JPH04268680A - Optical reader - Google Patents

Abstract

PURPOSE:To obtain the inexpensive optical reader with low power consumption while improving the operability by shortening the time for obtaining a binary signal capable of decoding and reducing driving clock frequency by adjusting the sensitivity of a light receiving result based on the decision of a detection level decision means. CONSTITUTION:An output signal BS of a line sensor 10 is supplied to a level detection circuit 9. In this detection circuit 9, a reference level Vo centered at the level range capable of binarization of the output signal of the line sensor 10 in a binary circuit 14 without fail is set. The signal BS of the line sensor 10 is envelope-detected, the level is compared with the reference level Vo, and a detection signal LS is outputted. This detection signal LS and a starting signal MST to be generated in a margin decision part 211 are supplied to a pulse width setting circuit. The pulse width of the next start pulse ST is increased or decreased based on the value of the former detection signal LS.

Description

[Detailed description of the invention]

0001

[Field of Industrial Application] The present invention relates to an optical reading device that performs multiple reading scans using a line sensor that receives light from a display optically recorded on a medium and converts it into an accumulated charge amount. In particular, the present invention relates to an optical reading device that has been improved to quickly optimize control of light receiving sensitivity.

0002

[Prior Art] In an optical reading device called a touch-type barcode scanner for reading barcodes attached to barcode labels, CCD sensors and CMOS
A line sensor such as a sensor is used, and the line sensor scans the barcode to obtain an analog signal whose level changes when the bar of the barcode is scanned. This analog signal is binarized by a binarization circuit and then supplied to a decoder to recognize the barcode.

FIG. 9 is a block diagram showing the basic configuration of an example of a conventional touch-type barcode scanner, in which 300 is a line sensor, and 301 is the above-mentioned binarization unit that binarizes the output signal of the line sensor 300. circuit, 302 is a binarization circuit 3
303 is a clock generator, 304 is the above-mentioned decoder that decodes the output signal of
is an m/n-ary counter.

In the figure, a clock generator 303 generates a clock φ1 of a predetermined frequency. This clock φ1 is supplied to the line sensor 300, and m/
It is supplied to an n-ary counter 304. This m/n-based counter 304 functions as a frequency divider, and when the period of clock φ1 is tφ, as shown in FIG.
This clock φ1 every period (that is, a period of ntφ)
A start pulse ST having a pulse width of m cycles (that is, mtφ) is generated. However, n and m are positive integers, and n>m. This start pulse ST is also supplied to the line sensor 300.

The line sensor 300 has an array consisting of a large number of photodiodes, which are light receiving elements, arranged in a line and a number of capacitors, which are power storage elements, corresponding to the number of photodiodes, and is supplied with a start pulse ST. During that pulse period, an amount of charge is accumulated in the photodiode according to the amount of light received from a medium such as a label with a bar code attached. For example, if the line sensor 300 is a CMOS sensor, the clock φ1 is supplied. The stored charge is sequentially read out from each capacitor each time the capacitor is loaded. This results in
Line sensor 300 generates an analog signal BS according to the barcode. That is, the start pulse ST sets the charge accumulation time by the line sensor 300, and the clock φ1
is a driving pulse for sequentially reading out each accumulated charge.

By the way, if the frequency division ratio of the m/n counter 304 is constant, the pulse width of the start pulse ST is constant, and the charge accumulation time in the line sensor 300 is constant, then the type of medium to which the bar code is attached is determined. The amount of accumulated charge varies depending on the distance from the medium to the light receiving surface of the line sensor 300, and as a result, the amplitude of the analog signal BS output from the line sensor 300 varies. On the other hand, the width of the bar of a barcode is not constant, but a plurality of different widths are defined, and the binarization circuit 301 must binarize the analog signal BS so as to represent the width of the bar. For this purpose, a threshold level is set so that the analog signal BS is binarized at the center level of its amplitude.

[0007] When such a line sensor 300 is used, there are variations in the light receiving to charging characteristics within one line sensor, and there are also variations between line sensors, and furthermore, there are Since variations occur in the illumination distribution, in order to set the light reception sensitivity, it is necessary to specify a very limited environment and set a certain detection time. Therefore, even if there is no variation in light reception to charging characteristics within one line sensor, if there are variations in the distance between the medium and the optical reading device or in the reflectance of the medium itself,
Since the amount of light received varies greatly, the amount of charge accumulated will vary as a result. Then, when the amount of accumulated charge that should change in a certain detection time becomes saturated, the analog signal BS also becomes saturated, and the binarization circuit 301 Even with a general automatic threshold control circuit (A.T.C.) that adjusts the threshold depending on the amplitude of the analog signal BS, which is the signal level, the threshold level cannot be set any higher for the saturated analog signal BS. Therefore, even if the display code of the medium includes black display, a binary signal indicating a white state is output during the entire detection time. Similarly, if the amount of accumulated charge in a certain detection time is large and there is little to change, the amplitude of the analog signal BS will also be insufficient, and the binarization circuit 301 will set the general automatic threshold of this binarization circuit 301. Even in the control circuit, the threshold level cannot be set lower than a certain value for the maximum value of the small analog signal BS due to detection sensitivity issues, so even if the display code of the medium includes white display, it is black during the entire detection time. Outputs a binary signal indicating the state.

As described above, when the detection time between each element of the line sensor 300 is fixed, even if an automatic threshold adjustment circuit is added to the binarization circuit 301, the binarization circuit 301
In this case, the binary signal supplied to the decoder 302 is erroneously discriminated in an inverse relationship with respect to reflection and non-reflection of the detected display symbol, that is, the bar code. Even if the erroneous discrimination state of this binary signal is partial, the binary signal supplied at this time will not be correctly decoded by the decoder 302, and the decoder 302 will enter the error determination routine. The bar and space lengths are judged incorrectly or incorrectly.

Therefore, in the conventional touch-type barcode scanner, the charge accumulation time of the line sensor 300 is made variable. Hereinafter, a means for varying the charge accumulation time will be schematically explained with reference to FIG. 11.

In the figure, N accumulation time setting circuits 3
051, 3052, . The decoder 302 decodes the output signal of the binarization circuit 301 and recognizes the barcode, but if the barcode cannot be recognized, a "decoding error" occurs.
The switch 306 is controlled as follows. This causes the switches 306 to operate in a predetermined order, for example, the accumulation time setting circuit 30
51, 3052, . . . to sequentially change the charge accumulation time in the line sensor 300, and search for the charge accumulation time at which the decoder 302 can recognize the barcode. The accumulation time setting circuit 3051 is selected next to the accumulation time setting circuit 305N, and such selection is performed in a ring pattern.

