US2928949A - Method of and apparatus for locating an identification mark attached to a written orprinted item - Google Patents

Description

March 15, 1960 s-r m ucH 2,928,949

METHOD OF AND APPARATUS FOR LOCATING AN IDENTIFICATION MARK ATTACHED TO A WRIT-TEN 0R PRINTED ITEM Filed July 5, 1957 3. f i A v F I g. I

! army METMRA GINO A??? wmasu 4 I y w F I g. 2

a/m/o mrPuF/ew Mama REG/f7 R I 2 3 I I 6 F I g. 3

INVENTOR ATTORNEY METHOD OF AND APPARATUS FOR LOCATING AN IDENTIFICATION MARK ATTACHED TO A WRITTEN R PRTNTED ITEM Karl Steinbuch, Fellbach, Germany, assignor to International Standard Electric Corporation, New York, N.Y., a corporation of Delaware Application July 5, 1957, Serial No. 670,111

Claims priority, application Germany July 13, 1956 3 Claims. (Cl. 250-206) oddly. In such a case the double pulses of adjoining lines do not start simultaneously with respect to the beginning of the item, but are staggered from lineto line, whereby however, the inset points will lie on one straight line when they indicate an identification mark.

In the following, the invention will be particularly described by way of example; as an item, there was been chosen a letter, on which the location of the postage stamp is required to be determined. Fig. 1 shows the the position of mail items is necessary in the course of processes in which the items have to be sorted in accordance with their respective places of destination, because they are generally capable of assuming one of four different positions, and the sorting can only be performed when the mail items are in a proper readable position.

Various methods have already been suggested for sensing and locating the postage stamps, which are aflixed to letters, by a mechanical scanning operation. Thereby, in one such known arrangement, the letters are e.g. led past a photo-electric scanning device, in which the scanning beam sweeps forwardly and backwardly vertically in relation to the direction in which the letters travel. As soon as the identification mark (stamp, etc.) is hit by the scanning beam, there is produced in the scanning device, e.g. in a photocell, an intensity fluctuation, which may then be used for characterizing the identification mark.

These intensity fluctuations of the photocell current may also be produced when the photocell is stationary and the mail item is continuously led past the cell. However, the evaluation of the intensity fluctuations in this way appears to be somewhat unreliable, because on the mail items there may also be other markings or signs (such as remarks regarding the address or sender, or the like), and which, under certain circumstances, may easily cause a similar intensity fluctuation as the identification mark itself.

An object of the present invention is the avoidance of faulty evaluations which may be caused by such other signs, etc. According to the invention, the items are scanned by means of one or more photocells respectively, and the photocell signal, which consists of impulses corresponding to the brightness variations on the front surface of the item, will be examined as to the presence of a double impulse, the pulse spacing of which corresponds to the width of the identification mark. The present invention results from the recognition that with the known methods of evaluating the difference of the direct-current portion during the scanning between the item and the identification mark, certain dark-colored letters etc. can not be reliably handled. Thus, for instance, with some letters the intensity diflerence between the dark'colored envelope and the postage stamps is very small. According to the invention, therefore, the edges or margins of the identification marks are evaluated, whereby, with a suitable design of the evaluating arrangement, there will be produced a double pulse with a de fined space between the two pulses, which may then be used for identification purpose.

In some cases it may be appropriate to carry out the scanning line-wise, that is, the scanning is effected simulintensity fluctuations of the photocell current, which are likely to appear on the scanning of a dark envelope. The two large negative pulses correspond to the edges of the postage stamp. The appearance of such a double pulse is registered by a suitable arrangement or device. Preferably there may be employed a timing-gate circuit, which only produces a control signal, or permits the generation of such a control signal whenever two pulses of about the width of the postage stamp pass through the gate circuit. In Fig. 2, there is shown such a timinggate circuit, which would meet this requirement, and in Fig. Ban alternative arrangement.

' Referring to Fig. 2, the photocell signals coming from the photocell 1 are fed via the amplifier 2 to the bandpass filter 3. This filter is so dimensioned that pulses, which correspond to the normal width of the stamp edges, are permitted to pass as strongly and as undisturbed as possible. The first pulse is fed at the same time to the gate 4 and to the delay network 5. The gate 4 is a coincidence gate having two inputs and requires the concurrent application of pulses of said inputs to cause it to open. At first the gate 4 is still closed, so that the first pulse is prevented from passing. In the delay network this first pulse is so delayed that the delayed pulse and the second pulse of the double pulse appear simultaneously at the gate 4, so that it will now open and, consequently, permit the second pulse to pass. The delay network is tuned or adjusted in such a way that in the delay time there are contained all stamp widths that are likely to occur. Of course, the delay time also depends on the speed at which the letters travel past the scanning device.

