DE3612901C1 - Processing circuit for signals from an IR detector - Google Patents

Description

The invention relates to a processing circuit according to the Ober Concept of claim 1 or claim 2.

DE-OS 25 10 474 is a multi-element radiation detector for a target-seeking institution known to target discrimination due to different radiation wavelengths of the individual radiation sources recorded simultaneously. This is supposed to a stray emitter to be hidden, which is a known and from Interesting real target has a different radiation frequency. For this the amplitude ratio of the different beams frequency frequencies selected behind frequency-selective reception channels tet. The signal processing problem preceding defined (point-shaped) radiation sources for their spectral values evaluation first of all from the ambient radiation (clutter) sepa This publication does not deal with the need to hurt.

In the multi-element infrared detector for detecting low-flying Target objects according to US Pat. No. 4,209,699 are provided by one Rough location, in the search phase with a large distance range to switch to a more precise location because it is at a greater distance the target object is relatively small in relation to the geometric one Resolution of the sensor. The search area is relatively sharp limited sectors assigned to the individual detector elements divided. If a target object is detected in a sector, switched it to the display device; at the same time the amplifiers arranged behind the detector element to a larger one Bandwidth switched when the reception level is a predetermined  Threshold exceeds what is the criterion for a larger approximation of the target object is taken. Also for this distance-dependent Change in bandwidth of the individual, different emitters assigned reception channels is left open, like the actual one Target object as such can be discriminated if it is rela tiv little different from ambient radiation (clutter).

As is well known (see the older patent application P 34 40 843.6) provides the scanning of a scanned area in accordance with this Radiation and radiation reflection conditions a very strong fluctuating signal at the output of each element of a multiple infrared detector, which occasionally has steep signal peaks overlay if there are also high-energy spotlights on the terrain be recorded. These are the only interests objects in the case of a search fuse ammunition with such a detector. For the acquisition of one Real target in the ambient radiation of the captured simultaneously Adaptive filtering and threshold evaluation take place on site the course of the signal amplitude (cf. the older application P 34 40 749.9). Because only those parts of the delivered by the detector Signals are intended to be used to obtain search detonator ignition information who are most likely to be alone in interesting real target are assigned. Among other criteria they also differ in terms of time Course of their amplitude significantly from those signal components, from the terrain of the target environment as the so-called Clutter-Sig nale are generated.

Therefore, the present invention lies before the actual analysis a real destination signature - based on the task with as little as possible equipment and circuitry the evaluation possible speed of the detector signals to improve noticeably that A target signature is provided for obtaining the ignition information in which the signatures of hot spot radiators (typical real targets)  as little as possible falsified by the surrounding Ge country radiation (the so-called clutter) is offset in terms of amplitude appear. The aim is a noticeable signal / clutter improvement in real time processing and with no significant change in point radiator signature can be achieved, so that possibly the circuitry Effort in the further evaluation circuit (for real target acquisition) can be reduced, but in any case the effectiveness of a smolder len evaluation is increased.

This task becomes more generic in a processing circuit Art solved according to the invention in that it according to the labeling part of claim 1 or claim 2 is designed.

After these solutions there is no direct (linear) processing processing of all detector element output signals; there is still one Limitation of signal processing to what is currently the strongest - So obviously on a spotlight and therefore probably on one target to be traced back - all currently detected detector element output signals. Rather, one will not linear signal preprocessing just behind the detector elements in such a way that an extreme signal (in particular that a hot spot radiator to be assigned to all as a target object output signals currently supplied by the detector elements) is set in relation to a reference information, which from the others (including or excluding the extreme signal) Output signals is obtained; where the extreme value signal is none must be a singular signal, but as an averaged amplitude of several extreme detector signals can be used.

The invention is based in particular on the knowledge that a single large-area detector element is not very suitable is within a larger radiating one detected by the detector (i.e. clutter generating) surface a small area radiator  (a so-called hot spot). Because even with very large temperature difference between the individual spotlights and the surrounding surface is the integral of that of the spotlight radiant power introduced far below that Performance contributed by the entire large area.