In FIG. 9, in order to switch the charge accumulation time as described above, it is sufficient to sequentially change the frequency division ratio of the m/n-ary counter 304. However, the frequency division ratio here can be expressed as m/n, assuming that the period of the start pulse ST in FIG. 10 is constant (that is, n = constant).
Changing the frequency division ratio means changing m (that is, changing the pulse width of the start pulse ST).

Next, the specific structure of the touch-type barcode scanner described above will be explained with reference to FIGS. 12 and 13. However, in the figure, 1 is a reading signal latch circuit, 2 is an illumination unit, 3 is an optical imaging unit, 4 is a 1-scan counter, 5 is a medium, 6 is a bar code, 7 is a start pulse generation circuit, and 8
1 is a pulse width setting circuit, 10 is a photoelectric conversion unit, 11 and 12 are frequency dividing circuits, 13 is an oscillation circuit, 14 is a binarization circuit, 15 is a scanning number constant memory, 16 is a scanning number counter, and 17 is a scanning number comparison 18 is an edge detection circuit, 19 is a timer counter, 20 is a count value memory, 21 is a count value memory control section, 211 is a margin judgment section, 212 is a first throw judgment section, 213 is an address counter, 21
4 is an address decoder, 215 is a reset circuit, 22 is a bit image converter, 23 is a character code bit image memory, 24 is a stop code bit image memory, 25 is a bit image memory control section, and 26 is a start/stop determination section. , 27 is a character conversion unit, 28
is a character code match comparator, 29 is an error processing circuit, 30 is a constant memory, 31 is a data match counter,
32 is a data matching frequency comparator, and 33 is an output data converter.

[0013] A barcode is a code in which characters such as letters and numbers are coded using multiple bars of different widths, so that a string of characters such as product prices and product names can be represented by an array of bars. It is. The code representing a character is hereinafter referred to as a character code.Although the number of bars representing one character code and the combination pattern of thick bars and thin bars differ depending on the barcode format, all formats are , bar rows representing margins (blank areas), start codes, and stop codes are provided before and after the rows of bars representing all character codes.

In the following, for clarity of explanation, FIG.
2. The format of the barcode read by the barcode scanner shown in Figure 13 is Interleaved 2of5.
shall be. In this system, one character is represented by five black bars, and one character is also represented by five white bars that are spaced between the black bars. One barcode represents a plurality of character codes, each character code being one digit.

In this type of barcode display, the part of the background color of the medium of a predetermined width on the left side of this barcode display is used as a start margin, and the part of the base color of the medium of a predetermined width on the right side of this barcode display is used as a stop margin. The margin is used to determine whether the barcode is displayed. The thin bar at the left end of the barcode display and the thin bar following it represent the start code, and the thin bar at the right end and the thick bar one position to the left represent the stop code. Then, between these start codes and stop codes, a character code is used to determine the display contents by a combination of a predetermined number of thin bars or thick bars (black bars) and thin spaces or thick spaces (white bars). is displayed. Here, assuming that the direction in which the character codes are arranged (that is, from the start code to the stop code) is the forward direction, barcodes are generally read in the above forward direction, but in the reverse direction. In such cases, the stop code is read as if it were a start code, and the start code is read as if it were a stop code. , the stop code pattern is different, and this makes it possible to determine the reading direction of the bar code.

In FIG. 12, the oscillation circuit 13 is always operating, and its output signal is divided by the frequency dividing circuit 11 to form the clock φ1, and by the frequency dividing circuit 12 to form the clock φ2. be done. The clock φ1 corresponds to the clock φ1 in FIG. 9, and the clock φ1 corresponds to the clock φ1 in FIG.
The clock φ2 is supplied to the photoelectric conversion unit 10 corresponding to 00, the 1-scan counter 4, and the start pulse generation circuit 7 corresponding to the m/n counter 304 in FIG.

A read start signal S is received from a host device (not shown).
When the reading signal latch circuit 1 is supplied, the reading signal latch circuit 1 is set and outputs the reading signal READ, which is supplied to the illuminating section 2 and the start pulse generating circuit 7. The illumination section 2 has an LED as a light emitting element, and lights up the LED in response to a reading signal READ to illuminate the printing area of the barcode 6 on the medium 5. The light reflected from this printing area is irradiated onto the light-receiving surface of the line sensor, which is the photoelectric conversion section 10, via the optical imaging section 3. An image is formed. Below, photoelectric conversion section 1
0 is explained as a line sensor.

The start pulse generating circuit 7 is operated under the following four conditions: (1) the scan end signal SED is supplied from the 1-scan counter 4; (2) the reading signal READ is supplied from the reading signal latch circuit 1;
(3) The pulse width setting circuit 8 has completed the pulse width setting and the set value has been sent. (4) The character conversion end signal CED is supplied from the character conversion section 27 (FIG. 13). When all of the following conditions are satisfied at the same time, a start pulse ST is generated. Such conditions are hereinafter referred to as pulse generation conditions.

In the initial state, the one-scan counter 4 receives the scan end signal SED, the pulse width setting circuit 8 receives the initial value set, and the character converter 27 (FIG. 13) receives the character conversion end signal CED. When the reading signal READ is started to be supplied from the reading signal latch circuit 1, the start pulse generation circuit 7 satisfies the pulse generation conditions and starts the pulse width determined by the initial value of the pulse width setting circuit 8. Generate pulse ST. This start pulse ST is supplied to the one-scan counter 4 and the line sensor 10.

The line sensor 10 receives a start pulse ST.
The image on the light receiving surface is changed by 1 every time clock φ1 is supplied.
The barcode 6 is scanned and read pixel by pixel. As a result, the line sensor 10 outputs an electrical signal BS having different levels for the black bar and the white bar of the barcode 6. In addition, the 1-scan counter 4 receives a start pulse ST.
As a result, the clock φ1 starts counting and the scan end signal SED is no longer output, and when the clock φ1 for one scan of the line sensor 10 is counted, the scan end signal SED is output again. This scan end signal SED is supplied to the start pulse generation circuit 7, the scan number counter 16, the count value memory 20, the margin determination circuit 211, and the address counter 213.

The pulse width of the start pulse ST determines the charging time of each capacitor forming a pixel in the line sensor 10, and the line sensor 10
The magnitude of the output signal level is determined. The pulse width of the start pulse ST is determined by the value set in the pulse width setting circuit 8, and this value is determined by the count value NS of the scan number counter 16 that counts the scan end signal SED output from the 1-scan counter 4. It's decided by then. Therefore,
The set value of the pulse width setting circuit 8 is changed every time the line sensor 10 scans, and accordingly, the pulse width of the start pulse ST changes every time the line sensor 10 scans.

