JPH0342654A - Original size detecting device for image reader - Google Patents

Abstract

PURPOSE:To surely detect original size by a simply structured device by providing dimension determining means which determine the leading and trailing edges of an original in the main scanning direction and original width according to an original presence or absence signal outputted from an original presence or absence deciding means. CONSTITUTION:A white plate 4 is in the reading position of the original 2, is arranged behind the original 2, and has reflectance higher than that of the brightest part of the image surface 2a of the original 2. The original presence or absence deciding means 63 detects the level difference between received light output from the white plate 4 of a received light sensor 6 and received light output from the image surface 2a of the original 2, and decides the presence of the original if the level difference is low. According to the original presence or absence signal outputted from the original presence or absence deciding means 63, the dimension determining means 66 and 67 determine the leading and trailing edges of the original 2 in the main scanning direction and original width. Thus, the simply structured device can surely detect original size.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a document size detection device for an image forming apparatus such as a small copying machine or a digital copying machine, or an image reading apparatus such as a facsimile machine or an image scanner.

[Conventional technology]

In general, in the above-mentioned image reading device, problems occur due to unevenness in illumination light, non-uniform light intensity due to Keller in the imaging optical system, or variations in sensitivity between multiple light-receiving cells such as embedded sensors and image sensors. Shape ink is unavoidable, and as a result, thin lines disappear, white spots blur, etc., resulting in a loss of image quality.

Therefore, by reading a reference white plate that provides uniform reflectance in advance, detecting and placing the unevenness and ripples, and correcting them during actual image reading, an electrical signal of an image with uniform sensitivity is obtained. Shape ink correction was performed.

Furthermore, when reading an image, it is necessary to detect the size of the document, especially its width, and the positions of the left and right edges of the document in order to correct the case where the center of the document and the center of reading of the image reading device are misaligned.

However, with conventional white or light-colored backs or rollers, it is difficult to distinguish between the edges of the document and the margins.
For example, as proposed in Japanese Utility Model Application No. 63-108257, the contrast with the margins of the document is emphasized by using a black roller, and the starting and ending edges of the document in the width direction are detected from the image reading output. There was a document size detection device that determined the width of the document.

[Problem to be solved by the invention]

However, in such a document size detection device that uses a black roller, when performing shading correction, it is necessary to separately prepare a reference white plate (or white roller) and move it to the reading position, which makes it difficult for the device to perform shading correction. There was a problem that the process became complicated and the cost increased by 1~.

Furthermore, this document size detection device is based on the premise that the margins, or edges, of the document are white or light-colored.Documents with black borders, documents with many dark photographs on the edges, or thick books There is a serious problem in that the correct size cannot be detected in the case of a document with black edges, which tends to occur when copying.

The present invention has been made in view of the above points, and an object of the present invention is to provide a document size detection device that can reliably detect the document size with a simple structure.

[Means to solve the problem]

In order to achieve the above object, the present invention includes a front light source that illuminates the image surface of a document, and a light receiver that receives reflected light from the front light source that is reflected by the image surface of the document and converts it into an electrical signal. In an image reading device equipped with a sensor, a white plate or a white roller, which is located at the document reading position and is arranged on the back side of the document and has a higher reflectance than the brightest part of the image surface of the document, and a white plate of a light receiving sensor. Or a document presence/absence determining means that detects the difference in level between the light output from the white roller and the light reception output from the image surface of the document and determines that there is a document when the level is low, and a document presence/absence signal outputted by the document presence/absence determination means. A dimension determining means is provided for determining the position of the starting edge and the ending edge of the document in the main scanning direction or the stripe width, depending on the width of the stripe.

In addition, the white plate or white roller is made of a translucent material, and a back light source that illuminates the translucent white plate or the translucent white roller from the back, and a front light source that illuminates the image surface of the document and the back light source are alternately used. A switching means for turning on and off may be provided.

[For production]

According to the document size detection device configured in this way, the light reflected by the white plate or white roller on the back side of the document is stronger than the light reflected by the brightest part of the image surface of the document, that is, the white background part, or the light reflected by the white plate on the back side of the document is stronger than the light reflected by the brightest part of the image surface of the document, or the original back light source is detected. is turned on, the front light source is turned off, and after the light receiving sensor detects the leading edge of the document in the sub-scanning direction and the positions of the starting and ending edges in the main scanning direction are determined, the back light source is turned off and the front light source is turned on. As a result, the output of the light-receiving sensor that detects the background is always at a higher level than the output of the light-receiving sensor that detects the image surface, regardless of the shading of the image surface of the document.

