DE3346589A1 - SCANING METHOD WITH AND INQUIRY METHOD OF DETECTOR ROWS AND RELATED ARRANGEMENT - Google Patents

Description

description

Scanning method with and interrogation method of detector lines as well as the associated arrangement

The invention relates to a method for sensing radiant energy according to the preamble of claim 1 and to a device for performing this method.

A method of this type is the subject of DE-AS 23 32 245. Let the video signals obtained here in parallel scanning can only be prepared for television using a comparatively complex standard converter. From DE-PS 22 50 251 and DE-OS 22 24 275 are then scanning methods for terrain profiles using

', 0 of a row of detectors known. In the first case, the row of detectors itself takes on the function of the scanning mirror by moving it over the terrain of interest. The information obtained is transmitted to delay elements, in part via ring switches. These publications also lack a reference to a TV-compatible reproduction of the information obtained.

The invention is based on the object of improving the generic method in such a way that the video signal thus obtained can be reproduced on a monitor in a TV-compatible manner without the interposition of a standard converter. This object is achieved according to the invention by the features mentioned in the characterizing part of claim 1. In this way you can get a TV-compatible video signal on the one hand and on the other hand - via the number, asr detector rows and their mutual offset - avoid aliasing ^ effects in the horizontal direction or meet the sampling theorem. The undesired aliasing effects arise if the query is not continuously but only at certain times. Then, in principle, multiple spectra occur, with the undesired sidebands forming the basic spectrum if the sampling theorem is not fulfilled

more or less strong upper lobe, so that false frequencies not present in the original spectrum occur, which are in
Make moiré effects (aliasing) noticeable.

The embodiment according to claim 3 has the advantage that immediately After the nesting process, the contents of the shift register can be displayed as a monitor line.

An advantageous further development according to claims 5 and 6 also appears it that the predetermined spatial frequencies are not disturbed by sidebands of the multiple spectrum.

Regardless of whether the detector rows are in the so-called CCD-FPA-j

Technology (= charge coupled device - focal plane array), i.e. with |

serial readout of the line raster signals, or in conventional J

Technology, i.e. with parallel leading out of the electrical version

Conclusions to be applied are expediently a |

Intermediate storage of the line grid according to claim 7 vorgeb |

beat. CCD is understood here to mean well-known semiconductor |

Charge transfer devices. j

The remaining subclaims also contain developments of Er- j

finding. f

In the following, exemplary embodiments of the | Described invention in more detail, wherein have the in the individual figures einan- \ of the corresponding parts the same reference numerals. It shows f

1 shows a two-line detector arrangement with delay |

member and totalizer, the individual elements of which I

spaced from each other by element width sina una their |

both lines parallel to each other and against each other- I,

others are offset by the element width,

FIG. 2 shows a four-line detector arrangement according to FIG. 1,

but with a line offset of only half
Element width,

3a shows a simplified two-line detector,

3b shows the two-line detector according to FIG. 3a, each with one parallel buffer and a parallel CCD line in a schematic representation,

3c shows the pulse sequence diagram belonging to the line detector in FIGS.

Figure 3d shows the back and forth 1 signals from the line detector monitor electron beam line written in accordance with FIGS. 1 and 3a to 3c,

FIG. 4a shows the propagation scheme of the individual line rasters at different points in time in an according to FIG. 2 designed multi-line detector in CCD-FPA technology,

4b shows the propagation scheme of the individual line rasters in a four-line detector designed according to FIG. 2 in conventional technology,

5 shows the four-line detector according to FIG. 2 with a parallel shift register and buffer similar to FIGS. 3b and

6 shows the application of the four-line detector according to the invention for TDJ operation in the scanning direction of Detectors.

1 shows a detector arrangement which consists of the two parallel detector rows I and II, each with D ^ individual elements of width x _D , the center of which is δχ ₀ . The two rows of detectors are offset from one another by the distance ΔΧς. The line length corresponds to the entire horizontal field of view of the scene to be recorded, which results from the column numbering 1 to 2 N of the object lines. The vertical field of view of the scene is recorded by the fact that the scanning mirror, which is not shown in the drawing for the sake of simplicity, shows the scene of interest

from in the TV field duration of e.g. 20 ms / - in the viewing direction - above

leads downwards continuously over the rigid detector arrangement.