When the start pulse generation circuit 7 is supplied with the scan end signal SED again from the 1-scan counter 4 and satisfies the pulse generation conditions, it again generates the start pulse ST with a pulse width according to the setting value of the pulse width setting circuit 8. This causes the line sensor 10 to start scanning and the 1-scan counter 4 to start counting. In this way, the line sensor 10
scans barcode 6 repeatedly.

The output signal BS of the line sensor 10 is shown in FIG.
After the level is binarized by the binarization circuit 14 corresponding to the 1 binarization circuit 301, the rising and falling edges are detected by the edge detection circuit 18. The edge pulse EG output from the edge detection circuit 18 is supplied to a timer counter 19 and an address counter 213. The processing circuits after the edge detection circuit 18 constitute the decoder 302 in FIG.

The timer counter 19 counts the clock φ2 from the frequency dividing circuit 12 using the edge pulse EG as a reset signal. Therefore, the timer counter 19 outputs a count value N representing the width of each bar of the bar code 6. This count value N is determined by the count value memory control section 2.
1 is written into the count value memory 20 controlled by 1.

The count value memory control section 21 is composed of a margin determination section 211, a first throw determination section 212, an address counter 213, an address decoder 214, and a reset circuit 215. Margin judgment section 2
11 is the count value N of the timer counter 19 for the blank area (start margin, stop margin) provided before the first bar and after the last bar of the barcode 6, and the count value of the blank area and the barcode area. The start and end of the barcode are detected by the ratio, and a start signal MST and an end signal MED are generated. Also, start margin,
When the stop margin is insufficient, a margin error signal MER is output and supplied to the reset circuit 215. 1
When the scan end signal SED is supplied from the scan counter 4, the margin determination circuit 211 does not operate. The first throw determination unit 212 detects the magnitude of the count value N output from the timer counter 19, and detects whether the interval between edge pulses EG from the edge detection circuit 18 is too narrow,
If the width is too wide, a first throw error signal is output and supplied to the reset circuit 215. If the barcode 6 printed on the medium 5 becomes narrower or wider due to smudges, scratches, or dust adhesion and is read, the first throw determination unit 212 outputs a first throw error signal. Output. When the margin error signal MER is supplied from the margin determination section 211 or the first throw error signal is supplied from the first throw determination section 212, the reset count 215 generates a reset signal and starts counting with the address counter 213. and the value memory 20.

When the scan end signal SED is no longer supplied from the 1-scan counter 4 and the line sensor 10 starts scanning, the address counter 213 detects the margin determination unit 21.
1, the start signal MST is supplied, and the edge signal EG from the edge detection circuit 18 starts counting from the initial value. The count value output from address counter 213 is decoded by address decoder 214 and supplied to count value memory 20 as address signal ADR. The count value memory 20 enters the write mode when the scan end signal SED is no longer supplied from the 1-scan counter 4, and the count value N output from the timer counter 19 goes to the address specified by the address signal ADR of the count value memory 20. Written sequentially.

When the count value N for all bars of the barcode 6 is written into the count value memory 20 and the end signal MED is output from the margin determination section 211, the address counter 213 stops counting and the initial value is set. be done. Next, after one scan of the line sensor 10 is completed, one
When the scan counter 4 outputs the scan end signal SED, the count value memory 20 enters the read mode, and the address counter 213 outputs the count value sequentially incremented by 1 from the initial value by the internal clock. This count value becomes the address signal ADR of the count value memory 20 in the address decoder 214. Therefore, from the count value memory 20, the written count value N
are read in the order they were written.

Note that when the reset circuit 215 outputs a reset signal, the address counter 213 is reset to the initial value and the count value memory 20 is cleared. The margin determination unit 211 continues to output the margin error signal MER until the next margin determination, and the first throw determination unit 212 continues to output the first throw error signal until the next scan end signal SED is supplied from the 1-scan counter 4. continue.

The count value N outputted from the count value memory 20 is supplied to a bit image converter 22, where it is compared with a preset threshold value and converted into a bit image BI representing the type of bar.

Next, FIG. 13 will be explained. The bit image BI from the bit image converter 22 is supplied to a character code bit image memory 23, a stop code bit image memory 24, and a bit image memory controller 25. Bit image memory control section 2
5 determines whether the bit image BI is a bit image forming a character code (character code bit image) or a bit image forming a stop code (stop code bit image) by determining the output order of the bit image BI. bit image B
When I is a character code bit image,
The bit image BI is divided into five pieces, each of which is 1 byte of data, and these data are sequentially written into the character code bit image memory 23 to create the bit image BI.
When is a stop code bit image, it is written into the stop code bit image memory 24. In the case of Interleaved 2 of 5, bit images for black bars and white bars are separated, and 1 byte of data is formed for each and written to the character code bit image memory 23. When the writing in the character code bit image memory 23 and the stop code bit image memory 24 is completed, the bit image memory control section 25 sequentially reads out the 1-byte data CCD from the character code bit image memory 23 and writes the data into the character code bit image memory 23 and the stop code bit image memory 24. Also, the stop code bit image SCD is read from the stop code bit image memory 24 and sent to the start/stop determining section 26.

In the start/stop determining section 26,
For the correct stopcode used in the barcode,
A bit pattern (hereinafter referred to as a registered stop code pattern) consisting of two types of bit images obtained when correctly read in the forward and reverse directions is registered in the ROM, and two bit patterns from the stop code bit image memory 24 are stored in the ROM. The bit pattern of the image data SCD (hereinafter referred to as a detected stop code pattern) is compared with the registered stop code patterns to determine which registered stop code pattern it matches. When there is a registered stopcode pattern that matches this detected stopcode pattern, the reading direction of the barcode 6 can also be determined based on this, and the data (stopcode data) corresponding to this matched registered stopcode pattern can be determined. S
SC is supplied to the output data converter 33, and a conversion direction instruction signal CDD is sent to the character converter 27.

When there is no registered stop code pattern matching the detected stop code pattern SCD, the start/stop determining section 26 outputs an error signal ERR1 and supplies it to the error processing circuit 29.