Therefore, by appropriately setting the threshold value, the document presence/absence determination means detects the level difference and reliably determines the presence/absence of the document, and determines the starting edge of the document in the main scanning direction according to the document presence/absence signal. Determine the position of the end, that is, the left and right edges,
Also, the document width can be determined from the difference.

Furthermore, when the document is black, the white plate or white roller can be used as it is as a reference white plate for correction of ink, which simplifies the construction of a complicated device.

〔Example〕

Hereinafter, an embodiment in which the present invention is applied to a small-sized copying machine will be described with reference to the drawings.

FIG. 1 is a schematic configuration diagram showing a small-sized copying machine which is a first embodiment of the first invention.

This compact copying machine 1 basically includes a contact image sensor unit 10 which is a double image reading device and an image writing unit 1-2.
It is composed of 0.

The contact image sensor unit 1 10 is a contact type consisting of a white plate 4 located on the back side of the original 2 (and transfer paper 3), a front light source 5 that illuminates the image surface 2a of the original 2, and a CCD with a light receiving sensor. line sensor 6 and freon 1-
An imaging optical system, such as S
It consists of 1.1 and 'A (Selfoc array) 11.

- The image writing unit 20 includes a thermal head 21, an ink sheet 22, a supply roll 23, and a take-up roll 24, and the thermal head 21 is provided with a thermal radar array 21a at its writing position Q. It is being

When copying, place the original 2 and the transfer paper filter with their backs facing each other, that is, with the image surface 2a and transfer surface 3a facing outward, and feed the small copying machine 1 with the original 2 facing up. Insert the paper from the paper slot 12.

For example, when the filler sensor 13 consisting of a microswitch detects the inserted original 2 and copy paper 3, a motor (not shown) is started and operates the paper feed roller pair 14 and the platen roller 15 to feed and transport the paper. At the same time, the front light source 5 is turned on and the image reading/writing standby state is entered.

A line sensor 6 that performs main scanning perpendicular to the conveying direction, which is a sub-scanning direction, detects the leading edge of the original 2 (and the stacked transfer paper 3) fed through the conveyance path 18a by the paper feed roller pair 14. Then, the read image data starts to be stored in a memory which will be described later.

Furthermore, when the leading edge of the copy paper 3 (and original 2) sent through the conveyance path 16b is sandwiched between the thermal head 21 and the platen roller 15 and reaches the man-in position ff1Q, it is inserted into the memory. Image data is output as a video signal to the thermal head 21, and at the same time the ink sheet 22
begins to be sent.

That is, the image data read by the line sensor 6 is stored in the memory for the time it takes for the original 2 and the copy paper 3 to reach the writing position Q from the reading position P, and is stored as delayed image data in the thermal head. 21.

Each element constituting the thermal array 21a of the thermal head 21 is heated in accordance with the video signal corresponding to its position, and the ink in the portion of the ink sheet 1 to 22 that is in contact with the heated element is evaporated. , a copy image is formed on the transfer surface ESa'2 of the transfer paper 3.

The transfer paper 3 on which the copied image has been formed is discharged together with the original 2 from the paper discharge port 17 via the conveyance path 16c by the platen roller 15.

FIG. 2 is a block diagram showing an example of a circuit for performing data processing and control in the first embodiment.

The control unit includes a CPU 30, which is a microcomputer that controls each part of the compact copying machine 1, and a DMA (direct memory access) controller 31, which controls memory input/output of various data. Lesbus.

It consists of a control bus, and is connected to each circuit via a path line (BUS) 32, shown as one line in the figure.

Connected to blocks, elements, etc.

The storage unit includes a ROM 33 in which programs, constant data, etc. are stored in advance, and a RAM 34 in which various variable data, image data, etc. are stored, and each is connected to a pass line.

The pass line also includes an image reading system 40. 1/○ circuit and converter for each of the image writing system 50.

A sensor group 36 provided in each part of the small copying machine via a counter timer and a 1/○ circuit 35. Operation unit 37 for inputting operator instructions and displaying information and alarms, paper feed roller pair 1
4 and a motor 38 which is a power source for the platen roller 15 and the like are connected to each other.

