NO813504L - PROCEDURE AND DEVICE IDENTIFICATION DEVICE - Google Patents

Abstract

1. Method of identifying contrasting line patterns on objects which appear in a random position and orientation and at random times on an image window and each comprise a characterising field on a surface facing the image window, which field includes symbols and at least one contrasting line pattern in at least one data track, which pattern characterises the position and the orientation of the data track and has a plurality of parallel lines with different spacings and/or line widths, the image window being scanned line-by-line and a binary video signal is produced which corresponds to the scanned contrasting line pattern and the subsequent symbols, the image window, in the first step of the method, being scanned at different angles until a prescribed contrasting line pattern is identified, in the second step of the method, the position and alignment of the data track relatively to the image window is identified and, in a third step of the method, a raster scan is effected in the direction of the data track and the symbols are read and decoded, with which, for identifying the contrasting line pattern (PIC), the interval lengths of overlapping light-dark intervals (I 1 to I 4) of the video signal are measured and represented in the form of binary counter readings (TC 1 to TC 4), the counter readings (TC 1 to TC 4) address a two-dimensional table, in which are contained an expectation field with criteria for a prescribed relationship of two successive interval lengths of the intervals (I 1 to I 4), the counter reading (TC N) corresponding to an interval length (IN; N = 1 to 4) addressing a line of the table and the counter reading corresponding to the following interval length (I N+1) addressing a column of the table, a signal (LPIC 1, LPIC2 , LPIC 3) is produced when the ratio of the two respectively following interval lengths is within a prescribed value range, and an identifying signal (PIC OUT) is supplied when in each case a signal (LPIC 1, LPIC2, LPIC 3) is produced with a number of successive addressing steps as prescribed by the contrasting line pattern (PIC), characterised in that one or more contrasting line patterns (PICm ; m = 1,2,3...) are scanned, both in the forward direction and the rearward direction, that the counter readings (TC 1m to TC 4m ) for each contrasting line pattern (PICm ) pass a forward expectation field with the forward scanning and a rearward expectation field with the rearward scanning, respectively, and that, in each case, a forward identifying signal (PICm OUT V) and a rearward identifying signal (PICm OUT R), respectively, is delivered when the corresponding contrasting line pattern (PICm ) has been recognised with scanning in the forward direction and rearward direction, respectively.

Description

The invention relates to a method and a device for identifying objects that appear in an arbitrary position and orientation and at arbitrary times on a picture window. On a surface facing the image window, each object displays an identification in the form of a data field which in at least one data track comprises contrasting characters with at least one contrasting line pattern that characterizes the position and orientation of the data track. The data track also has several parallel lines with different distances and/or line widths. The image window is scanned linearly and a binary video signal corresponding to the scanned contrast sequence is produced. In a first step, the image window is scanned from different angles or directions until a contrast line pattern is recognized. In a second step, the position and alignment of the data field in relation to the image window is determined, and in a third step, a raster scan is performed in the direction of the data track to read and decode the characters contained in the data track.

Such methods and devices are already known. The items to be identified are, for example, trade goods, department store items or the like, which are identified in machine-readable form. For this purpose, identifications on which characters in a machine-readable code are printed are fixed or placed on the objects or objects, e.g. The OCR code. Such identification can consist of quality, dimension or price indications, of the article number, etc. This identification data is in one way or another placed on the objects' surfaces.

Such identifications can only be machine-read with difficulty, as the objects differ in size, and as these identifications e.g. on adhesive labels are placed in different places of the object. When reading the identification, it cannot therefore be assumed that it is available at specific times in a specific position with a fixed orientation. The reading of these identifications cannot therefore be compared to the reading of punched cards or the like in which a card is available in a precisely defined reading position at precisely determined times. In the present case, exactly the opposite is the case. The data field on the object only appears more or less approximately in a certain place, and also. the alignment of the data field is relatively arbitrary.

Such methods and devices for the identification of objects are, for example, used at checkouts in supermarkets etc., to enable an automatic registration of the price and/or article number of the object that a customer wants to buy and for this purpose has brought to the checkout area. The goods, such as boxes of different shapes and sizes, bottles, cartons, boxes, etc., are then individually arranged on the image window, as only the surface that carries the data field used for product identification must be directed towards the image window. The data fields on the various objects thus appear with different alignment in different places within the image window. Furthermore, the data fields do not appear at fixed time intervals at the reading station. The reading station must therefore search for the data field and then read out the characters of the data tracks in the direction of the data tracks of the data fields. The characters read out can then be supplied as electrical signals to the till, which can print the price and possibly also the article number or the article group on the checkout tab.