In the character converter 27, a 1-byte bit pattern (hereinafter referred to as registered character) containing a bit image obtained when correctly read in the forward direction is converted to all character codes used in the bar code. (called a code pattern) is registered in the ROM,
The 1-byte data CCD bit pattern from the character code bit image memory 23 (hereinafter referred to as the detected character code pattern) is compared with the registered character code patterns, and it is determined which registered character code pattern it matches. Ru. In this case, when the barcode 6 is read in the reverse direction, the conversion direction instruction signal CDD from the start/stop determining section 26 causes
The detected character code pattern CCD is reversed and compared with the registered character code pattern.

[0034] When the detected character code pattern CCD matches any of the registered character code patterns,
The character converter 27 outputs character data CD for the matched registered character code pattern, and sends it to the character code match comparator 28 and the output data converter 33. When the detected character code pattern CCD does not match any registered character code pattern, the character converter 27 generates an error signal ERR2 and sends it to the error processing circuit 29. Further, when all of the detected character code patterns CCD for one scan of the barcode 6 are converted into character data CD, the character conversion section 27 generates a character conversion end signal CED, and generates a character conversion end signal CED as shown in FIG.
The start pulse generating circuit 7 of No. 2 is supplied. As a result, this start pulse generation circuit 7 generates a start pulse ST
The line sensor 10 (Fig. 12) generates the barcode 6.
Perform the next read scan.

The character code match comparator 28 receives one scan of character data CD from the character converter 27.
and compares this with the character data CD supplied from the character converter 27 in the next scan,
When all character data match, a match pulse is output and supplied to the data match counter 31. The data matching number counter 31 counts the matching pulses, and this count value is compared with the matching number set value stored in the constant memory 30 by the data matching number comparator 32. When the count value of the data matching number counter 31 exceeds the matching number setting value, the data matching number comparator 32 generates a read completion signal REND1 and sends it to the output data converter 33. If even one character data CD from the character converter 27 between the previous scan and the current scan does not match, the character code match comparator 28 outputs an error signal ERR3 and sends it to the error processing circuit 29, and also detects a data match. Clear the number counter 31.

The output data conversion unit 33 takes in the character data CD from the character conversion unit 27 and the stop code data SSC from the start/stop determination unit 26 every time the barcode 6 is scanned, and holds the latest data. When the read completion signal REND1 is supplied from the data matching frequency comparator 32, the character data CD and stop code data SSC held are converted into a predetermined format and sent to the host device (not shown). . At the same time, it sends a display instruction signal DIS to indicate the completion of reading to a display device (not shown) such as a lamp or buzzer, and also generates a reset signal RST to initialize itself. This reset signal RST is supplied to the reading signal latch circuit 1 (FIG. 12), etc., and resets them. When the reading signal latch circuit 1 is reset by the reset signal RST, it stops outputting the reading signal READ and stops reading the bar code 6. The illumination unit 2 is also turned off.

In this way, when the character data from the previous and subsequent scans match consecutively a certain number of times (for example, twice) according to the predetermined number of times set value in the constant memory 30, it is assumed that the bar code 6 has been read correctly and the reading ends. However, if this number of times does not match consecutively, reading of the bar code 6 is repeated. However, even if reading is repeated a certain number of times, the reading completion signal REN is not output from the data matching number comparator 32.
If D1 is not output, further reading of the barcode 6 is prohibited.

That is, in FIG. 12, the scanning number counter 16 counts up by 1 each time the scanning counter 4 outputs the scanning end signal SED. The count value of the scanning number counter 16 represents the number of scanning of the line sensor 10, and the scanning number constant memory 1
It is compared with a constant set to 5. When the count value of the scanning number counter 16 becomes equal to or greater than this constant, the scanning number comparator 17 outputs the reading end signal REND2, as shown in FIG.
The output data is supplied to the output data converting section 33 of.

Therefore, the output data converter 33 outputs the reset signal RST to initialize itself without sending character data or stop code data to the host device. As a result, the reading of the barcode 6 is deemed to have failed and the reading of the barcode 6 is stopped. Also,
A display instruction signal DIS indicating unreadability is sent to the display device.

The error processing circuit 29 receives an error signal ERR.
1, ERR2, and ERR3, the character conversion reset signal CERT is output, and the count value memory control unit 21 and bit image conversion unit 2 in FIG.
2 and character code bit image memory 2 in Figure 13.
3. Stop code bit image memory 24, bit image memory control section 25, start/stop determination section 2
6. Initialize the character conversion section 27 and the like. The character conversion unit 27 thereby outputs a character conversion end signal CED.

As described above, the line sensor 10 reads and scans the barcode 6 multiple times, and when the character codes match consecutively a number of times equal to the number of matches set in the constant memory 30, the barcode 6 is read correctly. This character code is sent to the host device to complete the reading operation, but during this time, the pulse width set by the pulse width setting circuit 8 is changed for each scan of the line sensor 10, and along with this, the , the pulse width of the start pulse ST output from the start pulse generation circuit 7 changes.

FIG. 14 is a block diagram showing an example of the pulse width setting circuit 8 in FIG. 12, in which 81 and 82 are constant memories, 83 is a multiplier, 84 is an adder, and 85 is a set value memory.

In the figure, a constant memory 81 stores a value (reference value) that determines the pulse width (reference pulse width) that should be set at the beginning of the start pulse ST generated by the start pulse generation circuit 7 (FIG. 12). is stored, and the constant memory 82 stores a value for a pulse width that is 1/16 times the reference pulse width (1/16 value).

When the power is turned on to the barcode scanner or from the output data converter 33 in FIG.
When ST is output, the reference value of constant memory 81 is loaded into set value memory 85, read out, and supplied to start pulse generation circuit 7 and adder 84 in FIG. Therefore, the pulse width of the start pulse ST that is first output from the start pulse generation circuit 7 corresponds to this reference value.

When the scanning of the line sensor 10 (FIG. 12) is completed, the 1-scan counter 4 (FIG. 12) outputs the scanning end signal SED.
Each time , the multiplier 83 multiplies the 1/16 value stored in the constant memory 82 by the count value Ns of the scanning number counter 16 (FIG. 12) at the timing of the previous edge of the scanning end signal SED. , an adder 84 adds the output value of the multiplier 83 and the setting value read from the setting value memory 85. In the set value memory 85, the set value stored there is rewritten with the output value of the adder 84 with a slight delay from the previous edge of the scan end signal SED, and this is immediately read out and sent to the start pulse generating circuit 7 ( 12) and the adder 84.

