JPS63211478A - Optical character reader - Google Patents

Abstract

PURPOSE:To obtain only a result of recognition when a moving speed is constant, by measuring an elapsed time after a line has gone into a visual field, and outputting a result of recognition when said time exceeds a prescribed time while the line is in the visual field. CONSTITUTION:An image of a form 3 is inputted by an image sensor 6, and a binary signal is obtained by a control/binarization circuit 7. Based on the binarization signal, a recognition processing is executed by each character recognizing means 51, a result of recognition of a line in a visual field is obtained, the result of recognition is saved in a recognition result storage means 52, and its contents R1 and a result of recognition R2 obtained newly are compared by a recognition result comparing means 53. A time counting means 54 counts the time when the same recognition result is obtained, and when the time exceeds a prescribed time, a signal of OK is sent to a conformity recognition result output means 56. When the signal of OK comes, the conformity recognition result output means 56 outputs a result of recognition to its signal. As for one line which goes into the visual field, the result of recognition is outputted 58 only once by a one line output means 57.

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention is capable of scanning characters, symbols, etc. (hereinafter, only characters will be described as a representative, but the same applies to symbols as well) by scanning paper with a hand-held scanner. The present invention relates to an optical character reading device that reads ``characters''.

[Conventional technology]

At supermarkets, department stores, etc., PO3 (Point of
5ales) system is widespread. Hand-held optical character readers are often used in this POS system.

As such a device, the present applicant has filed a patent application for the device shown in FIG. 2 as Japanese Patent Application Laid-open No. 41474/1983.

In FIG. 2, reference numeral 1 is a scanner, and with hand 2, paper 3 is
It reads the characters stored on the paper 3 just by placing it on the paper. The paper 3 is, for example, a price tag on which information for a POS system is stored. 4 is a light source, 5 is a lens system, and 6 is an image sensor, which requires a field of view for at least one line of characters stored on paper 3, and in the illustrated example, one line horizontally and one vertically. The field of view is about three times that of text. Reference numeral 7 denotes a control/binarization circuit that converts the analog signal that is the output signal of the image sensor 6 into binary signals corresponding to the character area and the background area, respectively, and sends the signals to the memory 8. The memory 8 stores the binarized signal of the entire field of view of the image sensor 6.

Confirmation processing is performed from the binarized signals stored in the memory 8, but in the illustrated example, only the binarized signals in which the characters are correctly within the visual field are subjected to the confirmation processing. That is, the third
The figure shows an example of scanning with the image sensor 6 and converting it into binarization, and this figure shows a case where characters are correctly included in the field of view 17. FIG. 4 shows a case where scanning is performed with characters protruding from the visual field 17. In this case, the recognition process described in detail later is not performed, and as the scanner 1 moves, the image sensor 6 scans again, binarizes it, and stores the binarized signal in the memory 8 to improve the reading speed. ing. Note that in FIGS. 3 and 4, some vertical lines indicating pixel divisions are omitted.

The visual field position detection circuit 9 determines whether or not to perform recognition processing. That is, the visual field position detection circuit 9 operates in parallel with storing the binary signal into the memory 8, and it is determined whether or not recognition processing is performed almost at the same time as the storage of the binary signal into the memory 8 is completed. is making a judgment. Furthermore, when performing recognition processing, the visual field position detection circuit 9 does not process the entire visual field;
The recognition area is determined so that only the area where characters exist is recognized.

An example of a specific configuration of the visual field position detection circuit 9 is shown in FIG.
The determination circuit 22 includes a ROM 25 and a register 26.

Each row OR circuit 21 performs an OR operation for each row between the second-order pixels (19 in FIG. 3, from the Bt column to the Bu column) approximately in the horizontal center of the field of view of the image sensor 6, i.e. , when the control/binarization circuit 7 sends a binary signal as shown in FIG. The signals are stored in the memory 8, and at the same time, the logical sum of the binarized signals in each row from the Bt column to the Bu column is determined by the respective row logical sum circuit 21.

