Background of the Invention
The invention relates to an evaluation circuit and, in particular, to a device for determining the point of passage of a moving object through a measuring plane.
Such an evaluation circuit can be used, for example, in a device for determining the position of a point on a ballistic curve within a given measuring plane as disclosed in German Offenlegungsschrift No. 2,402,204. In this device, for example according to FIGS. 5 and 6, two cameras are arranged at a known distance from one another and are inclined with respect to one another. The cameras are each provided with a row of photodetectors in their image planes. The signals from the individual detectors are processed in an evaluation circuit (not described in detail in the patent) and are then fed to a computer for determination of the position of the passage of a projectile through the measuring plane.
The detection, for example, of shading of one or a plurality of adjacent detectors in a row as a result of passage of the projectile--it is here assumed that the measuring plane is illuminated--is difficult when the illumination changes.
It is the object of the invention to provide an evaluation circuit which makes the compilation of such detector signals more reliable even under changing external environmental conditions.
In accordance with the invention, apparatus is provided for evaluating signals generated by first and second spatially separated rows of photoconductors positioned with respect to a measuring plane. The outputs of the photoconductors in each of the rows is scanned by a sampling circuit at a predetermined rate, and the outputs of each sampling circuit coupled to at least one evaluation circuit. Each of the evaluation circuits includes an analog-to-digital converter coupled to the output of the associated sampling circuit, and a pair of digital delay circuits having their inputs alternately coupled to the output of the A/D converter. A comparator is provided for comparing a threshold signal generated by a reference generator with the outputs of the digital delay circuits, the comparator generating a digital signal having a value corresponding to the difference between the ratio of the signals at the outputs of the delay circuits and the threshold signal. A computer coupled to the outputs of the comparators in each of the evaluation circuits determines the coordinates of an object traversing the measuring plane.
In accordance with the invention, which evaluates the difference between successively scanned signals from the individual detectors, the reliability with which a momentary event to be evaluated is detected is much greater. A delay of n.multidot.T can be effected in the analog region by means of an analog shift register, for example, a so-called bucket chain circuit. Or, the signals can be converted to digital signals which are then stored temporarily. Due to the same signal influence in both channels, it is of advantage for the two signals being compared to also convert the directly used signal and to then reconvert it (D/A conversion).
A further improvement in signal evaluation is realized, according to a further feature of the invention, in that the series of analog signals obtained from the detectors are fed to a peak value sensor and the output signal of this peak value or average value sensor is used as a comparison value to which the signals from the detectors are related (standardization). This relating can be effected directly in the analog/digital converter.
Since the rows of detectors, e.g. so-called diode arrays, preferably include a large number of individual detectors, a further feature of the invention provides that a plurality of, e.g. four, partial evaluation circuits are provided in parallel and their inputs are connected to the individual detectors of detector groups. If four partial evaluation circuits are used, one partial evaluation circuit is connected in succession, for example, with the first, fifth, ninth etc. detector of the row, another partial evaluation circuit is connected in succession with the second, sixth, tenth etc. detector of the row, etc. In that way, it is possible to either lower the sampling frequency or increase the frequency of sampling. The sampling frequency may lie at several MHz. However, it is also possible to associate, for example, the first 256 detectors with the first partial evaluation circuit, the next 256 detectors with the second partial evaluation circuit, etc.
If the number of partial evaluation circuits employed is a, their a simultaneously appearing signals must be converted into consecutively arriving signals. For this purpose, short-term memories having different lengths of storage times are provided in (a-1) partial evaluation circuits so as to be able to line up the signals one after the other within the available clock pulse time. The result is a sequence of signals whose appearance depends on the partial signals (0 or 1).
For the further evaluation of the n signals thus obtained, these signals are fed to a memory having n memory locations. There also is provided a counter which counts from 1 to the highest number of the n memory circuits or of the number of detectors in a row. If a change in light occurs in the measuring plane, e.g. shading by, for example, a projectile, several juxtaposed detectors experience a change in the impinging light intensity. Thus their signals are changed and the evaluation circuit generates different signals (e.g., a 1 each time)--as described above--at the locations associated with these detectors compared to the locations associated with the other detectors. The above mentioned memory circuit in conjunction with the counter now records the counting position of the counter at which this signal change occurs and the position at which it disappears again. Thus, it is determined, for example, that the signal change occurs at counter position b (of n possible counter positions) and that it disappears again at e.g. (b+10). It has now been determined that detectors b through b+10 of the row have noted shading, for example.
If--as mentioned above--a measuring plane is monitored by two cameras each having such a row of detectors and if the above-described evaluation circuits are connected thereto, the results obtained for the two rows of detectors are such that the coordinates of the passage point can be calculated according to known relationships.
Brief Description of the Drawings
FIG. 1 is a block circuit diagram of a device designed according to the invention for determining the coordinates of a passage point through a measuring plane monitored by two cameras;
FIG. 2 is an embodiment of the evaluation circuit included in FIG. 1;
FIG. 3 is an embodiment of other blocks of FIG. 1.
Description of a Preferred Embodiment
In FIG. 1, the numerals 1a and 1b identify two rows of photodiodes, e.g. so-called diode arrays, each including, for example, 1024 individual diodes, associated with two cameras (not shown). These rows of photodiodes are connected to sampling circuits 2a and 2b with which the individual diodes are scanned successively. In order to lower the sampling frequency, which is derived from the frequency of a clock pulse generator 3 (e.g. 16 MHz), sampling circuits 2a and 2b each simultaneously sample four juxtaposed diodes, and their signals are fed separately to amplifiers 4a.sub.1 to 4a.sub.4 and 4b.sub.1 to 4b.sub.4, respectively. With successive clock pulses, amplifier 4a.sub.1 is connected, for example, with diodes 1, 5, 9, . . . 1021 and amplifier 4a.sub.2 is connected with diodes 2, 6, 10, . . . 1022, etc. The sampling frequency need then be only one fourth as much (e.g. 4 MHz).