In addition, in order to directly input and output image data to the RAM 34 without going through the CPU 30, the DMA controller 31
Apart from the pass line filter 2, there are directly parallel signal lines for inputting timing signals to the CPtJ 30, and control signal lines for each converter of the image reading system 402 and the image writing system 50, etc.

The image reading system 40 includes a contact image sensor unit 10 (
Figure 1) line sensor 6 and line sensor drive circuit 4
1, A/D&S/P converter 42, and I10 circuit 4
3 and a counter timer 44.

The line sensor 6 consists of a CCD with 1728 dots arranged at a pitch of 8 dots/11 songs, for example, and its reading width is 21.6 nwII, so even if the short side of an A4 size sheet is main scanned at one time, it will still be slightly There is room for.

When the I10 circuit 43 receives a command from the CPU 30 or the DMA controller 1 to the roller 31 via the pass line 32, it transmits the command to the line sensor drive circuit 41 and outputs a start signal to the counter timer 44.

The line sensor driving circuit 41 drives the line sensor 6 to perform main scanning based on the command transmitted from the I10 circuit 43, and also amplifies the image data input from the line sensor 6 and outputs it as a video signal (analog). do.

When the counter timer 44 is started by the start signal from the 170 circuit 43, the counter timer 44 starts by a line (not shown).
The line sensor drive circuit 41 outputs a clock signal to the line sensor 6 from 1 to 17, which is set in advance.
When the count number corresponding to 28 dots is reached, a start signal is output to the line sensor driving circuit 41 and the A/D&S/P converter 42.

The A/D&S/P converter 42 inputs the video signal output from the line sensor driving circuit 41, performs A/D conversion, and then converts it into binary data according to a preset threshold value. The serial 8-bit data for dots is S/P converted and edited into 1 byte of parallel data, which is output to the RAM 34 and stored in memory according to instructions from the DMA controller 1-roller 31.

After outputting the 1728 dots 1 to 2 straightened data to the RAM 3; 4 as 2" 6-byte data for one line, a stop signal from the counter timer 44 causes the line sensor driving circuit 41 and the A/D&S/P converter to What is 42?
Returns to standby state in preparation for the next line scan.

The image writing system 50 includes a thermal head 21, a print timing generation circuit 51, an I)/S converter 52,
10 circuit 53 and a counter timer 54.

As described above, the image data of each line stored in the RAM 34 by the image reading system 40 is stored only in the time it takes for the superimposed original 2 and transfer paper 3 to reach the writing position Q from the reading position P. It is read out with a delay and output to the image writing system 50.

The I10 circuit 53 connects the CPU”S via the pass line 32.
When a signal to start writing one line is input from O, the command is transmitted to the print timing generation circuit 51, and the print timing generation circuit 51 instructs the counter timer 54 to start.

When the 1-line write start signal is output from the CPU 30, the subsequent control is performed by the DMA controller 31.
The data is taken over from the DMA controller 31 to the RAM 34 .
Control signals are output to the print timing generation circuit 51 and the P/S converter 52, respectively.

The 216-byte image data for one line outputted from the RAM 34 is converted into serial data byte by byte by the P/S converter 52 and outputted to the thermal head 21 as a 1728-dot video signal.

The print timing generation circuit 51 sends a latch signal, clock signal, etc. to the thermal head 21 in synchronization with the video signal.
Output to.

The counter timer 54 receives the clock signal from
When the count I~ number corresponding to the preset 1728 dots 1~ is reached, the thermal head 21 and P/S
By outputting a stop signal to the converter 52,
Writing for one line is completed.

In this way, for example, one line can be read.

Writing is performed at a cycle of 5ms each, and 8 lines/
Assuming that the line density is nun, the image on the image surface 2a of the original 2 is copied onto the transfer surface 3a of the transfer paper 3 at a speed of 25 nwn/sec, and is ejected from the paper ejection port 17 (FIG. 1). .

Although not particularly shown in FIG. 2, non-uniformity of the video signal is due to optical causes such as uneven illumination light and electrical causes such as variations in sensitivity of each light receiving element that makes up the line sensor 6. The shading correction for correcting the quality is performed as follows, for example, when the power supply line of the compact copying machine 1 or when the filler sensor 13 detects insertion of the original document 2.

