror about an axis substantially coincident with the surface of said mirror.

2. A scanner as claimed in claim 1, wherein the oscillating mirror is at a sufficiently small offset angle with respect to incident radiation that the mirror itself can be used as the aperture stop of the scanner without a significant change in pupil area throughout the scanned field.

3. A scanner as claimed in either preceding claim, wherein the field of view for specific applications is selected using auxiliary optics in the form of objective lenses rather than optical telescopes.

4. A scanner as claimed in any preceding claim, wherein the oscillating mirror is a plane mirror.

5. A scanner as claimed in any preceding claim, wherein the scanner is arranged to operate at or near unity magnification.

6. A scanner as claimed in any one of claims 1-5, wherein the optical system is substantially as described with reference to any one of Figs. 3 to 7 of the accompanying drawings.

7. A thermal imaging system incorporating a scanner as claimed in any preceding claim.

8. A thermal imaging system substantially as described with reference to Fig. 8 of the accompanying drawings.