For this purpose, the set value read from the set value memory 85 is initially equal to the reference value stored in the constant memory 81, but each time the line sensor 10 (FIG. 12) finishes scanning, this reference value is The value increases by 1/16 times the value. Therefore, the start pulse generating circuit 7 (FIG. 12) first generates the start pulse ST with the reference pulse width.
However, each time the line sensor 10 scans, the pulse width of the start pulse ST becomes 1/1 of the reference pulse width.
It will become longer by 6 times the width.

Next, the line sensor 10 shown in FIG. 12 will be explained. FIG. 15 is a configuration diagram of this line sensor 10, in which 101 is a pixel section, 102 is a readout shift register, 103 is a sweep shift register, 104 is a sample and hold circuit, 104a is a preamplifier, 104b is a sample switch, and 104c is a A capacitor, 104d is a buffer amplifier, 105 is an 8 frequency divider circuit, 106 is a sweep-out latch circuit, 107 is a read-out latch circuit, D1 to Dn
are photodiodes with PN junctions, S11 to S1n,
S21 to S2n are analog switches using MOSFETs. Further, FIG. 16 is a diagram showing the timing relationship of signals of each part in FIG. 15.

In FIGS. 15 and 16, one pixel is composed of one photodiode Di (where i is 1, 2, . . . , n) and a pair of analog switches S1i, S2i. The pixel unit 101 has n such pixels1.
arranged in columns. The analog switch S1i is controlled on and off by the sweep shift register 103, and the analog switch S2i is controlled on and off by the read shift register 102.

Clock φ from frequency dividing circuit 11 (FIG. 12)
1 is divided by an 8 frequency divider circuit 105, and the duty ratio is 1/2.
A shift clock φSH with a duty ratio of 1/16 and a sampling clock φSP with a duty ratio of 1/16 are formed. The sample switch 104b of the sample hold circuit 104 is turned on during the pulse period of the sampling clock φSP. Shift clock φSH is supplied to read shift register 102, sweep shift register 103 and read latch circuit 107.

The start pulse ST supplied from the start pulse generating circuit 7 (FIG. 12) is an "L" (low level) pulse, and the read latch circuit 107 is set at the rising edge (rear edge) of the start pulse ST. It is reset at the rising edge of the shift clock φSH immediately after the lever. The output PRD of this read latch circuit 107 is
Therefore, it is an "L" pulse with a pulse width from the trailing edge of the start pulse ST to the immediately following shift clock φSH. This pulse PRD is supplied to the read shift register 102. The sweep latch circuit 106 is set at the falling edge (front edge) of the start pulse ST, and reset at the rising edge (rear edge) of the output pulse PRD of the read latch circuit 107. Therefore, the output PEX of the sweep latch circuit 106 is an "L" pulse from the front edge of the start pulse ST to the rear edge of the output pulse PRD of the read latch circuit 107.

When the read latch circuit 107 and the sweep latch circuit 106 are in the reset state, the “L” pulse PR
When D and PEX are not output, all outputs of the read shift register 102 and sweep shift register 103 are "H" (high level), and each analog switch S2i for reading (where i = 1 to n) is off, and each sweep analog switch S1i is on. Such a state is shown in FIG. As a result, a constant bias voltage +V is applied to the cathode side of each photodiode Di via the analog switch S1i, and an amount of charge determined by the capacitance and the bias voltage +V is accumulated in each photodiode Di. This state is the sweeping state.

When the start pulse ST is supplied and the sweep latch circuit 106 outputs the "L" pulse PEX at the timing of the previous edge, the sweep shift register 10
3, this pulse PEX is sequentially transferred by the shift clock φSH, and thereby the analog switch S1
1, S12, ..., S1n are sequentially shift clocks φSH
It turns off with a delay of one period. The period during which each analog switch S1i is turned off is longer than the pulse width of the output pulse PEX of the sweep latch circuit 106 and is the shortest period among integral multiples of the period of the shift clock φSH. In this way, the state shown in FIG. 18 in which the analog switch S1i is off while the analog switch S2i is off is a pixel charge accumulation state, and the photodiode Di has a large amount of charge depending on the intensity of light incident thereon. A photocurrent flows, and the charge on the capacitor at the PN junction is discharged.

When the “L” pulse PRD is output from the read latch circuit 107 at the edge after the start pulse ST, the read shift register 102 outputs this pulse PR.
D is sequentially shifted by the shift clock φSH, and the analog switches S21, S22, . Such a state at the i-th pixel is shown in FIG. This state is a read state, and the accumulated charge of the photodiode Di is read out via the analog switch S2i and supplied to the sample and hold circuit 104. Analog switch S2i
The period during which the analog switch S1i is turned on corresponds to one cycle of the shift clock φSH at the end of the off period of the analog switch S1i, and the analog switches S1i and S2i are turned off at the same time to return to the sweeping state shown in FIG. Therefore, photodiode D
The amount of charge read from i corresponds to the amount of remaining charge in the capacitor discharged in the state shown in FIG. 18, and corresponds to the intensity of the received light.

In this way, the photodiode D1
, D2, . The pixel after this reading is in the sweep state shown in FIG. 17, and the bias voltage +V is applied to the cathode of the photodiode Di to perform charging again.

Signal B based on charges read out from each pixel
In the sample hold circuit 104, S' is amplified by the preamplifier 104a and then outputted to the sampling clock φ.
The output signal B of the line sensor 10 is sampled and held by a sample switch 104b driven by SP and a capacitor 104c, and is outputted via a buffer amplifier 104d.
It becomes S.

As described above, in the line sensor 10, the pixel charge accumulation time in each pixel varies depending on the pulse width of the output pulse PEX of the sweep latch circuit 106, and therefore the pulse width of the start pulse ST. , the amount of charge read out varies depending on the pixel charge accumulation time. For this reason, by changing the pulse width of the start pulse ST, the magnitude of the output signal BS of the line sensor 10 can be changed.

By the way, if the start pulse ST is input before reading is completed for all pixels, reading will be performed simultaneously for two different pixels, and normal reading will not be performed. For this purpose, in FIG.
A scanning counter 4 is provided, whereby the line sensor 10 detects that reading of all pixels has been completed, and the start pulse generation circuit 7 generates a start pulse ST based on this detection. There is.

[0058]

[Problems to be Solved by the Invention] In the conventional barcode scanner described above, the pulse width of the start pulse ST is sequentially lengthened every time the line sensor 10 scans, and the pixel charge accumulation time in the line sensor 10 is sequentially increased. The increasing action is repeated. According to this, the following problems arise.