As explained in more detail below (in connection with FIG. 2), the signal having the maximum amplitude at the moment is generally evaluated as the extreme detector element output signal, that is to say is set in relation to the environment (clutter) information. If the target area is very close, that is to say if the detector is at a very short distance from the point-shaped radiating target object, the detector optics result in a correspondingly larger projection of the point radiator into the plane in which the detector elements are arranged. Therefore, several adjacent elements are then excited by the radiation from a single spotlight. When the remaining distance to the target object becomes smaller, it is therefore more expedient than to evaluate the extreme signal not as a maximum amplitude, but rather to set the detector element output signal, which currently has a minimum amplitude, in relation to the surrounding signal. This is because when the number of detector elements detected by the spot radiator is no longer small compared to the total number of all elements present, the mean value of all detector element output signals approaches the level of the maximum signals attributed to the spot radiator. Therefore, the previously considered difference formation is no longer sensible, because the high reference voltage resulting from the averaging would also greatly reduce the target signature. In this case, however, a switchover to direct utilization of sum information (for example the mean value signal recorded in any case about all elements), which is output as the target signature to be further evaluated, may be appropriate; as long as a difference evaluation of the mean value with the minimum signal of all detector elements results in a more advantageous target signal-clutter ratio.

For finding the currently extreme (maximum or minimum) Amplitude detector element output can be the known arithmetic circuits for the analysis of digitized Signalver runs can be used. However, in order to reduce the circuitry Keep limits and falsify signatures before the actual one To avoid target signature analysis if possible, it is more appropriate a real-time analog signal ver. directly behind the detector elements work to separate out the current extreme amplitude Channel. For this there are amplifier-diode circuits previously known as such (see e.g. U.S. Patent 3,092,732).

In order to distinguish the signal hot spot signal more clearly from the clutter, is basically a subdivision of the detector as strong as possible to strive for adjacent individual elements. At the front lying signal preprocessing (from this detector sub-part resultant single element output signals) not a measure to detect the location of the point Strahlers in the currently recorded area (as in the older application P 34 10 942.0 mentioned) - such an evaluation can, however advantageously connect after the spotlight information under Application of the present solution was more clearly discontinued by the clutter. In the same way this is suitable - due to the present signal preprocessing better off the clutter - target signature especially good for wrong target temperature discrimination (see the older application P 35 22 456.8). Above all, however, is due to the larger distance between Target and clutter amplitudes a threshold evaluation for the De Definition of a moving trigger window easier.

As the environment information generating the clutter, the Signal preprocessing which, in the sense of averaging sum of all current detector element output signals are drawn, i.e. the sum of all current signal amplitudes divided by the number of signal amplitudes taken into account (Detector elements). If the sum signal so averaged from the moment strongest, that is to say attributable to a hot spot heater gate element output signal is subtracted for the  further target analysis the signature, which greatly improved ver ratio to the clutter amplitudes.

It has been shown that a further improvement of this S / C ratio is achievable if in from the mean education information that is currently maximum (i.e. all Probably due to a real target) detector off signal is not included. The averaging will then limited to those electrical signals from the other detector elements (which are currently do not carry a maximum output signal). Because that avoids in the resulting target signature an S / C loss versus best single element output signal: the result of subtractive Linking is now also not a fraction of the current Spotlight signals weakened more. This further earnings ver improvement, which is of the order of 3 db (compared to the signature with averaging including the extreme value) makes, however, requires increased circuitry or Data processing effort because determined separately must be which of the detector elements currently the maximum ampli tude delivers (which are excluded from averaging should). If, on the other hand, this further improvement in results (through Exclusion of the current maximum from the sum or mean bil dung) can be dispensed with, the result is an extremely simple Real-time analog circuitry for the S / C ver Improvement through different differentiation variants.

Additional developments and advantages of the invention result from the following description of the drawing with restriction highly abstracted to the essentials as a block diagram Typical implementation examples for solutions according to the invention. It shows:

Fig. 1 behind the detector elements, the signal pre-processing for obtaining a hot spot target signature with improved signal-to-clutter ratio; taking into account a switchover option for closer target approach;

Fig. 2 S / C ratios for different target signature wins, in qualitative dependence on the target distance; and

Fig. 3 in a modification of the processing circuit outlined in Fig. 1, an example of obtaining a distance-dependent optimized target signature.