The delay network may be of any known type and may consist of e.g. two mono-stable flip-flops which are arranged in series. The first one of the flip-flops is excited by the first pulse and is thereby caused to tilt into its unstable position. After a delay determined by its time constant, it is tilted back to the stable position, and thereby, causes the second flip-flop to be tilted into its unstable position which also, after its respective time constant, is tilted back. While the second flip-flop is in its unstable position, the gate 4 is open. The time constants are so determined that there will also be obtained a widening or enlargement of the pulse. This is necessary because different widths of postage stamps are in use, and because the gate 4 is supposed to be reliably opened to all such widths. In Fig. 3 there is shown an arrangement differing from that of Fig. 2 in that there is provided an additional arrangement for eliminating the writing in the scanning results. Due to the fact that the scanning is effected with the aid of a gap-type photocell, also all signs etc., such as hand-writing or the like, will be scanned and fed to the evaluating device. Thereby it may happen that in the writing there appear two pulses which are spaced apart each other at a relation lying within therange'of the'stamp width, and which may be so strong that they actually 'can'energize' theevaluating" device. The arrangement'for excluding the writing is now based on the experimental fact that the pulses generated by' thewriting are greaterin number" than those pulses which are generated by the stamp image.

The signals as coming from theband-pass'filter 3-of Fig: 2"are fed also to the full-Wave rectifier-*6, Fig. 3, and to the low-pass'filter 7. In this way,1th'e'inco ming signals are integrated. Upon reaching a'certain value the second gate 8 will be blocked. The gate 8 is agate of known-typehaving anenabling input 'and'an inhibiting input; The input of gate 8 which is coupled to the output of gate 4 is the enabling input and the input to gate'8 which is coupled to the'output of the filter7 isfthe inhibiting input. This value is determined in such a manner .that the integration. of 'the signals 'producedby the scanning of the'postage stamp'will be insufficient'for blocking or inhibitingthe gate, while the integration'of the writingsignals for the width of apostage stamp will be suificient to efiect such blocking. Thus only a double pulse, corresponding to the edges of the stamp, is'capable of transmitting asignal via both gatesii and'S' to the indicating device, in which the recognitionof a posta'ge stampis determined accordingly.

While I'have described above the :principles of my invention in connection with specific apparatus, it is" to b'e'clearlyunderstood that this description is made only b'y'way-of example and not as a limitation to the "scope of my invention as set forth in the objects thereof and inthe accompanying claims.

What is claimed is:

1. Apparatus for determining the position of .a written or printed item with respect to an identification mark applied thereto, for example, of a letter. with respect to thepostage stamp comprising light sensitive means for V 4 scanning a surface of said item, means for deriving a wave-form linesignal from said scanning means, gate means, time delay means intermediate said means for deriving said line signal and said gate means for delaying said signal a time approximately corresponding to the time taken to scan the Width of said identification mark, means for integrating the signals derived from said means for-deriving said line signal, said integrating means coupled to said gate means whereby signals corresponding to the edges and the area lying between of said identificati'on rnark' are ofinsufficient value to prevent opening of said gate means upon application of said edge pulses through saidv delay means, but signals derived from extraneous'marking on said item are of sufiicient value to inhibit :opening of said gate means.

2. Apparatus as claimed in claim 1 wherein said gate means comprisesa' first and a second electronic gate, said first gate comprising a two-input coincidence gate having-an-output, said second gate having an enabling input and"an'inhibiting input, the inputs of said first gate coupledto said means for deriving a wave form line signal and'said time-delay means, respectively, theenabling'input of said second gate coupled to the output of said'first gate and the'inhibiting thereof coupled to the output of said integrating means.

'An apparatus according to claim 1 wherein said integratingmeans comprises a full-wave rectifier and a lowpass filter'connected in series.

References Cited in the file of this patent UNITED. STATES PATENTS 2,616,983 Zivorykin Nov; 4, 1952 2,760,064 Bell Aug. 21, 1956 2.7771059 Stites Jan. 8. 1957 2,824,958 D'unn' Feb. 25, 1958

Images