An object line is swept over by the two detector lines I and II one after the other, that is to say at a time interval of a TV line duration. In this case, an object line raster corresponding to the respective detector line is formed at each sweep time. The two object line rasters created in this way, which are spatially (in the x direction) by δχ and temporally m the T ¥ -Zeilenciauer against- = 10 ", are nested spatially and temporally correctly with the help of an electronic device and result in the video signal for the electron beam of the TV system.

Ξ Shall, for example, spatial frequencies up to the detector spatial frequency

f _v η -ίγ ~ and if aliasing is to be avoided, then D

for example, according to FIG. 2, η = 4 and δχ = -L · - x _n to be selected. One is satisfied

11 with half the spatial frequency —3— f „ _n = ^ ₅ - —that is sufficient

£. X, UT \

two-line detector arrangement known from FIG. 1, where δχ = Xj ₃ .

The functional sequence is first described using the simpler version of Fig. 1 with η = 2 and δχ = X _n :

At time t ^ the detector line I is in the object line y ^ and forms the following object line grid, with "object" the scene designed in the detector level is meant:

The vertical scanning speed is adjusted so that the entire vertical deflection corresponds to the TV field time and the detector line II corresponds to the object line y, 64 us later than the detector line I achieved. At this point in time, the latter is in the object line y ~. The resulting grid is therefore the interlace method adapted to the TV system. The detector line I then forms the grid

^Y 1III: 2 4 6

If the raster y ^ is delayed by T _z = 64 μs and if clocks I and II are interleaved in time according to the lower part of FIG. 1 in the viewing direction, the result is the result at the output of the adder σ

This gives a u>. the raster twice as fine as it corresponds to the distance between the individual detectors of the detector rows I or II taken individually. The sampling interval is therefore x _D , so that by appropriate pre- and post-filtering, spatial frequencies up to -1— “—-—- can be transmitted alias-free, while without

this measures the alias limit

lies.

This detector arrangement can also be implemented using CCD-FPA technology. In this case, it is of interest whether the necessary clock frequency, with 6s within a CCD line, is able to shift the charge is feasible. This clock frequency f ^, so the reciprocal of the charge transfer time from main memory H via a buffer Z for the next main memory H, as shown as seen from Fig. 3b, must be chosen so that the entire detector row in the line trace t _H = 52 us of Monitor electron beam is read out.

The terms "main memory" and "intermediate memory" are used in the upcoming Embodiment in connection with the charge transport mechanism according to the bucket chain principle - according to which the CCD's work - used. The main memories are each assigned to a detector and take over from it Detector signals, while the buffer is only used for charging to serve. It should be emphasized that this is only an auxiliary concept without describing the physical process exactly.

For the clock period t _HH , therefore, one N _y -tel of the trace time t _{H is available} . The following therefore applies:

HH

^U HH

β φ

• I ■

- 11 -

Here is a numerical example for estimating the clock frequency at j

a detector number N _v = 300: t

^f HH - -5Tus * ³⁰⁰ - ⁵ ' ⁷⁶⁹

|:

Even with this relatively high number of detectors, a clock |

frequency reached within the scope of the future Ll-CCD-Detek- |

gate arrangements feasible lies. |

As already mentioned, according to FIG. 2, n _y = 4 and ax _s = - \ - x _D j

be selected if an aliasing-free resolution up to the detection ■

gate frequency f " _n = -J" is to be achieved:!

χ, υ Aq I

Referring to FIG. 2 is located at the time t _y1 here again the}

Detector line I in the object line Y ₁ and forms the following object f

line grid

The vertical sampling interval again corresponds to the TV
Field time, and the detector line II reaches the object line
by t ₂ = 64 microseconds later than the detector row I. The latter is closed
this point in time in object line y ₃ (interline method)
The detector line I then forms the grid

V _{1 n} : 1.5 3.5 5.5

Another 54 ps later, the grid is created
y _1in : 2 4 6 .....

and again 64 us later the grid

Y _{1 IV} : 2.5 4.5 6.5

φ

• * ff * ■

ψ ■ · · t

- 12 -

If the grid y ,, is delayed by τ = 3. t (t = 64 µs), the raster

Ί II

at χ * Z

t and the grid y * -tti around τ = 1

t opposite

the grid y _1IV and interleaves the bars I to IV according to the lower part of FIG

1 1.5

2.5

3.5

This results in a grid that is 4 times finer than the distance between the individual detectors and the frequency limit for alias-free transmission has been quadrupled, that is to say improved to —r-.