A thermally point-radiating target object 10 is to be identified by means of an infrared detector 11 , which is, for example, part of a detonator ammunition, in terrain 13 scanned in the form of stripes by recording its thermal radiation 17 . The radiation is projected via an optical system 18 as a thermal image into a detection plane 20 which is provided with a plurality of detector material elements 21 (21.1,..., 21.i,..., 21.n), which are preferably arranged next to one another in a straight line. is equipped. Each of these elements 21 delivers an electrical signal 22 in accordance with the intensity of a thermal excitation.

An ignition information 36 is to be obtained from the signals 22 in an evaluation circuit 23 if the object 10 is actually a target to be combated. For this evaluation, it is desirable that a target signature 41 in the preprocessing 61 of the signals 22 be leveled as little as possible or even falsified, on the other hand clearly differentiated from the clutter. Since a point-shaped emitter, in the form of a target object 10 , produces a point-like image in the projection onto the plane 20 compared to the extent of the area 13 detected by the detector 11 , the plane 20 is with as many elements as possible, as small as possible and arranged close to one another 21 equipped. The entirety of all output signals 22 is, if necessary after level increase in preamplifiers 33 , given to the inputs of an averager 50 , which can in principle be constructed as a summing amplifier 52 with a reduction circuit 53 corresponding to the number n of input signals 22 summed up. The contribution of the punctiform radiator is therefore largely lost in the mean signal 54 ; because the surrounding area 13 , which radiates much more weakly, extends over a substantially larger area than the point emitter, which in the integral leads to considerably higher radiation intensity.

As has been shown, a subtractive link 62 of the mean signal 54 provides only that signal 22 . i, which is assigned to the point source 10 of the target 10 (and therefore has a stronger intensity than all other detector element signals 22 ), with a target signature 41 which is unadulterated in relation to the ambient or mean value signal 54, a very steep and high level rise which stands out from the rest of the signal , which is therefore relatively easy to detect in the evaluation circuit 23 in order to be able to carry out the real target detection; namely, for example, via a characteristic curve temperature evaluation according to patent application P 35 22 456.8 from June 22, 1985 and / or via digital filtering and threshold value formation according to patent applications P 34 40 843.6 and P 34 40 749.9 from November 8, 1984 by the applicant.

To obtain the signature 41 , an extreme value transmitter 63 (in FIG. 1 a maximum transmitter 63 max) only switches through the currently maximum point source signal 64 max according to FIG. 1. It does not matter which of the detector elements 21 the radiation 17 of the hot spot target 10 is currently projected on; since no geometric filing information is to be obtained in the signal preprocessing 61 , but only the information, which is unadulterated from the terrain 13 relative to the surrounding radiation, according to the signal 22 of the currently most intense radiation excitation.

The extreme value transmitter 63 can simply have an interconnection of operational amplifiers 66 connected downstream of the individual detector elements 21 ; in such a way that (for maximum signal detection) the amplifiers 66, as entered in FIG. 1, are driven via their direct input (+) with the respective current signal 22, are decoupled on the output side by diodes 65 and are then fed back to all inverting inputs (-) , Since, by definition, the same voltage is present at the output of an operational amplifier 66 in the usual connection for the gain factor "one"; as at the input, and since all inverting inputs (-) connected in parallel now carry the same voltage, the diode 65 is forward-biased only behind that amplifier 66 , in which the voltage of a detector signal 22 applied to the non-inverting input (+) is greater is, as in all other operations, amplifier 66 - with the result that only this strongest of the detector signals 22 is switched through as the point emitter signal 64 max by the extreme value transmitter 63 max.