That is, after startup, the pulse width of the start pulse ST gradually increases from the reference pulse width until it reaches a pulse width that allows good barcode reading. If there happens to be a barcode reading error when the width of the start pulse ST becomes shorter, and the output signal BS of the line sensor 10 becomes impossible to decode, the pulse width of the start pulse ST continues to increase. , it is necessary to wait until the start pulse ST reaches a pulse width such that a good bar code can be read again. In other words, it takes a long time to read the barcode, which poses a problem in terms of operability.

In order to solve this problem, line sensor 1
The time required for one scan of 0 may be shortened, but for this purpose the frequency of the output clock of the oscillation circuit 13 must be increased. However, increasing the clock frequency increases power consumption in the oscillation circuit 13, which is particularly problematic in portable touch-type barcode scanners, which have a built-in battery as a power source. There are also problems in terms of noise countermeasures and line sensor sensitivity.

An object of the present invention is to solve this problem, and without increasing the frequency of the drive clock of the line sensor.
An object of the present invention is to provide an optical reading device that can quickly set an appropriate pixel charge accumulation time in a line sensor.

[0062]

[Means for Solving the Problems] In order to achieve the above object, the present invention provides level detection means for detecting the magnitude of the output level of a line sensor, and a pulse of a start pulse according to the detection output of the level detection means. A pulse width setting means for increasing or decreasing the width is provided, and the amount of increase and amount of decrease in the pulse width of the start pulse are made different.

[0063]

[Operation] Regardless of whether or not the output signal of the line sensor can be decoded, the output level of the line sensor is detected by the level detection means, and the start pulse is set to a pulse width according to this detection result. be done. For this reason, once the output level of the line sensor reaches an appropriate level for binary conversion, the output level of the line sensor is maintained at that appropriate level even if decoding is impossible.

Furthermore, even if the output level of the line sensor deviates from the appropriate level at the time of startup, the level detection means and the pulse width setting means will change this output level significantly and approach the appropriate level. When it approaches the level, it changes slightly and is set near the appropriate level.

[0065]

Embodiments Hereinafter, embodiments of the present invention will be explained with reference to the drawings. 1 and 2 are block diagrams showing an embodiment of an optical reading device according to the present invention, in which 8' is a pulse width setting circuit;
Reference numeral 9 denotes a level detection circuit, and parts corresponding to those in FIGS. 12 and 13 are given the same reference numerals and redundant explanations will be omitted.

In FIG. 1, the output signal BS of the line sensor 10 is also supplied to the level detection circuit 9. In this level detection circuit 9, the output signal B of the line sensor 10 is
A reference level V0, which is the center of the level range in which S can be reliably binarized by the binarization circuit 14, is set, and the output signal BS of the line sensor 10 is envelope-detected, and the envelope level and the reference level V0 are The result is “L” when envelope level ≦V0.
, a detection signal LS that becomes "H" when the envelope level>V0 is output. This detection signal LS and the start signal MST generated by the margin determining section 211 are supplied to the pulse width setting circuit 8'.

The pulse width setting circuit 8' determines the pulse width of the start pulse ST generated by the start pulse generation circuit 7, and when this optical reading device is in the initial state, the pulse width is set in the level detection circuit 9. In other words, when reading the ideal barcode 6, the envelope level of the output signal BS of the line sensor 10 reaches this reference level V0.
The pulse width of the start pulse ST is determined so as to match the pulse width, and the initial value that produces this pulse width is set in the pulse width setting circuit 8' and sent to the start pulse generation circuit 7 as the pulse width setting value NST. Therefore,
The pulse width of the start pulse ST that is first output from the start pulse generation circuit 7 corresponds to this initial value.

The barcode 6 is read by the line sensor 10, and the envelope level of the output signal BS is compared with the reference level V0 in the level detection circuit 9. When the envelope level≦V0, the detection signal L is “L”.
When s is output from the level detection circuit 9, the pulse width setting circuit 8' receives the start signal MS from the margin determination section 211.
At timing T, a value larger than the previously set value is output as the pulse width setting value NST, and the pulse width of the next start pulse ST is increased. On the other hand, when the envelope level>V0 and the "H" detection signal Ls is output from the level detection circuit 9, similarly,
The pulse width setting circuit 8' makes the pulse width setting value NST smaller than before, and reduces the pulse width of the start pulse ST.

In this way, when the envelope level of the output signal BS of the line sensor 10 goes up or down with respect to the reference level V0 of the level detection circuit 9, the pulse width setting circuit 8'
Accordingly, the pulse width of the start pulse ST increases or decreases. As a result, the envelope level of the output signal BS of the line sensor 10 fluctuates near the reference level V0, and this output signal BS is reliably binarized by the binarization circuit 14, but in this case, the pulse width setting The amount of increase and amount of decrease in the pulse width of the start pulse ST due to the change in the pulse width setting value NST of the circuit 8' are set to be different. This is due to the following reason.

Even if the above-mentioned initial value is set in the pulse width setting circuit 8' in the initial state of the optical reading device, the line sensor 10 may be The envelope level of the output signal BS is the reference level V of the level detection circuit 9.
It may be significantly different from 0. In such a case, the pulse width setting value NST output by the pulse width setting circuit 8' changes sequentially every time the line sensor 10 scans the barcode 6, but the pulse width setting value NST output by the pulse width setting circuit 8' changes sequentially every time the line sensor 10 scans the barcode 6. In order to shorten the period, it is better to increase the amount of change in the pulse width setting value NST in this case. After that, if the envelope level jumps over the reference level V0, the envelope level will change toward the reference level V0, but by reducing the amount of change, the envelope level will not deviate greatly from the reference level V0. . As a result, when the envelope level exceeds the reference level V0 again, the envelope level changes greatly and exceeds the reference level V0 again, but returns to the reference level V0 again. In this way, the output signal BS of the line sensor 10 quickly and reliably reaches a level that can be binarized from the initial state, and this state is maintained thereafter.

Note that the portion shown in FIG. 2 is the same as the conventional portion shown in FIG. 13, so a description thereof will be omitted.

FIG. 3 shows changes in the envelope level due to the above operation, and BS(0) represents the envelope level in the initial state (first barcode scanning of the line sensor 10). Further, the amount of increase in the envelope level is twice the amount of decrease. In this case,
When the envelope level changes from BS(0) to BS(1) and then to BS(2), it becomes larger than the reference level V0. Therefore, the envelope level is BS(3)
When it decreases to BS(4) and then decreases to BS(4), the reference level V
becomes smaller than 0. Therefore, the envelope level increases again to BS(5). Below, line sensor 1
This operation is repeated as long as 0 repeats the reading scan of the bar code 6, and the envelope level of the output signal BS of this line sensor 10 is near the reference level V0.