To further comment, the register 24 is first cleared, and when the signal in the B column of the L1 row is input to the OR gate 23, it is ORed with the output signal of the register 24, and the result is stored in the register 24. If the process is repeated up to the Bu column, the result of the logical sum operation of the LL rows will be stored in the register 24. When the logical sum of up to 3u columns is determined, the contents of the register 24 are sent to the determination circuit 22, and at the same time, the register 24 is cleared. This results in L
The result of the logical sum from row 1 to rows t and p is sent to the determination circuit 22. The result of the logical sum for each row shows whether or not there is a binarized "black" in each row.

The configuration and operation of the determination circuit 22 to which a signal representing the result of the logical sum is supplied are as follows.

That is, the signal representing the result of the logical sum of each row in the determination circuit 22 is input to the ROM 25, and the contents read from the ROM 25 are stored in the register 26. Since the state transition indicating a change in the black and white state is coded and written in the ROM 25, the register 26 stores the black and white state up to that row. The signals from the register 24 indicating the result of the logical sum of the next row and the register 26 indicating the previous state are input to the ROM 25, and the new state signal is stored in the register 26. SET, EN in ROM25
A code corresponding to the DSCLEAR signal is written. When the signal SET changes from white to black, EN
D is output when black changes to white by more than a certain number (corresponding to the height of the character) Vt, and the character position is from the SET line to the END line. On the other hand, when the signal SET and the signal END are not output up to the bottom rows t and p, the CLEAR signal is output to the control circuit 16, and the control circuit 16 clears the memory 8. In this way, subsequent recognition processing is not performed, and a new binarized signal is acquired by scanning again.

In the above example, when calculating the logical sum for each row,
The reason why only the center part is used instead of the symbol for the entire horizontal row of the field of view is to avoid the influence of noise, such as when the paper 3 is smaller than the field of view or when black marks are printed on the paper 3 outside the characters to be read. This is to minimize it as much as possible. When the white of the paper 3 is sufficiently larger than the field of view, the logical sum of all horizontal columns may be taken.

The new line detection circuit 10 determines the character position from the signals SET and END sent from the visual field position detection circuit 9, and compares it with the character position determined by the SET and END signals sent in the next scan, and determines the amount of change in the character position (in pixels). number) or signal CLEA
A judgment is made based on whether the character string has protruded once from the field of view 17 using R, and if it is determined that a new character string exists, a signal NEW is sent to the recognition processing circuit 15. After the signal NEW is sent, the next recognition process is performed.

With this new line detection circuit 10, when reading a character string with more lines than the field of view, it is only necessary to move the scanner 1 from top to bottom or from bottom to top.

That is, since the scanner 1 is placed on the paper 3 and the characters are read,
If left for a long time, the same part will be scanned and read over and over again, but the new line detection circuit 10 prevents this.

Next, cutting out one character will be explained.

Since the binarized signal stored in the memory 8 is the entire field of view of the image sensor 6, the amount of data has a larger number of pixels both vertically and horizontally than the number of pixels that can be processed by the recognition processing circuit 15 that recognizes and processes one character. In other words, it is necessary to convert each character into a binary signal.

The number of pixels that can be processed by the recognition processing circuit 15 will be explained assuming that the number of pixels that can be processed by the recognition processing circuit 15 is n horizontally and m vertically (column direction) in FIG.

First, a binarized signal of p in the horizontal direction and pixel number P of the entire field of view 17 in the vertical direction is transferred to the single-digit bath sofa 1). That is, the binarized signals from the B1 column to the Bn column in the L1 to Lp rows are stored in the one-digit buffer 1.
).

The vertical position detection circuit 12 reduces the binary value of the single-digit bass sofa 1). The vertical position detection circuit 12 determines the vertical position of the character, and the recognition processing circuit 15 determines the number m of vertical pixels (determined so that the character position is located approximately at the center). As a result, the vertical direction m is determined from the binary signal stored in the -digit buffer 1) and transferred to the single character buffer 14 as one character data mxn. In the example shown in FIG. 3, the binary signal contained in the area from Ll to Lm rows and Bl to Bn columns is treated as one character data in the one character buffer 14.
Transfer to.