The data field placed on the object is provided with a contrasting line pattern or product identification code (PIC) which comprises several parallel lines with different distances and/or different line widths. The contrast line pattern serves to highlight the data field, e.g. the printed label, in a reliable and unambiguous manner in relation to those

other signs and line patterns that may be present on the object's surface in the data field's surroundings. With regard to the data traces, within the data field, the contrast line pattern also has a predetermined position and orientation and can therefore be used to determine the position and orientation of the contrast line pattern - and thus the data traces - in relation to the scanning angle of the scanning lines, and then in the direction of the data traces to produce a scanning raster as in

vertical direction crosses the characters within the data tracks. The secure recognition of the contrast line pattern thus represents a significant step in the omnidirectional reading of such data fields which are printed with clear characters.

From DE-OS 2 338 561 there is known a method and a device of the nature indicated at the outset, whereby the recognition of the contrast line pattern only occurs when the contrast line pattern is scanned essentially perpendicular to the direction of its lines and the pulse sequence thereby contained in the video signal is equal to a predetermined pulse sequence which corresponds to the contrast line pattern used. It is thus a correlation method. It is a particular disadvantage here that the scanning of the image field for locating or locating the contrast line pattern must take place in very small angular increments until a scanning line intersects the contrast line pattern vertically, whereby the search time is undesirably long. In addition, blurred printed edges of the contrast line pattern's lines make themselves disturbingly noticeable and sometimes prevent a searched contrast line pattern from being recognized as such since the pulse order contained in the video signal deviates from the stored comparison pattern. As a result, either the search sequence is repeated or the contrast line pattern cannot be recognized at all, and the associated data field then remains unread.

The purpose of the invention is to provide a method and a device of the type indicated at the outset which enables fast and reliable recognition of contrast line patterns and which works to a large extent unaffected by blurring at the edges of the contrast lines.

According to the invention, the above-mentioned purpose is achieved by a method of the nature indicated at the outset in that the interval length of overlapping light and dark intervals of the video signal is measured for recognition of: the line of contrast pattern (PIC), that successively measured intervals are compared with each other and a comparison signal is produced with a first amplitude when the two interval lengths being compared are in a predetermined relationship to each other that corresponds to the contrast line pattern's corresponding distances, and that a recognition signal (PIC OUT) is emitted, when a comparison signal with the first amplitude.

A device for identifying objects in accordance with the specified method comprises an opto-electronic scanning device which outputs a binary video signal which corresponds to a linearly scanned image field and consists of a series of light and dark intervals. Such a device further comprises decoder devices for recognizing a scanned contrast line pattern which characterizes the position and orientation of at least one data track of a data field and contains several parallel lines with different distances and/or line widths. The device further comprises devices for aligning the scan raster parallel to the data track and for reading the characters of the data track scanned with the scan raster. According to the invention, such a device is characterized by the fact that the decoder devices contain a counting circuit which receives the video signal and which counts the interval lengths of successively overlapping intervals of video signals, that the decoder devices contain at least one comparison table which, via a gate circuit, receives successively counted interval lengths in pairs and emits a comparison signal with a first amplitude when the two interval lengths compared with each other are in a predetermined mutual relationship that corresponds to the relationship between the corresponding intervals of the contrast line pattern, and that a utilization or evaluation coupling is arranged which emits an identification signal (PIC-OUT) in the case of an uninterrupted generation of a predetermined number of comparison signals with the first amplitude.

The advantages of the invention lie in particular in that the recognition of the contrast line pattern only depends on whether the ratio between successive, mutually overlapping interval lengths of the video signal lies within narrow value ranges, so that the contrast line pattern also in a scan that takes place at an angle in relation to the direction of the contrast lines, occurs in a safe manner as the successively formed ratios between successively measured interval lengths of the video signal are invariant in relation to the scan angle. For reliable recognition of the contrast line pattern, it is thus sufficient to - at an arbitrary angle, continuously - increase the contrast line pattern by which all contrast lines are completely crossed. The search for the contrast line pattern can thus take place in relatively large angular increments, i.e. relatively quickly.

As the decoding of the video signal for detection of the contrast line pattern with regard to video signals must take place in real-time processing, i.e. essentially simultaneously with the production of the video signal, in order to ensure that a contrast line pattern is recognized at any point within the video signal, the determination of whether the ratio between successively measured interval lengths lies within a predetermined range of values takes place, not by means of a division, but by input in a two-dimensional comparison table, also called a diyision table-, which emits a comparison signal with a predetermined first amplitude only when both interval lengths compared with each other lie in a predetermined field of expectation of the table within which the quotient between the interval lengths compared with each other has a value range that corresponds to the value range for the corresponding interval lengths of the sought contrast line pattern. In this way, the time- and hardware-consuming division method is avoided and the real-time operation with regard to the video signal is easily maintained.