That is, the front light source 5 is turned on and the light is reflected by the reference white plate (which also serves as the white plate 4 in this embodiment)] 5 The light is received by the line sensor 6, and the analog output of each element is A/D converted. , the CPU 30 calculates a correction coefficient proportional to the reciprocal thereof and stores it in the correction coefficient area of the RAM"S4.

By multiplying the output from the next line sensor day by the correction coefficient for each element, image data of a reference white plate or a document with uniform reflectance can be obtained as data with uniform levels.

Therefore, if the circuit for multiplying this correction coefficient is a digital calculation, it may be provided after A/D converting the video signal (analog) output from the line sensor deactivation circuit 41; if it is an analog calculation, for example, the line sensor Opening circuit 4
1 and read out the correction coefficient from the RAM 34 as D.
/A It is sufficient to input the converted data and perform the calculation.

In the embodiment shown in FIG. 2, the line sensor opening circuit 41
The following description will be made assuming that the analog video signal outputted from the .

Figure 3 shows the relative relationship between the reading position P and the document 2. It is an explanatory diagram showing an example.

In the figure, the left and right vertical lines indicated by broken lines are the first and final XMth (XM=172) of the line sensor 6, respectively.
8) is the trajectory of the point read by the element in the sub-scanning direction.

Further, the left and right edges of the document 2 are respectively designated 2r- and 2R, and the hatched portion 2b in contact with the left edge 2L is a dark gray portion.

Three dot-dash lines A, 13. C shows three examples of the relative positions of the reading position P with respect to the document 2.

Figure 4 shows waveforms of one line of video signal (analog or digital may be used) after shading correction, and Figure 5 shows waveforms of one line of binarized video signal with element numbers plotted on the horizontal axis. 3, (A) to (C) correspond to A to C shown in FIG. 3, respectively.

In FIG. 4, the saturation level (maximum output value if digital) is shown by a two-dot chain line, and the threshold value is shown by a broken line.

As shown in FIG. 3, FIG. 4(A) shows the video signal at the relative position A before the document 2 reaches the reading position P, and all the elements read the white plate 4, which is also the reference white plate. The output level is slightly lower than the saturation level.

FIG. 4(B) shows a video signal at a relative position B where the white area of the original 2 is located at the reading position P. It is lower than the level of white board 4.

The threshold value is set at a midpoint between these two levels, and the A/D&S/P converter 4 is used to read the image.
2 (FIG. 2).

Similarly, FIG. 4(C) also shows a video signal at relative position C.

The waveform diagrams shown in FIGS. 5(A) to 5(C) are
The video signals shown in Figures (A) to (C) are binarized using threshold values, and as shown in Figures 5 (B) and (C), the dark gray part 2b or the black solid part is The waveform is the same regardless of the front right side of the document 2, and the coordinates XL, X, and R corresponding to the positions of the left and right edges 2L1 to 2H of the document 2 can be reliably detected.

In FIG. 2, the video signal output from the line sensor opening circuit 41 is input to the H/D&S/P converter 42 as described above, and also passes through the A/D converter 61 constituting the document position detection system 60. The coordinates X L + X R are input to the document position detection circuit 62 and detected.

When the document position detection circuit 62 detects the end of one line scanning based on the stop signal input from the counter timer 44, it outputs a signal to the CPU 30 and stores the coordinate data XL+xR in the RAM 34 via the pass line 32.

The middle = W of the document 2 is obtained by calculating (XRXL) by the CPU 30.

FIG. 6 is a circuit diagram showing a specific example of the configuration of the document position detection system 60.

The document position detection system 60 includes an A/D converter 61 and a document position detection circuit 62. A dip switch (DIPSW) 64 that sets a threshold value, a binary-1 bit counter 65, and a latch circuit 8 that is a dimension determining means that latches the output of the counter 65 depending on the falling and rising edges of the document presence/absence signal, respectively.
8.67.

The counter 65 which scans the line sensor 6 dot by dot by clock signal (CLOCK) and outputs the contents to the latch circuits 66 and 67 is cleared before one line is scanned.

When one line scanning starts, a video signal (VIDEO) is input in synchronization with the clock signal, is A/D converted by the A/D converter 61, and is sent to the digital comparator 6.
Input to the A terminal group of 3.