The clock frequency f _HH is the same as with the two-line arrangement if the same number is used for the detector lines N is chosen. For example, for Ν _χ = 300 f _m = 5.769 MHz. It should be noted that all detector lines take up new object line rasters every t = 64 με, while those before time 3t, 2t and It for detector line I or before time 2t _z and 1t _z for detector line II or before time It at detector line III generated raster are still on the corresponding delay line (or in the shift register or memory) to then after and after being processed.

The operations described so far apply to the first one Field of the TV standard. In the second field duration these processes are repeated, only that now the line grid from the lines with even-numbered index, namely

^etc -

It is not necessary for the detector arrangement to have one performs mechanical-optical line offset jump. Only the object line signal takeover points have to be shifted accordingly with respect to the vertical scanning movement - from an electrical point of view in terms of time.

With the help of the delay and signal acceptance times, the described method not only a double line jump, but a multiple interlace can also be implemented. With a 4: 1 line jump is then possible, also in a vertical direction to increase the frequency limit for aliasing-free transmission up to the spatial frequency of - determined by the detector geometry, where y _{D is} the dimension of the detector in the vertical direction.

The described detector arrangement for IR detectors can be used in conventional

licher technology, i.e. the individual detector lines are taken from the associated Dewar vessel indicated in Fig. 4b, from which the detectors are surrounded, which leads to certain difficulties with the large number of detectors and their lines brings itself. The detector arrangement can also use CCD technology be carried out with so-called FPAs. In the latter case, the Realization of the delay elements internally on the focal plane With the help of CCD shift registers as well as externally. With conventional detector technology, only the external solution comes into question.

Let us first consider the internal solution: When arranging after In the FPA case, a CCP line (not shown here) is assigned to each detector line in FIG. 1. It is sufficient to read out the CCD I accordingly later than the CCD II. A pulse sequence plan Fig. 3c shows.

In the arrangement according to FIG. 2 there are either - as shown in FIG. 4 - additional 3-fold registers Ia, Ib and Ic or 2-fold registers Ha and Ub or 1 for detector lines I, II and III -fach register Ia with serial transport of the line grid necessary to achieve the delay times. Or, as shown in FIG. 5, multiple buffer memories B with a parallel but staggered transport between the buffers and subsequent serial readout from the respective line CCD are necessary.

334658S

For the external solution, on the other hand, the following applies: First is the usual arrangement to be provided by one CCD line per detector line. The necessary Delays can then be implemented either with conventional analog delay lines or with external shift registers or analog RAM memory (= random access memory) can be implemented.

Solution according to FIG. 3 for the two-line transmission te ktor of FIG. 1

3 shows a delay arrangement for a two-line detector including a pulse sequence diagram. As mentioned, in this simple case it is sufficient to read out the CCD I correspondingly later compared to the line raster I recording, while the CCD II is read out practically without delay.The two line rasters are fed alternately to the monitor electron beam control electronics. In this solution, according to FIG. 3b, each of the two detector lines I and II is assigned a buffer line B for intermediate storage and a CCD line for charge transport. For the sake of a better overview, only the three individual _{detectors D ^, D2 and D 3} per detector line have been assumed. Accordingly, there are only three buffers B and three CCD modules, each with a main memory H and an intermediate memory Z. The individual detectors are via the electronic switches Sch 1j or Sch I ₁₁ with their assigned buffers B and these in turn via the electronic switches Sch 2j or Sch 2jj connected to the associated main memory H.

In the direction of view above the pulse sequence diagram of FIG. 3c, the detector positions in FIG. 3a are at times t. t ₃ t ₅ and t j are shown. The signal acceptance of the object line y ^ via the detector line I occurs at time t . via the detector

y '»line II at time t, etc. between t. and t - one recognizes the relatively long intermediate storage of the line signal in the buffer B ₁ ; it is not transferred to main memory Hj until shortly before t ₃ , so that buffer Bj is available for a new signal transfer _{at time t 3.} Here 1 3 5 consisting - - _γ between (* ιj ^ beginning ^and fryiI ^ Ehde the entire object line I ^{is then} read out a corresponding process takes place in the elements of Detek-.