As described above, in this embodiment, while detecting the output level of the line sensor 10 and determining whether or not this is a suitable level for binarization,
Since the output level of 0 is adjusted, even if decoding happens to become impossible in a good binarization processing state, this state will be maintained and decoding will be performed immediately. Moreover, the time required from the initial state until the binarization process is reliably performed can be shortened. Therefore, the bar code can be read quickly without increasing the frequency of the driving clock as in the prior art, improving operability and preventing an increase in power consumption.

FIG. 4 shows the pulse width setting circuit 8' in FIG.
14 is a block diagram showing a specific example, in which 86 and 87 are constant memories, 88 is a switch, and 89 is a control circuit, and parts corresponding to those in FIG. 14 are given the same reference numerals.

In the figure, the above initial value is stored in constant memory 81, and a value 1/8 times this initial value (hereinafter referred to as 1/8 value) is stored in constant memory 86. 87 has a value -1/16 times this initial value (hereinafter -
1/16 value) are stored respectively. Control circuit 8
9 is a start signal MS from the margin determination section 211 (FIG. 1) in response to the detection signal Ls from the level detection circuit 9 (FIG. 1).
It operates at timing T and controls the switch 88 to set the value of 1/8 of the constant memory 86 or -1/ of the constant memory 87.
16 values are provided to adder 84. The adder 84 adds this and the pulse width setting value NST from the setting value memory 85 and writes it to the setting value memory 85 to obtain the pulse width setting value NST.
Update.

In the initial state, the initial value of constant memory 81 is written into setting value memory 85, and this initial value is supplied as pulse width setting value NST to adder 84 and start pulse generating circuit 7 (FIG. 1).

When the detection signal Ls is "L", the control circuit 89 causes the switch 88 to select the 1/8 value of the constant memory 86 at the timing of the start signal MST, and the adder 84 selects this and the pulse width setting value. NST is added and written into the setting value memory 85, and the pulse width setting value NST is increased by a value 1/8 times the initial value. Detection signal Ls is “H”
Similarly, the switch 88 stores the constant memory 8.
The adder 84 adds this to the pulse width setting value NST and writes the result into the setting value memory 85. As a result, the pulse width setting value NST becomes 1/1 of the initial value.
It becomes smaller by 6 times the value.

In the embodiment described above, the level detection circuit 9
The output level of the line sensor 10 was determined by setting a reference level V0 to the output signal BS of the line sensor 10.
A level range that can be reliably binarized may be set, and the output level of the line sensor 10 may be determined based on this. Examples for this purpose will be described below.

FIG. 5 shows an example of such an embodiment, in which 7' and 8'' are counters, 9' is an amplitude comparator, and 34
1 and 2 is a decoder consisting of a processing circuit after the edge detection circuit 18, and 35 is a decoder.

In FIG. 5, the amplitude comparator 9' corresponds to the level detection circuit 9 of FIG.
correspond to the pulse width setting circuit 8' and the start pulse generation circuit 7 in FIG. 1, respectively.

The output signal BS of the line sensor 10 is binarized by the binarization circuit 14, and is supplied to the decoder 34, as well as to the amplitude comparator 9', where it is binarized by the binarization circuit 14. When the “H” pulse is output, the preset reference values VH and VL (however,
The amplitude is compared with VH>VL). These reference values VH, V
L determines the appropriate amplitude range of the output signal BS that can satisfactorily decode the output signal of the line sensor 10, and when VH>BS>VL, the amplitude of the signal BS is appropriate. The threshold value VTH of the binarization circuit 14 satisfies VTH<VL.

The amplitude comparator 9' detects at least the output of the line sensor 10 when the signal BS has a small amplitude and no "H" output pulse is output from the binarization circuit 14, or at the timing of this "H" pulse. When the signal BS has a portion where VL≧BS, the up signal UP is output, the count value of the counter 8" is multiplied by K1 (>1), and when the signal BS has a large amplitude, the " BS to the output signal BS of the line sensor 10 at the timing of the output pulse of "H".
When there is a part where ≧VH, the down signal DOWN
is output and the count value of the counter 8'' is multiplied by 1/K2 (K2>1), and when VH>BS>VL, no signal is output and the counter 8'' is held at the same count value. These output signals UP of the amplitude comparator 9',
DOWN is cleared by start pulse ST.

The counter 7' is preset with the count value of the counter 8'', and the clock φ is calculated from this preset value.
Count 1. This count value is supplied to the decoder 35. The decoder 35 outputs an "L" signal for a period from the preset value to the first specified value, and outputs an "H" signal when the count value of the counter 7' exceeds the first specified value. This “L” output from the decoder 35
The signal is the start pulse ST. At the rising edge (rear edge) of this start pulse ST, the counter 7' loads the count value of the counter 8'' as a preset value, and when the clock φ1 is counted up to the second specified value, the clock φ1 is started again from this preset value. Each time the counter 7' starts counting clock φ1 from the preset value, a start pulse ST is output from the decoder 35, and the pulse width of this start pulse ST is determined by the edge timing after the previously generated start pulse ST. Counter 8 preset to counter 7'
” according to the count value.

Here, the count value of the clock φ1 from the preset value of the counter 7' to the above-mentioned first specified value is expressed as m.
, the clock φ from the preset value to the second specified value
When the count value of 1 is n and the period of clock φ1 is Tφ, the pulse width of the start pulse ST is m·Tφ and the period is n·Tφ, both of which differ depending on the preset value of the counter 7'. However, (n-m) is constant and corresponds to the number of pixels of the line sensor 10.

Now, if the counter 7' is an up counter and its preset value is a negative value and the first specified value is zero, then when this preset value becomes K1 times as high, the pulse of the start pulse ST The width becomes K1 times, and when the preset value becomes 1/K2 times, the pulse width of the start pulse ST also becomes 1/K2 times.

In the initial state, the initial value is stored in the counter 8'', and this initial value is preset in the counter 7'.Thereby, the decoder 35 starts the reference pulse width according to this initial value. The pulse ST is output, and after that, the counter 8'' is output every time the line sensor 10 scans.
The count value changes and the pulse width of the start pulse ST increases or decreases.