The lateral position detection circuit 13 determines the lateral position of the binarized signal to be stored when the next binarized signal is stored in the minus digit bath sofa 1). At this time, the characters may be moved horizontally one pixel at a time and stored in the negative digit buffer 1), but the same character may be stored many times, which requires a long processing time. Therefore, the horizontal position detection circuit 13 detects the number of pixels up to the next character, shifts it horizontally by this number of pixels, and
, the binarized signals of the pixels of vertical p are stored in the one-digit bath sofa 1).

The contents of the single-digit bassophon are, for example, shown in Figure 6 (al, (b)
It changes like l, tel.

The recognition processing circuit 15 recognizes a single character from the binary data stored in the single character basso 14, and may be a known one such as that described in Japanese Patent Publication No. 59-6418. extracts feature data of character constituent parts for each row or column in the vertical and horizontal character areas, and outputs character recognition results corresponding to the type and appearance order of this feature data.

[Problem that the invention seeks to solve]

In conventional optical character reading devices, when reading a character string with more lines than the field of view, it is sufficient to simply move the scanner 1 from top to bottom or from bottom to top. However, if the moving speed at that time is too fast, the character image captured by the image sensor 6 may be blurred, and correct recognition results may not be obtained. Inability to obtain a correct recognition result is due to unreadable results (generally called rejects) and misreadings (generally called rejects).
misrecognition). If the paper cannot be read, the recognition processing circuit 15 will output a code indicating the beam jet, and the operator will be able to see it immediately and just apply the scanner to the paper again.If the paper is misrecognized, for example, a code will be displayed on the paper. 7”
is written, the recognition processing circuit 15 outputs a code indicating "1", so the operator can read the recognition processing circuit 15.
It is not possible to tell that it is a misrecognition just by observing the recognition results from In order to check whether erroneous recognition has occurred, the operator must compare the output result of the recognition processing circuit with the paper.

This places a heavy burden on the operator.

The present invention was devised in view of this point, and its purpose is to obtain recognition results only when the moving speed is constant.

[Means for solving the problem] First, the method used in the present invention is adopted as a method for obtaining recognition results only when the moving speed of the scanner is below a certain value.
This will be explained using figures. Figure 7 is for simplicity of explanation.
The field of view is one line in size and the scanner is moved from top to bottom.

fa) indicates the size of the visual field, and H is the height of the visual field. fbl indicates the size of the character, and the height of the character is h. If you move the scanner from top to bottom at a speed of V, a line enters the field of view (fc) and fdl until it exits.
The time T is T-(H-h)/V (1). When the scanner is moved at the upper limit of the moving speed V,,X, the time from when a line enters the field of view until it exits T s 1) 1
becomes T□, -(H-h)/:V,□ ...(2) Now, if T>Tl141) holds, then C1) (2
From 1, v〈■1. II...(31 holds true. Also, when the scanner is tilted and the character line is hit, the actual height h" of the character line becomes h">h,
(Fig. 7 (el), when the scanner is moved at a speed of V, a line enters the field of view (fl) until it exits (a is the time T°
is T"-(H-h")/V...(4),
Now, if T'> T, 5, holds, then (21(41
Therefore, V < V, , (H-h')/(H-h) < Va
ax --- (51 holds true. From the above considerations, calculate the elapsed time t after the line enters the field of view, and if th<lin is exceeded while the line is in the field of view, then the speed of the scanner is It was found that ① is less than v,,7.The present invention measures this elapsed time t, and t is T a
This realizes a processing device that outputs a recognition result when i.sub.in is exceeded.

Next, the present invention will be explained in detail.

FIG. 1 shows the configuration of the present invention.

Reference numeral 1 denotes a scanner, which is placed on paper 3 with hand 2.