According to Fig. 1 61, the signal-Uorverarbeitung a link 62 for forming the difference between on the one hand this insulated hot-spot signal 64 max and on the other hand (as a reference potential in accordance with the current total background radiation) the mean value signal 54. Thus, the hot-spot signal is maximum practical output 64 without being attenuated by the link 62 as the interest for further evaluation circuit 23 target signature 41, that is, in an analogous real-time processing without significant change in the hot-spot signature 41 a substantial s / c improvement achieved. In principle, the polarities at the inputs of the operational amplifier 66 in the extreme value transmitter 63 and / or at the inputs of the subtractive link 62 can also be reversed, which likewise brings about a signal / clutter improvement for the target signature 41 . However, the most favorable conditions for larger distances d to the target object 10 result from the polarities entered in FIG. 1 and explained above.

Basically, the formation of the difference with the mean value signal 54 results in a damping of the target signature 41 of interest. This becomes larger if, due to the decreasing distance d, the radiation 17 of the target object 10 is projected onto a plurality of adjacent elements 21 via the optics 18 . The mean value signal 54 increasingly approaches the maximum signal 64 max, the more elements 21 are irradiated when the target object 10 is imaged in the detection plane 20 and therefore make strong contributions to the averaging, ie the signal-clutter ratio increases to smaller ones d values. The difference provided by the link 62 is correspondingly smaller, as can be seen in FIG. 2 by the upper curve shape towards smaller d values. In order to avoid that the target signature 41 is lost when the detector 11 approaches the target object 10 , a switching device 67 is provided in the basic illustration according to FIG. 1, which finally prevents the formation of a difference (in the linkage 62 ) and z. B. directly provides sum information from all detector element signals 22 as the target signature 41 to be processed from now on. To control this Umschaltein device 67 can be queried, for example. In the extreme value transmitter 63 , whether a plurality of operational amplifiers 66 for switching their diodes 65 are simultaneously activated (so positive at the polarity shown in the drawing); or it is queried after the mean value generator 50 or after the link 62 (only symbolically indicated in FIG. 1) whether the mean value signal 54 exceeds a predetermined threshold value or whether the target signature 41 falls below a threshold value. To simplify the exemplary embodiment shown graphically in FIG. 1, the switchover device 67 is simply provided as a disconnect switch between the extreme value transmitter 63 and the linkage 62 , so that after the switchover device 67 is opened, the mean value signal 54 (with the opposite sign) as the target signature 41 on the evaluation circuit 23 pending; if necessary, an amplification with the factor n including sign reversal would have to be carried out if the sum of all detector signals 22 is to be processed further in the same relation and polarity as before after the difference-forming link 62 - but the information required about the target signature 41 is also already completely contained in the mean value signal 54 .

The curves in FIG. 2 show that different differences between information obtained from the detector signals 22 lead to a differently steep drop in the useful information amplitude (in the form of the signal-clutter ratio in the target signature 41 ) when the distance d between the detector 11 is reduced and result in the target object 10 . The subtraction of the mean value signal 54 from the detector maximum signal 64 max considered in connection with FIG. 1 produces, at a large target distance d, if the point target 10 actually only irradiates one of the detector elements 22 , favorable conditions for the further evaluation of the target signature 41 . This signal-clutter ratio is somewhat less favorable if the instantaneous minimum extreme value signal 64 min (cf. FIG. 3) is always subtracted from the instantaneous maximum signal 64 max in order to obtain the target signature 41 . Subtracting the minimum extreme value signal 64 min from the background or mean value signal 54 results in the most unfavorable but a constant signal-clutter ratio for a wide distance range, because the minimum detector signal 22 differs significantly less from the mean value signal 54 than the maximum to be assigned to a point target 10 Detector signal 22 . Only at very small detector-target object distances d do larger fluctuations also occur with regard to the minima of the detector signals 22 , and the difference with the mean value 54 then leads to a relatively steep drop in the useful amplitude of the target signature 41 ( FIG. 2, left). As a result of these circumstances, at smaller distances d below a limit distance d ', obtaining the signature from the difference between the mean value signal 54 and the maximum signal 64 max is less favorable than forming a difference between a minimum detector signal 64 min and the mean value signal 54 because then area 68 hatched in FIG. 2 below the crossing point of those two curves which represent the cheapest or most unfavorable target signature 41 for large distances d can now still be evaluated at small distances d.