The comparison table is preferably designed as two-dimensional non-volatile memory (PROM). The different possible discrete values of the first measured interval length address different lines or rows in the constant memory. The different discrete values of the subsequently measured interval length address different slots or columns in the constant memory. The expectation field is predetermined in such a way that it includes all the storage locations or memory cells in which the quotient between the interval length assigned to the line address and the interval length assigned to the column address lies in a predetermined value range. If a memory cell within the expectation field is addressed, the constant memory emits a comparison signal with a first amplitude which signals a recognition of the contrast line pattern. If, on the other hand, the memory cell determined by line address and column address lies outside the expected field, a comparison signal with a second amplitude is emitted which signals that the quotient between the compared consecutive interval lengths lies outside the predetermined value range. When a comparison signal with the second amplitude occurs, the decoder devices are then reset to zero and are then ready for carrying out a new recognition and decoder step.

Preferably two and two interval lengths to be compared with each other are entered into a separate comparison table with a separate expectation field and there checked with regard to their quotient value. Each individual comparison table is preferably realized as a separate constant memory. (PROM) to simplify the control logic of the decoder devices.

According to a preferred embodiment of the invention, the first constant memory for comparison of the first and second interval length, as well as the second constant memory for comparison of the second and third interval length etc. comprise 8-bit memory positions or memory cells. In the constant memories, eight comparison tables can then be placed, each with their own expectation field, as the first table e.g. is formed by the first bit of the memory cells, the second table is formed by the second bit of the memory cells, etc. By means of selective addressing and selective reading of the table in question used, it is possible to optionally recognize up to eight different contrast line patterns with one decoder connection, whereby the applicability of the device according to the invention is increased.

A first expectation field is preferably provided for identifying the interval length of a certain contrast line pattern which is scanned successively in the forward direction. A second expectation field is provided for identification of this interval length which is measured successively in the backward direction. The contrast line pattern is scanned in "forward direction" when the contrast lines are read, for example, from left to right. The opposite direction is called the "backward direction", i.e. in the backward direction the contrast lines are scanned opposite to the normal reading direction, i.e. from right to left. If scanning is carried out in the forward direction, and if the addressed lines and columns of the successively measured interval length lie in the first expectation field, an identification signal PIC OUT V is emitted which contains the information that the contrast line pattern has been identified in the forward direction. If, however, the contrast line pattern is identified by scanning in the backward direction, a second identification signal PIC OUT R is emitted which is different from the first identification signal PIC OUT V.. The first method step for finding the contrast line pattern is thus significantly accelerated, i.e. the number of different angle scanning steps is reduced when the contrast line pattern are identified in forward and backward directions. The time for carrying out one reading cycle is significantly reduced.

As different contrast line patterns can be identified with a decoding circuit, it is possible to link a certain information content to the contrast line patterns used. For example, a used contrast line pattern may contain information about the shape of the character pattern in the data tracks. For example, a special contrast line pattern can always be used if the first data track contains only alpha characters,

for example the name of the object to be identified. Alternatively, individual contrast line patterns may be printed in front, behind or below the data tracks to signal that the data tracks have a predetermined, corresponding shape.

If a special contrast line pattern is identified, a corresponding, individual identification signal is emitted, and this identification signal can be used to start corresponding control functions. For. for example, the identification signal can successively activate an alpha character decoder and a numeric character decoder.

When the scanning beam runs over a dark colored area of the data field, the binary video signal has a

first amplitude, Hi, and it has a second amplitude, Lo,

when the scan beam runs over a bright, character-free area of the data field. The assignment of the amplitudes Hi and Lo is arbitrary, and the opposite assignment of the two amplitudes to light and dark areas of the data field is also possible.

The intervals of the video signal whose length is to be determined preferably extend from a rising edge to the next rising edge, as well as overlapping from the falling edge located between these rising edges to the next falling edge of the video signal. The third interval then extends from the second ladder flank to a subsequent ladder flank, etc. In that the intervals extend from ladder flank to ladder flank. respectively. from hem flank to hem flank, it is ensured that blurred edges, of the contrast lines of the contrast line pattern - which usually for all contrast lines run in the same direction and uniformly - essentially do not affect the interval lengths, so that the decoder device also recognizes "contrast line patterns which have been produced under different printing conditions.

To increase the recognition security, contrast line patterns can be used with a signal- or sign-free start or precursor zone of predetermined length, the precursor person being arranged in front of the first contrast line. The measurement of interval length then preferably starts first

when a character-free precursor zone of the predetermined length appears in the video signal.

The measurement of the interval lengths is preferably terminated when an arbitrary interval length exceeds a predetermined maximum value which is equal to the maximum interval length occurring in the contrast line pattern. Likewise, a continuous measurement of the interval lengths is preferably terminated when the ratio between two consecutive, measured interval lengths lies outside the predetermined value range. In both cases, the utilization is interrupted at the earliest possible time and the decoder device is again reset to its active state.

The . continuous measurement of the interval lengths is likewise preferably terminated when a character-free intermediate zone of a predetermined duration corresponding to the maximum scanning distance between the lines of the contrast line pattern is sensed. The decoder device is then reset to the active state and is then ready for execution of a new cycle.