4 is a light source, 5 is a lens system, 6 is an image sensor, and 7 is a control/binarization circuit; The analog signal, which is the output signal of the photoelectric conversion element in the image sensor, is converted into a binary signal corresponding to the character area and the background tilt. 51 is a means for recognizing characters included in the field of view from the binary signal. 52 is means for storing recognition results.

53 is means for comparing the recognition result R1 stored in the recognition result 52 with the recognition result R2 obtained from each new character recognition means. When R1 and R2 do not match, the timer 54 is initialized. Further, the time measuring means is configured to measure the elapsed time t.
is measured, and when t exceeds Tm1n and a timer up signal is output, if R1 and R2 match, an OK signal is transmitted to the conformance recognition result output means 56 via step 55. This is a means for extracting only the corresponding recognition result when an OK signal is output to the conformity recognition result output means 56. - The line output means 57 is a means for finally outputting the recognition result 58 only once for the line within the field of view. - In addition to the method of repeatedly outputting 58 for one row from the recognition result obtained from the matching recognition result outputting means 56, the row outputting means 57 can output matching recognition results as described in the embodiment. There is also a method in which once a recognition result is obtained from the means 56, the recognition processing of each character recognition means 51 for that line is stopped.

[Effect]

An image of the paper 3 is captured by the image sensor 6,
A binary signal is obtained by the control/binarization circuit 7. Recognition processing is performed by each character LUm means 51 based on the binary signal, and the recognition result of the line in the visual field is changed.

The image sensor repeatedly captures images, and recognition processing is performed repeatedly. The recognition result is stored in the recognition result storage means 5.
2, and the content R1 is compared with the newly obtained recognition result R2 by the recognition result comparing means 53. The timing means 54 measures the time during which the same recognition result is obtained,
When R1 and R2 match and the time exceeds a certain time, an OK signal is sent to the matching recognition result output means 56. Compatibility recognition result output means 56 when an OK signal is received
outputs the recognition result for that signal. - The line output means 57 outputs the recognition result 58 only once for one line that has entered the field of view.

〔Example〕

FIG. 8 shows an embodiment of the present invention using a microprocessor, and FIG. 9 shows a flowchart of processing contents in the microprocessor.

In FIG. 8, the parts numbered 1 to 16 have the same functions and configurations as the parts with the same numbers in the description of the prior art in FIG. 2. However, 8.1). 12.13.14.15.
16 is an embodiment of each character recognition means shown in FIG. 151, and unlike the case of the prior art shown in FIG. Performs processing to recognize each character.

The microprocessor 81 operates as shown in FIG.
It has 56 functions. The program of the microprocessor 81 is stored in the ROM 82, and the recognition result storage means 5
The recognition results for realizing 2 are stored using the RAM 83. The timer 84 is an example of the timer 54, and the microprocessor 81 can read out the elapsed time since initialization from the microprocessor 81.

Gate 86 and SR flip-flop 85 are an embodiment of the single row output means of FIG. 157.

When a row of maple leaves enters the field of view, processing proceeds as follows. First, the visual field position detection circuit 9 detects that the row is within the visual field. If the row is a new row that has come into view, the new row detection circuit 10 sends a NEW signal to the set input of the SR flip block 85. S
When the R flip-flop receives the NEW signal, it outputs the NEW' signal. The period when the NEW' signal is output is 1)
Each character in the field of view is repeatedly recognized by the means for recognizing each of the 16 characters. The NEW' signal is transmitted from the microprocessor 81 to the F!
This continues until the recognition complete signal is transmitted to the reset input of the SR flip-flop.

The content to be processed by the microprocessor 81 is
This will be explained using figures. First, ■ initialize the recognition result storage area. Next, (2) importing the recognition results. (2) is a process of importing the recognition results again if the recognition results include unrecognized (reject) characters. This is a process for not outputting unrecognizable characters (87 in FIG. 8), and if unrecognizable characters can be output, the process ``■'' is unnecessary. ■
This is a process of comparing the recognition result imported in ■ with the contents of the recognition result storage area, and if they match, proceed to ■.