Recognizing these circumstances, as shown in simplified form in FIG. 3, it is preferably provided below the limit distance d ', at which the conditions mentioned are reversed, that the target signal 41 can no longer be obtained from the difference between the maximum signal 64 max minus the mean value signal 54 , but then from the difference of the mean value signal 54 minus the minimum signal 64 min.

As already explained above with regard to the extreme value transmitter 63 , only the input polarity of the operational amplifiers 66 and the diodes 65 need to be reversed in order to obtain a minimum signal 64 min, and the extreme value transmitter 63 delivers that of the input-side detector signals 22 as the minimum signal 64 min currently has the smallest amplitude. In parallel to the difference formation explained in FIG. 1 by means of the link 62 , the difference is now formed from the mean value signal 54 and the minimum signal 64 min in a further link 62 . Of the thus obtained two target signatures 41 'that is only about a maximum determiner 69 respectively output as processed further target signature 41, the amplitude (and hence the signal-clutter ratio) is greater than the other. This means that for distances above the limit distance d '( FIG. 2) the target signature 41 ' obtained via the maximum transmitter 63 max and below this distance d 'the target signature 41 ' obtained via the minimum transmitter 63 min is output, so that in the course of Reduction of the target distance d the characteristic curve for the target signature 41 , which is fully drawn out in FIG. 2, is traversed.

The maximum determiner 69 can be constructed in the same way as the maximum extreme value transmitter 63 max explained above, only the two input items 41 ′ being to be compared with one another by means of two coupled operational amplifiers 66 ; which, via their output diodes 65, switch through that input variable as the preprocessed target signature 41 which currently has the larger voltage amplitude.

Claims (7)

1. Processing circuit for signals (22) (11) individual with a number of detector elements (21) in an IR detector behind an imaging optics (18) for obtaining the hot spot signature (41) of a real target (10), by signal preprocessing ( 61 ) with a subtractive link ( 62 ) for the output of an extreme value transmitter ( 63 ) and the output of an average value generator ( 50 ), the extreme value transmitter ( 63 ) and the average value generator ( 50 ) each being connected downstream of the detector elements ( 21 ) , 2. Processing circuit for signals ( 22 ) of an IR detector ( 11 ) with a number of individual detector elements ( 21 ) behind an imaging optics ( 18 ), for obtaining the hot spot signature ( 41 ) of a real target ( 10 ) by signal preprocessing ( 61 ) with a subtractive link ( 62 ) for two extreme value transmitters ( 63 ) which are connected downstream of the detector elements ( 21 ). 3. Processing circuit according to claim 1, characterized in that a difference-forming logic circuit ( 62 ) at the nega tive input a mean value generator ( 50 ) and at the positive input a maximum extreme value generator (63 max) are connected upstream, both of the detector element signals ( 22nd ) are fed. 4. Processing circuit according to claim 2, characterized in that a difference-forming logic circuit ( 62 ) at the positive input a maximum extreme value transmitter (63 max) and at the negative input a minimum extreme value transmitter (63 min) are connected upstream, both of the detector element Signals ( 22 ) are fed. 5. Processing circuit according to claim 1, characterized in that a difference-forming logic circuit ( 62 ) at the positive input a mean value generator ( 50 ) and at the negative input a minimum extreme value generator (63 min) are connected upstream, both of the detector element signals ( 22nd ) are fed. 6. Processing circuit according to claim 5 and one of claims 3 or 4, characterized in that the Zielsig naturen ( 41 ') supplied by the logic circuits ( 62 ) are connected to a maximum determiner ( 69 ) for outputting a preprocessed signature ( 41 ). 7. Processing circuit according to one of the preceding claims, characterized in that a fixed assignment of operational amplifiers ( 66 ) to the input signals ( 22 or 41 ') is provided as an extreme value transmitter ( 63 ) or as a maximum detector ( 69 ), each Amplifier output is guided via a diode ( 65 ) to the extreme value transmitter or to the maximum detector output and this is fed back to a respective second input with a different polarity from the first input of all of the amplifiers ( 66 ).