The contrast line pattern further preferably comprises a character-free trailing zone with a length such as e.g. corresponds to the length of the character-free precursor zone. A contrast line pattern recognition signal is then preferably only emitted when the video signal likewise contains a character-free trailing zone of the predetermined length.

In order to increase the redundancy or redundancy of the recognition process and thus reduce the error probability, the same line is preferably scanned n times, and a contrast line pattern recognition signal is only emitted when a contrast line pattern is recognized n times in succession.

The length of the mutually overlapping intervals of the video signal is preferably measured digitally and, for this purpose, counted in a counting circuit. The counter circuit comprises a timing control circuit which produces gate pulses having the same length as the corresponding intervals of the video signal. For measurement of. the length of a respective interval of the video signal, a respective counter is arranged which is fed to a port input with the port pulses for the respective interval. The counting inputs of the counters are connected together with a clock or pulse generator which emits sync or clock pulses to the counters during the occurring gate pulses. The end position of the individual counters then corresponds to the respective time length of the respective gate pulses and thus to the corresponding intervals. Consecutive counted^interval^count values address the corresponding constant memory, after the later interval count value has been counted, whereby the count can persist also for further interval count values.

As already mentioned, each addressable cell or position in the constant or read-only memories (ROM) stores an n-bit data word, n being preferably equal to 8. The expectation field for the r'th contrast line pattern PIC is stored in the m'th bit of the n -bit data word by holding the respective bit "Hi". The outputs from the constant memories are connected to a gate circuit which emits an output signal after the addressing of the constant memories, said output signal specifying which of the possible contrast line patterns PICm has been identified. If, for example, a first count is measured that addresses the line in the first constant memory PRQM 1, and if a second count is then measured that addresses the column in the first constant memory PROM 1, the memory cell defined by the addressed line and column emits a n-bit data word. The same applies to the second and third counts that address a cell in the second constant memory, and the same applies to the third and fourth counts that address a cell in the third constant memory.

The gate control circuit which is connected to the outputs of the PROM memories preferably comprises n parallel AND gates,

n > m > 1; m= 1, 2, 3, .... The number of input terminals to the m'th AND gate is equal to the number of comparison tables or expectation fields, i.e. the number of constant memories. The input terminals of the rn'th AND gate receive the m'th bit

of the n-bit data words that are read from the various constant memories.

The rn'th AND gate emits an output signal LPICm, m = 1, 2, 3 .... when all its input terminals receive the value Hi, i.e. when the m'th bit from the various constant memories signals that the quotient of the successive measured counts that are compared in the respective constant memory lie within the measured expectation field.

The facility is preferably organized so that

an identification of the scanning direction relative to the line sequence of a contrast line pattern PIC is possible. For this purpose, the successively measured interval length of the used contrast line pattern PICm as it is scanned in the forward direction (from left to right), is linked to a first expectation field. The successively measured interval length of the same contrast line pattern PICm as it is scanned in the backward direction (ie from right to left), is linked to a second expectation field - (comparison table) . When the contrast line pattern is identified in the forward direction and

in the backward direction, a respective identification signal PICmOUT V and PICmOUT R is emitted which thus also indicates the scanning direction. The advantage of this embodiment of the invention is that the order or sequence, i.e. 1, 2, 3 of the contrast lines can be identified when this sequence is scanned in the direction 1, 2, 3, and when this sequence is scanned in the opposite direction, i.e. in the sequence 3, 2, 1. Using this feature, a field rotation of a maximum of 180° - at different discrete angles - is sufficient to find any contrast line pattern, so that the search time is reduced to approx. half of the normal value.

The invention will be described in more detail in the following in connection with design examples with reference to the drawings, where fig. 1 shows a first arrangement of a contrast line pattern within a data field with a data track, fig. 2 shows a second arrangement of. a contrast line pattern within a data field, fig. 3 shows a third arrangement of a contrast line pattern within a data field, fig. 4 shows a light-dark distribution for different contrast line patterns perpendicular to the contrast lines, fig. 5 shows a section of the video signal as a function of time, the section corresponding to the contrast line pattern according to fig. 4 (a), fig. 6 shows a block core of the decoder device's counting circuit, fig. 7 shows a block core of the decoder device's comparison table, fig. 8 shows a further embodiment of the decoder device's comparison table, fig. 9 shows a block core of the decoder device's utilization circuit, fig. 10 shows a schematic representation of the structure of the comparison table, fig. 11 shows a pulse plan for the pulses generated in the timing control circuit of the counter circuit, fig. 12 shows a pulse plan for the pulses processed within a selection circuit, and fig. 13 shows a circuit diagram of a further embodiment of the comparison table according to fig. 8, comprising a gate control circuit for selectively reading out different expectation fields from each constant memory.

In fig. 1 - 3 different identifications 50 are produced, e.g. adhesive price labels, such as is attached to a container, a package or one or another item of merchandise and appears in any position and alignment on a picture window. The image window is determined e.g. by means of the optical opening or aperture of a light point scanning device, e.g. a Vidikon tube, with the help of which the picture window is scanned first linearly, and then in a linear scanning grid.