Each output line of the dip switch 64 for setting the threshold value is pulled up by a pull-up resistor R, and the threshold value is set according to the ON/OFF state of each element switch of the dip switch 64. is applied to the B terminal group of the digital comparator 63.

Digital comparators 6 and 6 compare the data input to A and B, respectively, and output an original presence/absence signal that becomes I-I'' when A>B.

That is, as shown in FIGS. 5A to 5C, the document presence/absence signal becomes H'' when there is no document 2, and becomes L'' when there is.

Since the latch circuit 66 latches the contents of the counter 65 in response to the fall of the document presence/absence signal, it latches the data XL indicating the position of the left edge 2L of the document 2.

Similarly, the latch circuit 67 latches data X and R indicating the position of right m2R since it latches with a rising edge.

When one line of scanning is completed, the counter 65 is cleared by an end signal from the counter timer 44 and stands by for the next scanning.

The data XL and XR latched in the latch circuits 86 and 87 are stored in the RAM'M'34 through the pass line 32 according to a command from the CPU 30, and the IgW of the original is determined by (XR-XL). This is as already stated, and below, determining the positions X L + X R of the left and right edges 2L and 2R of the original 2 and their width W will be described as follows.
"Document size detection".

Specifically, the process of detecting the document size described above is performed as follows.

When the original 2 and the transfer paper 3 are superimposed and inserted through the supply port 12 of the small copying machine 1, and the filler sensor 13 detects this, the paper feed roller pair 14 and the platen roller 15 are moved by the motor 38. The paper is fed and conveyed.

At the same time, the 21 thermal head t' array 21a of the thermal spatula 1 is preheated to a temperature that does not evaporate the ink on the ink sheet 22, and the temperature is detected by the thermistor 21b, which is one element of the sensor group 36, and is controlled to a constant temperature. ing.

Further, ten front light sources 5 are turned on to the contact image sensor unit 1, and the line sensor 6 is repeatedly reading the white plate 4. During this time, the shading correction coefficient can also be determined.

When the original 2 (and the transfer paper 3) reaches the reading position P, the leading edge of the original is read and a signal indicating that the original is present as shown in FIGS. 5(B) and 5(C) is output.

There is a slight delay between when the filler sensor 13 detects the leading edge and when the pair of paper feed rollers 14 starts feeding, so during that time the leading edge of the document 2 is in the nip between the pair of paper feed rollers 14. The edge of the leading edge is regulated so that it is perpendicular to the conveyance direction. However, taking into account some errors, the position of the document is determined after scanning several lines after the document presence signal is output. You can.

When the size of the original 2 (and the copy paper 3) is small, its center does not necessarily coincide with the center of the reading width.

When you stand next to the paper feed slot 12 of the compact copying machine 1 and look down from above, if the reading scan is performed from left to right and the image data is output in that order during writing, the writing scan will be from the right. It must be done to the left.

Therefore, if the center line of the document 2 is closer to the right than the center line of the reading width, the left side margin (XL
-1) is the right side margin CXM-XR) = (1728XR)
However, when writing, the left side margin when viewing the transfer paper 3 from the lower transfer surface 3a is smaller, so (1728-X
If the idle space R) is not provided, a problem will arise in that the written image will be shifted to the right from the center by twice the center setting error on the transfer surface 3a of the transfer paper B.

In this way, information on not only the width of the document 2 but also the positions of the left and right edges is important and needs to be detected accurately.

As described above, the white plate 4 has a higher reflectance than the white part of the original 2.
(The same effect can be achieved even if a white roller is provided for conveying the original M2 and the copy paper 3.) The first embodiment using the same method has been described.

The reflectance of normal paper is not so high even though it looks white, so it is easy to make a white plate 4 with a higher reflectance.

However, if high-quality art paper for color printing or photographic paper with a baryta layer containing barium sulfate is used for the manuscript, it is not easy to obtain a white plate with a higher reflectance than that, and the digital comparator 6 There is a risk that the threshold settings of Party B may become delicate and the detection may become uncertain.

The second invention is to reliably detect the document size even in the case of a document using paper with such high reflectance.

FIG. 7 is a schematic configuration diagram showing a second embodiment according to the second invention, and the same parts as those of the first embodiment shown in FIG.