^- 10 ™

gate line II. The only difference to the processes in I is that Bj ₁ passes on his load immediately after receiving the signal, while Bj has carried out the relatively long interim storage. ^{Object line II is then} read out between (t _m ) ^^ - which differs from (Ι _η ) ^^ only by t _HH - and (t ^ nJg _n ^ g - here consisting of 2 4 6 - and with object line I. nested automatically to the sequence 12 3 4 5 6.

In Fig. 3d the back and forth signals are the tonitor electron beam line shown, from which it can be seen that the signal transfer and the intermediate storage is carried out in the flyback time will. The ratio of the outward and return travel of this line (52 to 12 μs) corresponds to in the present exemplary embodiment For technical reasons, it is only indicative.

For exter n s and internal solution according to Fig. 4 for the Vierzeil ene Φ

detector of Fig. 2 f

I.

Are the necessary delays of 3t, 2t and 1t according to I ₁

Fig. 2 carried out with serial shift registers, then applies to the}

external and internal solution the principle according to FIG. 4, from which j

how the line grid y.r y.jj y, ™ changes during the §

different times t ^ to t, propagate and at time |

ty at the respective output of the detector lines I, II, III, IV for |

Nesting are available. =

In the external solution, the dividing line is between CCD- f

Detsktcrsncrdnung ynd outside visit at the sit "extern" marked f

neten and in the case of the internal solution marked with "internal" j

third line. In the latter case, the additional slider *

gister I I. and Γ, or II. and II _K or III. just like the s

a, u c a D a I

Shift registers I, II, III and IV constructed as CCDs, whereby I

per sub-arrangement I, II, III not linearly one behind the other, but \

Z = B = parallel next to I, II and III, but electrically self- \

understandable in series. The two terms "external" and "internal" ·

• · ■

S »

• t «· t

- 16 -

relate here to the arrangement of the delay elements outside or inside the image field plane (focal plane), i.e. outside or inside the dewar. In particular, the detector arrangement in accordance with Fig. 4b constructed in a conventional technology, the sub-arrays I ₁ II, III and IV are to be understood as the left of the dividing line lying.

For the internal solution according to FIG. 5 for the four-line detector of FIG. 2

Instead of the serial shift register between the detector rows I. Ms η can also be provided with parallel buffers B or shift registers be. The basic structure is shown in FIG. 5. From this you can see the position of the individual line grids at the various points in time t * to ty.

TDJ operation (= time-delay-integration)

The method described also makes it possible to carry out the TDJ operation, which is known per se, in the scanning direction of the detector arrangement. The arrangement required for this is shown by the four-line detector according to FIG. 6. For better clarity, only three detectors are provided here per line - perpendicular in the viewing direction. In this case, the scanning speed must be coordinated so that the distance Ay _DZ is covered in the line duration t (i.e. not only Ay _n, as without TDJ). From one line trigger point to the next, m signal transfers are to be carried out if m is the number of TDJ detectors.

Claims (17)