Each of the identifications 50 has a data field which in at least one data track 51 contains contrasting characters 52 for identification of the item that is provided with this identification. The contrasting characters preferably consist of clear characters in one of the known, machine-readable font types, e.g. in OCR-A font or OCR-B font. In a predetermined position and orientation in relation to the data track, a contrast line pattern 54 is arranged - also called position identification code PIC - which comprises several parallel contrast lines with different distances and/or line widths. The contrast line pattern is according to fig. 1 placed in front of the data track, according to fig. 2 under the data track and according to fig. 3 at the end of the data track. The contrast line pattern 54 is perpendicular to the contrast lines designed asymmetrically to characterize the data field with regard to the beginning and end of the data tracks. Those in fig. 1 and 2 shown contrast line patterns, which are hereinafter referred to as PIC patterns, comprise a character-free precursor zone 56 and a character-free successor zone 58.

Although only line patterns with three lines are shown, patterns with more than three lines can also be used. Furthermore, the patterns - unlike the illustrations on fig. 1 - 3 - also be placed in a different position and have a different orientation in relation to the data tracks. It is also possible to provide two or more line patterns on one identification field 50.

As shown in fig. 1, the image window becomes, respectively an image corresponding to the picture window, e.g. on a Vidikon tube impact plate, scanned step by step with a step angle a of at least one scanning line 60. It is essential that the line or PIC pattern before reading the data traces is first recognized in a safe manner and its position and orientation in relation to the scanning beam 60 is determined, as the characters or signals contained in the data track can then be read using a subsequent raster scan in the direction of the data track. Fig. 4 shows the light-dark distribution perpendicular to the line direction for different PIC patterns which contain three lines and which are all asymmetrically structured and can therefore be used in accordance with the invention. Fig. 5 shows a section of the video signal which, when scanning a PIC pattern according to fig. 4(a) is produced as an electrical, binary signal, dark areas of the PIC pattern being assigned to the amplitude Lo. Any light-dark fluctuations within the individual lines and distances of the PIC pattern are eliminated immediately after the scan in the electrical signal. Within the video signal section according to fig..5, a character-free precursor zone is drawn which corresponds to the zone 5 6 in fig. 1, a first interval Tl from the first riser flank to the second riser flank, a second interval T2 from the first riser flank to the second riser flank, a third interval T3 from the second riser flank to the third riser flank and a fourth interval T4 from the second riser flank to the third hem flank, and also a trailing zone corresponding to the zone 58 in fig. 1.

The decoding of the PIC pattern, e.g. according to fig. 5, takes place in accordance with. the delta-distance method, in which successive and mutually overlapping intervals are tested, i.e. Tl, T2 and T2, T3 and T3, T4 are in a predetermined relationship to each other given by the PIC pattern to be decoded. If the value of the quotients between successive, mutually overlapping interval lengths lies within predetermined value ranges, whereby the range widths are given by the print blurs and digitization blurs, the sought-after PIC pattern is very likely present.

Fig. 6 shows the counting circuit which forms the input to the decoder device according to the invention and which counts the interval lengths T1 to T4 and makes these available as binary values for further processing. The counter circuit contains a timing control circuit 2 to which the video signal

VIDEO is supplied and which on a first output emits a first gate signal Tl from a first rise edge to a second rise edge of the video signal, on a second output emits a second gate signal T2 from a hem edge following the first rise edge to a second hem edge of the video signal, on a third output transmits a third gate signal T3 from the second rise edge to the next, third rise edge, and on a fourth output emits a fourth gate signal T4 from the second hem edge to a next, third hem edge, cf. the pulse plan according to fig. 11.

The gate signals Tl to T4 are individually supplied to a gate input Gl - G4 to a respective counter 6, 8, 10, 12. Each counter 6-12 receives at its counter input CT1 - CT4 clock pulses from a clock or pulse generator 14, which . pulses are counted by the counters 6-12 as long as the gate signal Tl - T5 is supplied. The counter position that can be output on the counters' 6-12 outputs TC1 - TC4 then constitutes a measure for the length of the gate signals Tl - T4.

In the timing control circuit 2, a trigger signal E2 is generated from the falling edge of the gate signal T2, a trigger signal E3 is generated from the falling edge of the gate signal T3 and a triggering signal E4 is generated from the falling edge of the gate signal T4 and emitted on separate outputs. From the trigger signal E4, a signal PVENT is further produced whose length corresponds to the character-free trailing zone after the end of the gate signal T4, and which is emitted at a separate output.

The counter circuit further contains a reset circuit 4 which receives the video signal VIDEO and on a separate input receives an external reset signal RESET IN at the beginning of each scanning line. The reset circuit 4 emits a reset signal RESET to the counters 6 - 12 reset inputs RS1 - RS4 and to the timing control circuit 2 and resets both the counters 6 - 12 and the timing control circuit 2 to the active output state when the video signal contains a character-free section - amplitude Lo - which is>. greater than the maximum scanning distance between the lines of the PIC pattern, this maximum scanning distance being given by the maximum distance within the PIC pattern multiplied by the

maximum permissible scanning angle.