If you perform the process of ■ after performing the process of ■, nothing is written in the recognition result storage area, so be sure to proceed to ■. This is a writing process.

(2) is a process for initializing the timer 84. After the processing in (2), the process proceeds to (2) and the recognition result is taken in again.If the process proceeds from (2) to (2), it is determined in (2) whether the timer value exceeds TIain in equation (2). If Tll1n is exceeded, at ■, the recognition process completion is sent to the gate 86, and each character recognition process from 1) to 16 is stopped, and at ■, the recognition result taken in at ■ is output (87 in Figure 1). , the recognition result storage area is located on the RAM 83, and is an embodiment of the recognition result storage means 52, as can be seen from the above description. !
! ! The identification result comparison means 53 is implemented in the process (2).

For example, Ttai n is as follows.

Now, assume that the height H of the visual field in FIG. 7 fal is 10fi, and the height h of the characters is 3fi. If the upper limit speed Vain at which the scanner can be moved is 2 Om/sec, the time Tll1n that must be measured by the counter circuit is:
Tai+t” (H-h)/Vain” 0.35
sec Next, an embodiment in which the time is measured by counting the number of recognition results obtained by the microprocessor 81 instead of measuring the time by the timer 84 is shown in Fig. 1O, and the processing in that case is shown in Fig. 1). Explain using.

In the flowchart of FIG. 1), the process marked with ■ in FIG. 10 is omitted in order to output unrecognizable characters as well. In addition, a counting variable I is used in place of the timer 8 in FIG. 8, and by counting the number of times the recognition result has been taken in in (■), the time since I is initialized in (■) is measured. ■ is the count variable ■ value is the number of counts N
This is a process of determining whether the elapsed time since initialization of ■ in ■ exceeds Twin by determining whether it exceeds Twin. The count number N is determined as follows.

Now, the time required to obtain one recognition result is Tc.
cycle, the time it takes to obtain a recognition result once is Tcycle X I. In order for this to exceed pmin, Tcycle ．．
If it is 35 seconds, it will be N -5.

As a special example, if Tcycle > Twin, it becomes N-0. In this case, if the same recognition result is obtained continuously, the scanner's moving speed V is less than V win.
Although the part that measures time is not explicitly shown in Figure 12, it is possible to detect that the same recognition result is obtained repeatedly by using ■, and at the same time, this recognition result continues for T+win or more. is detected. In other words, ■ indicates the recognition result comparison means 53 and the time measurement means 5.
It also serves as 4.

Finally, an explanation is added regarding the recognition result comparing means 53 (■ in FIG. 9, ■ in FIG. 10, ■ in FIG. 1). Generally, in order to determine whether lines that have been repeatedly recognized are the same, it is necessary to compare all characters in the lines. However, in the OCR price tag for PoS according to the JIS standard as illustrated in FIG.
P.V.

The characters X and ¥) are different from each other. Therefore, when the object to be read by the optical character reading device is such a price tag, the recognition result comparing means can determine that the repeatedly recognized lines are the same simply by checking the identity of the function characters in the lines.

〔Effect of the invention〕

According to the present invention, it is possible to realize an optical character reading device that does not output recognition results even if the moving speed of the scanner is too fast. This eliminates the need to check the recognition results when the movement speed is too fast, and allows the scanner to read characters with peace of mind while moving, significantly reducing the operational burden.