Amplifiers 4 are followed by partial evaluation circuits 5a.sub.1 to 5a.sub.4 and 5b.sub.1 to 5b.sub.4. Their structure will be described below in connection with FIG. 2.
FIG. 2 shows the details of one of the evaluation circuits 5. A terminal 20 is connected, for example, with amplifier 4a.sub.1, and terminal 21 is connected with generator 3. Thus the analog signals A from the sampled diodes are supplied in succession to terminal 20. The individual diodes themselves already generate considerably different signal amplitudes from the same light influence. Changes in the illumination and aging phenomena in the diodes also change the output signals. Nevertheless, a change in intensity produced, for example, by shading, must be recognized as such and evaluated at each individual diode. The circuit of FIG. 2 is provided for this purpose.
The analog signals A at terminal 20 are fed to an analog/digital converter 22. This module 22 includes a quotient former. For this purpose, the converter module 22 additionally receives the signal from a peak value sensor 23 which feeds a signal A' corresponding to the peak value of successive analog signals to module 22. There, the quotient A/A' is formed in order to suppress changes in brightness and then the resulting value is digitalized.
The output signal from converter 22 is reconverted to an analog signal in digital/analog converter 24a and is fed simultaneously, via a switching device 25, to one of digital memories 26a and 26b comprising 1024/4=256 memory locations. Memories 26a and 26b are delay circuits consisting of digital shift registers.
Switching device 25 causes the 256 digital signals to enter into, for example, shift register 26a during a sampling cycle of the diode array, shift register 26a simultaneously feeding its previous contents to a digital/analog converter 24b. In the next following diode sampling cycle, switching device 25 then feeds the output signals of converter 22 to shift register 266 whose previous contents are transferred to digital/analog converter 24b. In this way it is accomplished that at the outputs of converters 24a and 24b there are present analog signals B and B' which were derived from the same diodes but were obtained there at two points in time offset by twice the sampling cycle (2 T). The selection of 2 T was made to assure that a signal obtained in one sampling cycle and to be evaluated is compared with a (previous) signal state during which no signal to be evaluated appeared. If, for example, shading has only a brief effect in one cycle (e.g. only during 30% thereof), the signal may possibly not be sufficient to exceed the threshold. During the next sampling cycle the period of dwell is then assumed to be sufficient. If the previous signal were used as a reference (comparison) the threshold would again not be exceeded. However, by referring to the signal that appeared 2 T earlier, the threshold is exceeded. After detection of the event, evaluation is blocked for a period of time.
Comparison circuit 27, which functions as a comparator, now forms the quotient of output signals B and B'; then, this quotient is compared with a threshold which may be set via a device 28 and, depending on whether the threshold was exceeded or not, a 1 or a 0 signal is generated which is available at terminal 29. A sequence of 256 signals are obtained at terminal 29 during one sampling cycle T of which each signal is a 1 or a 0.
The detectors must have a certain amount of memory retention to assure that a change can be detected. This memory retention must be dimensioned in such a manner that a change can still be detected if the sampling of one diode takes place offset by almost T with respect to the occurrence of the event.
Output terminal 29 of the evaluation circuit 5a.sub.1 and correspondingly also the output terminals of the other partial evaluation circuits are connected with a first signal storage block 6a (FIG. 1), in which successively appearing signals are generated from the four parallel occurring signals from circuits 5a.sub.1 to 5a.sub.4. This will be explained in connection with FIG. 3 in which the four input terminals of block 6a are marked with the numeral 30. The signals are separately fed into memory 31 and are called out at time intervals of 1/4 of the clock pulse time .DELTA.T (e.g. .DELTA.T=1/4.1/4 MHz). In this way, a signal sequence is produced in output line 32 at the clock pulse frequency of 16 MHz with consecutive signals--in the correct sequence--being derived from the output signals of adjacent diodes.
Block 6a may also include a gate 33 which, by actuation at terminal 34, suppresses the signal in line 32, for example if in block 6b no change in signal occurs in the signal sequence at the same time, i.e. the second camera did not record a corresponding event.
The output of block 6a is connected via a line 7a with a further block 8a in which the signal changes are associated with the diodes of the row. According to FIG. 3, this block 8a includes two memory devices 35 and 36 as well as a counter 37, which counts each time (at a clock pulse frequency of 16 MHz) from 1 to n (1024) and begins at counter position 1 with the beginning of sampling of the first diode of the array. If, for example a "1" appears for the first time during one cycle T at the output of gate 33 or in line 7a, respectively,--let it be assumed that, for example, shading produces such a signal in the shaded diodes--then the momentarily existing counter position Z.sub.min of counter 37 is retained in memory 35 (e.g., Z.sub.min =510). If this "1" signal ends after a few clock pulses, e.g. at counter position 520, this second position Z.sub.max is retained in the memory due to inversion in inverter 38. This result indicates that shading was detected from the 510th to the 520th diode. These counter positions Z.sub.min and Z.sub.max are available as digital values at terminals 39 and 40.
The digital values are fed--as shown in FIG. 1--via line bundle 9a and correspondingly for the other camera via line bundle 9b to a computer 10 which is able to calculate according to known relationships the penetration point in the measuring plane from the known distance of the cameras and their known angles of inclination as well as from the measuring values Z.sub.min and Z.sub.max and the corresponding values from the other channel (line bundle 9b).