The counter circuit further contains an overflow detection circuit 40 which receives an overflow signal from the counters 6-12 overflow outputs OV1 - OV4 and then emits a reset signal OV RESET and then resets the decoder device to a new one. ready state.

Fig. 7 shows an embodiment of the decoder device's comparison table. The comparison table contains a constant memory (PROM) 28. The PROM unit 28 is organized so that the counter's 6 counter position TC1 addresses separate lines in a first memory matrix, and : . the counter's 8 counter position TC2 addresses separate slots in the first memory matrix. The counter positions TC1 and TC2 are supplied to the PROM unit 28 via a gate coupling 16, 18 by means of the trigger signal E2 after counting the counter position TC2. Within the first memory matrix, an expectation field is preset which comprises the memory cells in which the quotient between line address and column address lies within a predetermined value range. This value range corresponds to the quotient between the first interval length and the second interval length of the PIC pattern used. If a memory cell or memory position within the expectation field is addressed using the counter positions TC1 and TC2, a comparison signal with a first amplitude is emitted, e.g. Hi, which signals that information corresponding to part of the PIC pattern used was searched for.

To similarly perform a comparison between the gate pulses T2 and T3, the counter position TC2 for the second counter 8 addresses the lines in a second memory matrix, and the counter position TC3 for the third counter 10 addresses the columns in the second memory matrix. The addressing takes place after the counter position TC3 has been counted and the counter positions TC2 and TC3 with the help of the trigger signal E3 have been transmitted via the gate connection 20, 22 to the PROM unit 28. Within the second memory matrix, an expectation field is also preset which includes the memory cells whose quotient between line address and slot address lies within a predetermined range which is equal to the value range for the quotient between the intervals of the PIC pattern corresponding to the gate signals T2 and T3. If a memory cell in the expectation field is addressed, emitted

one

a comparison signal with a first amplitude, e.g. the amplitude Hi. When addressing a memory cell outside the expected field, a comparison signal with another ampli-tu de Lo is emitted.

The comparison between the third counter's 10 counter position TC3 and the fourth counter's. 12 counter setting TC4 occurs in the same way when addressing lines and columns in a third memory matrix which likewise contains an expectation field, as when addressing a memory cell within the expectation field, a comparison signal with the first amplitude Hi is emitted. The addressing of the third memory matrix takes place after the fourth counter's 12 counter position TC4 has been counted and the trigger signal E4 transmits the counter settings TC3 and TC4 via the gate connection 24, 26 to the PROM unit 28. The gate connections 16 - 26 consist of AND gates.

The readout of the comparison signals LPIC1, LPIC2 and LPIC3 takes place by means of the readout signals EL2, EL3 and EL4 which are provided by delaying the trigger signals E2, E3 and E4 in a delay coupling 30, cf. also the pulse plan according to fig. 12. The reading may only take place after the addressing of the first, second and third memory matrix has been carried out.

Alternatively to the embodiment according to fig. 7, the first memory matrix can be realized using a first constant memory, PROM1, the second memory matrix can be realized using a second constant memory, PROM2, and the third memory matrix can be realized using a third constant memory, PROM3, as it then advantageously three separate constant memories with significantly less memory capacity can be used.

The constant or read memory 28 comprises memory cells or storage locations, each of which has a capacity of n bits, ie. n = 8. Since only one bit of each memory cell that defines the expectation field is used to provide an expectation field, up to n different expectation fields for n different PIC patterns can be accommodated at the same time, preferably n = 8, and the first for expectation fields in PROM1 is placed in the respective first bit of the memory cells, the second expectation field is placed in the second bit of the memory cells, etc. The same applies to PR0M2 and PR0M3. The line and column addressing for a particular PIC pattern must then take place selectively to the corresponding bits of the memory cells. Furthermore, a switch 32 is connected after the PROM unit 28 which selectively reads out the comparison signals LPIC 1, LPIC 2 and LPIC 3 from the respective bits of the memory cells, and emits the sequential comparison signals at its output. LPIC 1, LPIC 2 and LPIC 3 formed exploit signal MUX PIC to the evaluation or exploit coupling.