Here, it is assumed that K1 >K2. Assuming that the initial value set in the counter 8'' is the minimum pulse width of the start pulse ST, in the first scan of the line sensor 10, the amplitude comparator 9' outputs an up signal UP, and the counter 8'' Multiply the count value by K1. Then, the count value is successively multiplied by K1 until BS>VL. In this case, as the count value increases, the amount of change in the count value, ie, the amount of change in the pulse width of the start pulse ST, gradually increases. As a result of multiplying the count value of counter 8'' by K1, VH>BS>VL does not hold.
If BS≧VH directly, the amplitude comparator 9' outputs a down signal DOWN and multiplies the count value of the counter 8'' by 1/K2. In this case, the count value obtained by multiplying by 1/K2 is It is larger than the original count value multiplied by K1 immediately before.For example, if the count value of counter 8'' is now N(0) and this is multiplied by K1, the count value becomes K1·N(0). This count value K
1 ・When N(0) is multiplied by 1/K2, K1 /K2
>1, so K1 ・N(0)×1/K2 >N(
0). That is, as a result of sequentially increasing the pulse width of the start pulse ST, the output signal BS of the line sensor 10
If the amplitude of the pulse exceeds the proper amplitude range defined by the amplitude comparator 9', the pulse width is decreased by an amount less than the last increase. From this, the start pulse S
The pulse width of T will quickly fall into an appropriate range.

FIG. 6 shows changes in the preset value of the counter 9', where the initial value N(0) is set to 2 and the level of the output signal BS of the line sensor 10 is set to the reference level VL to VH.
The range of preset values to be within the range determined by
5, K1 = 3, and K2 = 2. In this case, N(1)=6, N(2)=3, N(2)=3, N
(3)=9, N(4)=4.5.

FIG. 7 shows an embodiment modified from FIG. 5, in which only the down signal DOWN is output from the amplitude comparator 9', and the up signal UP is formed from the decoding error signal of the decoder 34. This is what I did. That is, when the level of the output signal BS of the line sensor 10 is low, the output signal of the amplitude comparator 9' is "L", and the down signal DOWN is not output. this"
The output signal of "L" becomes "H" by the inverter 36 and is output to the AND gate 37. At this time, the line sensor 1
When the level of the output signal BS of 0 is too low and the decoder 34 cannot decode the output of the binarization circuit 14, the decoder 34 outputs a decoding error signal DER of "H" to the AND gate 37. As a result, and gate 37
An "H" signal is output from the counter 8 and is supplied to the counter 8 as an up signal UP.

Furthermore, as shown in FIG. 8, the amplitude comparator 9'
Only the reference level VL is set and only the up signal UP is supplied to the counter 8'', and the down signal DOWN to the counter 8'' is formed from the decoding error signal DER output from the decoder 34 as shown in FIG. You can do it like this.

Note that the decoding error signal DER in FIGS. 7 and 8 is the error processing circuit 29 in FIG.
The character conversion reset signal CERT outputted by CERT can be used.

Furthermore, the numerical values given in the description of each of the above embodiments are merely examples, and the present invention is not limited to these numerical values.

[0093]

Effect of the Invention The present invention sets the pulse width control signal so that the variable amount of the pulse width is different between the amount of decrease and the amount of increase based on the magnitude determination of the detection level determination means for the light reception result. Since the sensitivity is controlled and adjusted, it does not take long to obtain a decodable binary signal, and the time required for successful decoding is shortened, significantly improving operability and increasing the drive clock frequency. Since there is no need to do so, it is possible to provide an optical reading device with low power consumption and low cost.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing a part of a first embodiment of an optical reading device according to the present invention.

FIG. 2 is a block diagram showing the remaining parts of the first embodiment of the optical reading device according to the invention;

FIG. 3 is a diagram showing changes in the output level of the line sensor in FIG. 1;

FIG. 4 is a block diagram showing a specific example of the pulse width setting circuit in FIG. 1;

FIG. 5 is a block diagram showing the main parts of a second embodiment of the optical reading device according to the present invention.

FIG. 6 is a diagram showing changes in the preset value of the counter in FIG. 5;

FIG. 7 is a block diagram showing the main parts of a third embodiment of the optical reading device according to the present invention.

FIG. 8 is a block diagram showing the main parts of a fourth embodiment of an optical reading device according to the present invention.

FIG. 9 is a block diagram showing essential parts of an example of a conventional optical reading device.

FIG. 10 is a timing chart showing the relationship between the clock and start pulse in FIG. 9;

FIG. 11 is a block diagram showing the main parts of another example of a conventional optical reading device.

12 is a block diagram specifically showing a part of the optical reading device shown in FIG. 11. FIG.

13 is a block diagram specifically showing other parts of the optical reading device shown in FIG. 11. FIG.

14 is a block diagram showing an example of the pulse width setting circuit in FIG. 12. FIG.

FIG. 15 is a diagram showing the configuration of a line sensor in the optical reading device.

16 is a timing chart showing signals of each part in FIG. 15. FIG.

FIG. 17 is a diagram showing a pixel sweep state in FIG. 15;

18 is a diagram showing the signal charge accumulation state of the pixel in FIG. 15. FIG.

FIG. 19 is a diagram showing a read state of pixels in FIG. 15;

[Explanation of symbols]

6 Barcode 7 Start pulse generation circuit 7’ counter 8’ Pulse width setting circuit 8” counter 9 Level detection circuit 9’ Amplitude comparator 10 Line sensor 14 Binarization circuit 34 Decoder

Claims (1)

[Claims] Claim 1: A display section optically recorded on a medium;
The amount of light from the display section is received by a plurality of light-receiving elements, charges are accumulated in a number of power storage elements corresponding to the number of light-receiving elements, and a detection signal of a level corresponding to the amount of charge stored in the power storage elements is generated. A line sensor that sequentially outputs a line sensor, a start pulse generating means that supplies a start pulse signal to the line sensor that specifies the accumulation time from each of the light receiving elements by pulse width, and an operation reference time of the start pulse generating means is specified. Also, read operation reference signal generation means for specifying a read reference time for sequentially reading out the detection signals from the line sensor, and the detection output from the line sensor based on a start pulse signal of the start pulse generation means. detection level determination means for determining the magnitude of the level by comparing the level of the signal with a reference value; and reducing the pulse width of the start pulse output by the start pulse generation means based on the magnitude determination result of the detection level determination means. a pulse width setting means for increasing the pulse width, and a pulse width control signal for causing the pulse width setting means to set the start pulse corresponding to the large judgment and small judgment of the detection level judgment means with different pulse width reduction and increase amounts. and a pulse width control means for outputting,
The line sensor is characterized in that it is configured to cause the line sensor to perform a plurality of reading scans with the pulse width of the start pulse in accordance with pulse width control signals of different decrease and increase amounts corresponding to the magnitude determination of the detection level determination means. Optical reader.