As shown in FIG. 7, in the second embodiment, instead of the white plate 4, a translucent white plate 7, which is white and translucent when viewed by reflected light, and is made of, for example, a milky white glass or a milky plastic board. , and a back light source 8 that illuminates the translucent white plate 7 from the rear side.Although not shown, only the front light source 5 is turned on when calculating the shading correction coefficient, and only the back light source 8 is turned on when detecting the document size. The device is equipped with switching means to do this.

Note that when reading the original 2, it is usually okay to leave the back light source 8 on, but if the original 2 and the copy paper filter are both made of translucent or transparent material, such as tracing paper or a vinyl sheet, If we consider the case where the back light source 8 is turned off, it is better to turn off the back light source 8.

With this configuration, the time to obtain the shading correction coefficient and the time to read the image are the same as in the first embodiment, but since the document 2 becomes a silhouette when the document size is detected, the video for one line is The signal itself (even without shading correction) becomes a signal with a large level difference close to the binarized signal in the first embodiment (Fig. 5),
The threshold value can be easily set and the document size can be reliably detected without increasing the level of the background portion where there is no document.

To explain in detail the operation of each part during copying work, the JM document 2 (and the overlaid transfer paper 3) is transferred from the paper feed port 12.
When the filler sensor 13 detects the leading edge of the document 2, only the backlight source 8 is first turned on and waits for the leading edge of the document 2 to reach the reading position P.

If you want to calculate the shape ink correction coefficient each time a document is inserted, first turn on only the front light source 5, read the light reflected by the translucent white plate 7, and then turn off the front light source 5 when the shading correction coefficient is obtained. , the back light source 8 is turned on and the document 2 waits for the leading edge of the document 2 to arrive.

When the line sensor 6 detects the leading edge of the document 2, the size of the document is detected after reading several lines, and when the detection is completed, the back light source 8 is turned off and the front light source 5 is turned on to start image reading.

Various methods are known for switching means for turning on and off the front light source 5 and the back light source 8 in accordance with each of the timings described above, so a description thereof will be omitted.

Although the present invention has been described above using a small-sized copying machine as an example, it is not limited thereto, and can be applied to image reading devices such as large-sized digital copying machines, facsimile machines, and image scanners.

7 In addition, the light receiving sensor is a line sensor, an image sensor (2
In addition to using a plurality of light receiving element arrays as in the case of the above example, a single light receiving element equipped with an optical axis deflecting means may be used. Furthermore, the first invention can also be applied to the case where the image plane is scanned with a spot such as a laser beam.

〔Effect of the invention〕

As explained above, according to the present invention, fffj'1
Therefore, it is possible to provide a document size detection device that can reliably detect the document size with a simple structure.

[Brief explanation of drawings]

FIG. 1 is a schematic configuration diagram showing a first embodiment according to the first invention, FIG. 2 is a block diagram showing an example of a circuit for data processing and control, and FIG. 3 is a diagram showing the reading position and original document. Figures 4 and 5 are waveform diagrams showing examples of one line of the video signal and its binarized signal, and Figure 6 is the original. A circuit diagram showing a configuration example of a position detection system. FIG. 7 is a schematic configuration diagram showing a second embodiment according to the second invention. 2... Original 3... Transfer paper 4...
White plate 5... Front light source 6... Line sensor (light receiving sensor) 7... Translucent white plate 8... Back light source 10...
・Contact image sensor unit (image reading device)

Claims (1)

[Scope of Claims] 1. Includes a front light source that illuminates the image surface of the document, and a light receiving sensor that receives reflected light from the front light source that is reflected by the image surface of the document and converts it into an electrical signal. In the image reading device, a white plate or a white roller disposed on the back side of the original at the reading position of the original and having a higher reflectance than the brightest part of the image surface of the original; Document presence/absence determining means for detecting a level difference between the light receiving output by the white plate or the white roller and the light receiving output by the image surface of the document, and determining that there is a document when the level is low; and the document presence/absence determining means. A document size detection device for an image reading device, characterized in that a dimension determining means is provided for determining the positions of a starting edge and a trailing edge of the document in the main scanning direction or the width of the document according to an output document presence/absence signal. 2. The document size detection device according to claim 1, wherein the white plate or the white roller is made of a translucent material, and a back light source that illuminates the translucent white plate or the translucent white roller from behind; the back light source and the document. 1. A document size detection device for an image reading device, comprising: switching means for alternately turning on and off the front light source that illuminates an image surface of the image reading device.