The decoder device's utilization circuit is shown in fig. 9. A buffer 34 receives the utilization signal MUX PIC and stores the comparison signal LPIC 1 - which indicates that the value TC1/TC2 lies within a predetermined range - as well as the comparison signals LPIC 2 and LPIC 3. The reading takes place by means of an input of the readout signals EL2, EL3 and EL4 which essentially occurs simultaneously with the comparison signals LPIC 1, LPIC 2 and LPIC 3, cf. the pulse plan according to fig. 12. After all comparison signals are stored in the buffer as storage signals LPIC 1<1>, LPIC 2', LPIC 3', these storage signals are output to an AND gate 36 which outputs an output signal LPIC when all storage signals LPIC 1' etc. have a first amplitude which corresponds to the first amplitude of the comparison signals LPIC 1 etc, cf. fig. 12. The output signal LPIC is supplied to an output circuit 38 which receives the video signal VIDEO and from the control circuit 2 receives the waiting signal PVENT. The output circuit 38 emits a recognition signal PIC OUT when the video signal VIDEO during the presence of the waiting signal PVENT remains at the amplitude Lo which characterizes a character-free substrate. In this way, it is ensured that the decoded line pattern is followed by a character-free trailing zone that is equal to the PIC pattern trailing person 58.

The output circuit 38 is reset by means of the external reset signal RESET IN and then emits a reset signal RESET A to the intermediate memory 34 and likewise resets this for a renewed operating operation. The buffer is further reset to zero using the overflow zero position signal OV RESET when a counter 6-12 signals an overflow (flood). Fig. 10 shows in schematic representation how the comparison table, e.g. The PROM unit's 1 partial comparison table for comparison of the quotient TC1/TC2 is organized. The table is presented as a storage matrix and the table's lines or rows are consecutively denoted by binary addresses. The columns or columns of the table are likewise consecutively designated with binary addresses, as according to a preferred embodiment of the decoder device according to the invention a 5-bit representation has been chosen. All memory cells or memory positions with a specific value of the quotient between line address and column address lie on one line, the so-called expectation line, around which lies the expectation field within which all storage cells are located whose address quotients lie within a predetermined range of values. The counts TC1 - TC4 are likewise output as 5-bit words. The count TC1 addresses the table's lines and the count TC2 addresses the table's columns.

Fig. 13 shows a circuit diagram of a further embodiment of the comparison table according to fig. 8. The comparison table contains a first constant or read memory PROM1 whose lines are addressed by the measured first count TCl and whose columns or columns are addressed by the second measured count TC2. A second non-volatile memory PROM2 is also included. whose lines are addressed by the measured second count TC2 and whose slots are addressed by the measured third count TC3, and further a third constant memory PROM3 whose lines are addressed by the measured third count TC3 and whose slots are addressed by the measured fourth count TC4 . Each constant memory PROM1, PROM2, ... thus provides a two-dimensional line and column-organized storage matrix to realize comparison tables for respective, two successively measured counts of the contrast line pattern.

Each cell or position in the memories PROM1,

PR0M2, .... which is addressed by its line and column, stores an n-bit data word where n > 1, dg preferably n = 8. For identification of m different contrast line patterns, with

n£m > 1, the expectation field for an n'th contrast line pattern PICm/m = 1, 2, 3, ...., is stored in the m'th bit of the n-bit data word by writing the value Hi into the respective bit, while the nth bits of memory cells outside the expectation field have the amplitude Lo.

The output signal from the constant memories PR0M1, PRQM2,,.... is fed to the gate control circuit 28 which has n parallel AND ports 29. The AND gates 2 9 have as many input terminals as there are constant memories PROM1, PR0M2> ...

Interconnectors 2 9a are arranged between the output of each constant memory PR0M1, PR0M2>, ... and the gate control circuit, which emits the data word received from the constant memory in bit-parallel form at its output. The interconnects 29a have n outputs of which the first output transmits the first bit, the second output transmits the second bit, the third output transmits the third bit etc. of the received n-bit data word. The n AND gates 29 are connected to the intermediate connections 29a so that the m'th AND gate on its various input terminals receives as an input signal the rnth bit of the data words received from the various constant memories. Each AND gate emits an output signal LPICm when all its input signals have the amplitude Hi, i.e. when the rnth bit in all constant memories has the amplitude Hi, and thus marks that the actually read contrast line pattern PICm lies in the measured expectation field and is identified without failure.

Thus, when the first AND gate 29 emits a pulse, the first contrast line pattern PIC is read and identified. If the measured AND gate 29 emits a pulse, the measured contrast line pattern PICm is read and identified. The output from the n AND gates is connected to a selection circuit (not shown). This selection circuit is adjustable to provide an output signal only when the selected contrast line pattern is reached. The PIC is read and identified.