ELTRO GHBH GESELLSCHAFT FOR STRAHLUNGSTECHNIK 6900 Heidelberg, Kurpfalzring 106 claims 1. Scanning method with and electronic interrogation method of detector lines using an iR scanning mirror, where ä) the radiation energy on one from several individual detector elements composite IR detector meets the corresponding electrical signals after optoelectronic conversion gives up, b) the scanning mirror, the radiation energy in the vertical RICH \ Q processing on the IR detector and directs c) the distances between the individual detector elements of a detector line are equal to the dimensions of the individual detector elements in the scanning direction characterized in that d) the scanning mirror the radiant energy of the entire vertical Field of view continuously on the horizontal and parallel detector lines (I to η) expanded and from line to line with constant mutual Offset (δχ) provided individual elements (D, to Dj.) leads and e) the line grid (s) resulting from the query at certain points in time in an electronic evaluation after a corresponding delay, is (are) nested with the grid of the following line (s). 2. The method according to claim 1, characterized net that the radiation energy or corresponding signals for predetermined spatial frequencies are filtered before and after the screening is carried out. 3. The method according to Aj, ^ uch 1 and 2, characterized in that that a) for a two-line detector, the first line raster to briefly cached in buffers (BI) before transferring the second line grid and then in a shift register (CCD I) (= charge coupled device), i.e. a delay is achieved, b) the second line raster is temporarily stored in further buffers (B II) shortly after it has been taken over by the detector arrangement and is transferred to a further shift register (CCD II) and c) the two grids by alternately serial reading out the associated CCD shift registers are nested and used as a monitor line, with the transfer and takeover pulses fall in the time segment of the monitor electron beam return (Pig. 1 and 3c). 4. The method according to any one of the preceding claims, characterized in that a known TDJ operation (= time delay and integration) is carried out in the scanning direction, according to which the scanning speed is adjusted so that in the line duration t, the vertical distance Δν _{η7 is moved back} from the detector line of a line block (I to IV) to the corresponding line of the next block, and from a line trigger point to the next m signal transfers are carried out (FIG. 6). 5. Detector arrangement 2 to carry out the method according to a
of the preceding claims, characterized in that the individual elements (D ₄ to D _U J to hori- 1 NX zontal and parallel detector lines
(I to η) expanded and these lines each with a constant
Are offset from one another. 6. Detector arrangement according to claim 5, characterized by an offset (δχ) of the detector lines (I to π) the <is as half the reciprocal of the predetermined one Spatial frequency. 7. Detector arrangement according to claim 5 or 6, characterized in that the line rasters are in an in- J internal CCD (= charge coupled device) - shift register or in I. I can be buffered in an external shift register. § 8. Detector arrangement according to claim 7, characterized in that in the electronic evaluation | tion of each detector element (D ₁ to 0 _Νχ ) a ^f speaking detector signal receiving main memory (H) to-; is arranged, between the individual main memories of a | Line an intermediate I serving for charge transport memory (Z) and between the detectors and their trains- I arranged main memories at least one also the intermediate j Storage buffer (B) is provided, where all | Memory in a common summer (σ) open (Fig. 3b, i 4, 5). S. 9. detector arrangement according to claim 8, characterized in that each between the detectors (D ^ to
Dm) and the associated buffers (B) on the one hand and between
these buffers (B) and the associated main memories (H)
on the other hand, an electronic switch (Sch) is provided for each can be seen (Fig. 3b). 10. Detector arrangement according to one of the preceding claims, characterized in that with detector lines In CCD-FPA technology (= charge coupled device-focal plane array) additional delay elements (Ia to HIa) are internally available the focal plane are provided with the help of CCD shift registers or externally (Fig. 4). 11. Detector arrangement according to one of claims 1 to 9, characterized in that in the case of individual detector elements (D, to Dj.) In conventional technology, the electrical connections are led out of a Dewar vessel surrounding the detectors and are connected to the respective first of the delay elements ( I; II. * III _a ) are electrically connected in parallel feed (Fig. 8). 12. Detector arrangement according to claim 10, characterized in that the internal solution for the serial Transport of the line grid with η detector lines the line I das (n -1) times, line II (n -2) times and so on for additional Has registers (Fig. 4). 13., detector arrangement according to claim 10, characterized in that that between the detectors (D ^ to D ») and the respective associated main memory (H) connected in series and locally a continuously increasing number from line to line of temporally offset transporting buffers (B) is provided (Fig. 5). 14. Detector arrangement according to claim 10 and 11, characterized in that that for the external solution, the delays from detector line to detector line with CCDs, are commercially available analog delay lines, shift registers or RAM memories (= random access memory). 15. Detector arrangement according to one of the preceding claims, characterized in that the detector center distance between the individual detector rows is local constant either twice the detector height or one whole Numerous multiples η and time of the TV line duration or corresponds to an integral multiple η thereof. 16. Detector arrangement according to claim 15, characterized in that the delay of the line raster in the individual lines with η = 1 and η = IV (n _y -1), (n _y -2), (n _y -3) or with η 1 η (η -1), η (η -2), η ( n _y -3) times the monitor line duration. 17. Detector arrangement according to one of the preceding claims, which is characterized in that it is not is rotated by 90 ° in the TV standard working monitor.