[Brief explanation of the drawing]

FIG. 1 shows a configuration example of an optical character reading device according to the present invention, FIG. 2 shows a configuration example of an optical character reading device according to the prior art, and FIG.
4 and 4 are explanatory diagrams conceptually showing the field of view of the image sensor, FIG. 5 is a block diagram showing the specific configuration of the field of view position detection circuit, and FIGS. 6 (a) to (C). is an explanatory diagram conceptually showing the contents of one character memory, Figures (31 to (6) in Figure 7 are diagrams showing the relationship between visual field height H and character height h, Figure 8, and Figure 1O) Examples of the present invention, FIG. 9, and FIG.
), Figure 12 is a flowchart of the microprocessor processing, and Figures (a) to (C) of Figure 13 are the OC for PO8.
This is an example of an R price tag. 1...Scanner, 2...Hand, 3...
...Paper, 4...Light source, 5...Lens system, 6...Image sensor, 7...
Control/binarization circuit, 8...Memory, 9...
...Visual field position detection circuit, 10...New line detection circuit, 1)...-digit buffer, 12...Vertical position detection circuit, 13... Lateral position detection circuit, 14
...Character buffer, 15...Recognition processing circuit, 16...Control circuit, 17...
Field of view, 18...-Number of horizontal pixels in character recognition processing, 19...Pixels that are ORed in the visual field position detection circuit, 21...Normal sum circuit for each row, 22...
・Judgment circuit, 23...OR gate, 24...
...Register, 25...ROM, 26...
...Register, 51...Each character recognition means,
52... Recognition result storage means, 53...
Recognition result comparison means, 54... Timing means, 55.
... Gate, 56 ... Compatibility recognition result output means, 57 ... - line output means, 58.87 ...
...Final recognition result output, 81...Microprocessor, 82...ROM, 83...
...RAM, 84...Timer, 85...
...SR flip-flop, 86...gate. Patent applicant: Sumitomo Electric Industries, Ltd. Agent: Kama 1) Text 2 1) Recognition results obtained in Figure 12 R2■ Figure 13 (a) (C) Procedural amendment (voluntary) 1988 5 2) Name of the invention Optical character reader 3, person making the amendment Relationship with the case Patent applicant address 5-15 Kitahama, Higashi-ku, Osaka Name (r,
Ne+ (213) Tahide Denki Kogyo Co., Ltd. 4, Agent + ILR 1-18 Nihonbashi, Minami-ku, Osaka 542
Contents of amendment No. 12 (+) "Accurate processing" in the fourth and second lines from the bottom of No. 3 of the specification will be corrected to "recognition processing." (2) On page 13, line 8, “Next,” (Figure 7 (e)
), J becomes r (Figure 7 (C)), and corrects it to J. (3) In the 13th true, line 10, “(darkness is)” is changed to “((to)
I will correct it to '. (4) The 8th line from the bottom of page 14 will be corrected to "in recognition result 52" and "in recognition result storage means 52." (5) In the first line of page 15, "56 should be corrected to OKJ." (6) In the ninth line from the bottom of page 16, "Examples are shown."
I will show an example and amend it to ``. (7) In the 9th line from the bottom of No. 17, ``mijino in r'' will be corrected to ``in character.'' (8) Correct "Go forward..." at the beginning of the 4th line from the bottom of No. 18 to "Go forward, if you do not -t, proceed to ■." (9) Correct "first time" in line 3 from the bottom of No. 18 to "first time". 0[D Insert "Tonaru," between the second line from the bottom of No. 19 and the last line. 00 "Character" in the 3rd and 7th lines of page 22 will be corrected to "code".

Claims (2)

[Claims] (1) In a hand-held optical character reading device that optically reads characters, symbols, etc., an image sensor is formed by arranging photoelectric conversion elements in a planar manner and can capture multiple characters, symbols, etc. for one line within its field of view. a binarization circuit that converts the analog signal output from the photoelectric conversion element into a binarized signal corresponding to a character area and a background area; a means for recognizing each character in the visual field; and a recognition result. recognition result storage means for temporarily storing the recognition result storage means, recognition result comparison means for comparing the contents of the recognition result storage means and the newly obtained recognition result, and timekeeping means for measuring the time elapsed since the recognition result was obtained. , and outputs a recognition result when the same recognition result is repeatedly obtained for a predetermined period of time or more. (2) The optical character reading device according to claim 1, wherein the means for measuring time is a counter that counts the number of times a recognition result is repeatedly obtained.