Claims (6)

1. Procedure for the identification of objects that appear in arbitrary positions with arbitrary orientation and at arbitrary times on a picture window, and on a surface facing the picture window has an identification in the form of a data field, which in at least one data track includes contrasting characters with at least a contrasting line pattern that characterizes the data track's position and orientation, and has a number of parallel lines with different spacing and/or line width. in that the image window is scanned linearly and a binary video signal corresponding to the scanned contrast sequence is produced, in that the image window in a first step is scanned at varying angles until a contrast line pattern is detected, in that the position and alignment of the data field in relation to the image window is determined in a second step , and a raster scan is performed in the direction of the data track and the characters contained in the data track are read and. is decoded in a third step, the contrast line pattern being identified by measuring the length of the overlapping light and dark intervals of the video signal, the successively measured interval lengths are compared and a comparison signal is produced which has a first amplitude when two interval lengths being compared have a predetermined mutual relationship which corresponds to the corresponding distance in the contrast line pattern, and an identification signal is emitted when the comparison signal is received during each of a number of successive comparison steps determined by the contrast line pattern, the comparison of successively measured interval lengths being performed in pairs in a two-dimensional comparison table within which the possible, discrete values of a interval length., is assigned to table lines, and the possible, discrete values of the next closest measured interval length are assigned to table columns, and an expectation field is determined that includes the table places in which the quotients between the two successively measured, compared interval lengths lie within a given value range, and a reference signal is produced which has a first amplitude when the compared interval lengths correspond to a table place within the expectation field, with different expectation fields for corresponding, successively measured interval lengths for different contrast line patterns being included in the individual comparison table, characterized by an individual identification signal for each contrast line pattern is emitted when the corresponding contrast line pattern is identified. 2. Method according to claim 1, characterized in that each individual comparison table comprises a first expectation field that corresponds to successively measured interval lengths of a contrast line pattern in the forward direction, and a second expectation field that corresponds to successively measured interval lengths of this contrast line pattern in the backward direction, and that a first identification signal (PIC OUT V) is emitted when the contrast line pattern is identified by scanning in the forward direction, and a second identification signal (PIC OUT R) is emitted when the contrast line pattern is identified by scanning in the backward direction. 3. Method according to claim 1, characterized in that each identification that appears on the image window comprises a number of different contrast line patterns (PICm , m = 1, 2, 3, ...), and that an individual identification signal (PICm OUT, m = 1, 2, 3, ...) is emitted when the corresponding contrast line pattern is identified. 4. Device for the identification of objects that appear in arbitrary positions with arbitrary orientation and at arbitrary times on a picture window, and on a surface facing the picture window displays an identification in the form of a data field which in at least one data track includes contrasting characters with at least one contrasting line pattern which characterizes the position and orientation of the data track, and has a number of parallel lines with variable distance and/or line width, as the image window is scanned linearly and a binary video signal corresponding to the scanned contrast sequence is produced, as the image window is scanned in a first step at varying angles until a contrast line pattern is detected, the position and alignment of the data field in relation to the image window is determined in a second step and a raster scan in the direction of the data track is performed and the characters contained in the data track are read and decoded in a third step, which device comprises an optoelectronic scan device p om contains a rotatable scanning grid which at the output emits a binary video signal corresponding to the linearly scanned image field and in binary form contains the scanned line's contrast pattern, a decoder for identifying the contrast line pattern which identifies the position and orientation of at least one data track of a data field , and a device for aligning the scanning ras- . tered in parallel with the data track and for reading the searched characters of the data track, the decoder comprising a counting circuit which receives the video signal and counts the length of successive, overlapping intervals of the video signal, at least one reference table for pairwise reception of successively counted interval lengths via a gate circuit and for outputting a comparison signal having a first amplitude when the comparison interval lengths have a given ratio corresponding to the ratio between the corresponding intervals of the contrast line pattern, °9 an utilization or evaluation circuit for producing an identification signal during a sequence of a given number of comparison signals each having the first amplitude, as the reference table comprises two-dimensional constant memories (ROM), as possible, discrete count values for two arbitrary, later interval lengths are addressed to the associated lines or slots in the constant memories, and selectively addressable, different expectation fields in the memories include those memory cells where the quotient between line address and column address lies within a given value range, so that an associated, individual comparison signal is emitted from each constant memory when a memory cell in one of the expectation fields is addressed, characterized in that each addressable memory cell in the constant memories (PROM 1, PROM 2, PROM 3) stores an n-bit data word (n £ 1), that the expectation field for a measured contrast line pattern (PICm , n m > 1) is stored in the rnth bit of the n-bit data word by making this bit "Hi", and that the constant memories are connected to a gate control circuit .(28) which emits an output signal LPIC m (m = 1, 2 , 3, ••?), when the constant memories are addressed and3 the searched contrast line pattern... (PICm) is identified. 5. Device according to claim 4, characterized in that the gate control circuit comprises n parallel AND gates, n > m > 1; m = 1, 2, 3, as the m'th bit of the data words stored in the constant memories is connected to! the inputs of the m'th AND gate which emits the output signal LPICm when all inputs are "Hi". 6. Device according to claims 4 and 5, characterized in that a first expectation field is provided for the successively measured interval counts of a contrast line pattern that is scanned in the forward direction (for example scanned from left to right), that a second expectation field is provided for the successively measured -interval counts of said contrast line pattern which is scanned in a backward direction (scanned from right to left), and that an identification signal is emitted when the contrast line pattern is identified, the identification signal containing information characterizing the relative scanning direction.