JPH03165175A - Document detection device and document detection method in image reading device - Google Patents

Abstract

PURPOSE:To detect the position and the width of an original in an installed area without use of an original guide by detecting optically a border between an end of a white color part of the original placed to the installed part of the black color member and the black member. CONSTITUTION:A white color member 22 is provided on a read position A and a black color member 23 is provided so that the width in the sub scanning direction X is a prescribed value and the length is the original read region or over in the main scanning direction. Then when a line l of an original 24 at a position B is read by a read means 20, a video waveform signal is outputted by a CCD 25 and binarizing processing is implemented by a white black binarizing processing section 29 via an amplifier 26, an A/D converter 27 and a picture correction section 28. The binarized data is read on by one bit from a line buffer 30 to detect a band of white/black data to detect a border between a white level of the original 24 and a black level of the black member 23 thereby detecting the position and the width of the original 24. Thus, when the original is placed in the installing position, the position and width of the original is accurately detected without use of the original guide.

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of Industrial Application) The present invention relates to an image reading device such as a facsimile, a copying machine, and a printer.

(Prior Art) Conventionally, a plurality of originals are separated one by one and fed to a reading position, the fed original itself is moved, and image data is read by an optical reading means at the reading position. In image reading devices equipped with an automatic document feed mechanism (ADF), the width of the document is detected by a detection means such as a photo interrupter by placing the document against a specified document guide. .

(Problems to be Solved by the Invention) In the conventional apparatus described above, the detection means can only detect a standard size original, and restrictions are placed on the placement of originals. Furthermore, since there is no means for detecting the amount of skew in the document, there is a problem in that if a large skew occurs, the document may jam or image data may not be read properly.

SUMMARY OF THE INVENTION An object of the present invention is to provide an image reading device that can accurately detect the position and width of a document if the document is at the installation position.

(Means for Solving Problem 11) In order to achieve the above object, the present invention separates a plurality of originals and feeds them to a reading position, and moves the fed original itself to the reading position. In an image reading device that reads image data using a reading means, a black member is provided at the position where the original is placed, and the boundary between this black member and the white edge of the original is optically detected to determine the position and width of the original. It is characterized in that it is configured to detect.

(Function) Since the above method is adopted, by optically detecting the boundary between the white edge of the document located in the installation area of the black member and the black member, the document can be moved in the installation area without using a document guide. The position and width of the document can be detected.

(Example) Embodiments of the present invention will be described below based on the drawings.

FIG. 2 is a perspective view showing the appearance of the image reading device.
2 is an automatic document feeder (ADF) placed on the top of the main body 1; 3 is a pressure plate provided on the ADF 2 for pressing a document; and 4 is a paper discharge tray.

FIG. 3 is a block diagram schematically showing the structure of the image reading apparatus of the present invention, in which the ADF 2 includes a semicircular roller 6 provided on one side of the document table 5; A separating roller 7a and a feed roller 7b separate and feed the stacked originals transported one by one, a pair of transport rollers 8 and 9 transport the separated originals, a guide roller 1O1 a guide roller 10, and a pair of transport rollers. a presser plate 12 that is provided between the contact glass 11 and the contact glass 11 and presses the original against the contact glass 11 without any gaps; It is composed of a discharge roller 13 and the like.

In addition, the main body 1 of the image reading device includes a light source 142 and a reflection mirror 1.
5.16.17, lens 18, CCD (charge coupled device)
A reading means 20 consisting of a reading section 19 consisting of the following is provided.

In the figure, the ADF 2 is the contact glass 1 of the main body 1.
1 is in a closed state. The documents stacked on the document table 5 of the ADF 2 are conveyed by a semicircular roller 06 to a separating core a7, separated and fed one by one by a separating roller 7a and a feed roller 7b, and transferred to a pair of conveying rollers 8. The contact glass 11 and the holding plate 12 are moved by the guide rollers 10 and
sent between. At this reading position of the presser plate 12, the reading means 20 performs a known image data reading operation on the original. After this reading operation, the document is moved by the transport roller pair 9 and the ejection roller 13 to J[
One sheet is ejected to tray 4.

FIG. 1 is a configuration diagram showing an embodiment of the presser plate 12 shown in FIG.
is upstream of the white member 22 in the conveying direction of the original 24,
The white member 22 is provided at the reading position A, and the black member 23 has a protrusion of 10 to 20 mm in the sub-scanning direction X at the above position. and the length is longer than the document reading area in the main scanning direction.

In the same figure, the light source 14 and reflecting mirrors 15, 16, and 17 constituting the reading means 20 are moved to the positions indicated by two-dot chain lines, so that the reading operation of the reading means 20 is performed at position B. Next, when the original 24 is conveyed to the position B by the guide rollers 10 or the like, the black member 23 is positioned above the installation position of the original 24. In this state, data corresponding to the effective reading width in the main scanning direction at position B is read by the reading means 20 from below the original 24, and the original 24, which will be described later, is read.
Detection of the position and width of is performed. After detecting the position and width of this document 24 and setting the reading area in the main scanning direction,
Light source 14 and reflective mirror 15.16.1 of reading means 20
7 to the position indicated by the solid line, and at the reading position [A, a known image data reading operation for the document is started.

FIG. 4 is a block diagram showing a control system for detecting the position and width of a document, in which 25 is a C0D (charge coupled device) that constitutes the reading section 19 in FIGS. 1 and 3, 26 is an amplifier, and 27 is a control system for detecting the position and width of a document.
is an A/D (analog/digital) converter, 28 is an image correction section, 29 is a monochrome binarization processing section, 30 is a crawl-in buffer, 31 is a CPU (central processing section), and 32 is a reading means 20
A drive circuit for braking the drive motor 33, 34 is the ADF2
36 is an image rotation calculation element, and 37 is a page memory.

Here, at position B in FIG. 1, the original 24 as shown in FIG.
When the line Q of is read by the reading means 20, as shown in FIG.
A video waveform signal as shown in a) is transmitted to the CCD 2 in FIG.
5, and an amplifier 26. After passing through the A/D converter 272 and the image correction section 28, the image is binarized in the black-and-white binarization processing section 29. The binarized data is as shown in FIG. 6(b) (here, white is 1 and black is 0). Then, by reading the binary data one bit at a time from the line buffer 30 and detecting the band of white/black data, and detecting the boundary between the white of the original 24 and the black of the black member 23, the position of the original 24 is determined. Detect width. At this time, several bits of white dust or black dust may be deposited on the black member 23 or the white original 2.
4. Therefore, it is necessary to prevent false detection due to dust.

Next, the detection operation for detecting the position and width of the original 24 will be explained with reference to the flowchart of FIG.

In the figure, the size of the garbage is up to 5 bits,
Also, if a continuous 20-bit white or black is detected, the original 2
4 or black member 23 is detected, first, a bit counter counts the total number of bits read from the line buffer 30, a white counter counts the number of white bits, and a black counter counts the number of black bits. , reset the white dust counter that counts the number of bits corresponding to the size of white dust and the black dust counter that counts the number of bits corresponding to the size of black dust (
0) Do (SL-1). Each of the above counters is c in Figure 4.
It is built into p-u3t.

Next, 1 bit of data is read from the line buffer 30 (SL-3) and the bit counter is incremented by 1 (
SL-2). Therefore, to detect a white belt, first check whether it is white or black (1 or 0), and if it is white (YES in SL-4), the white counter is set to +1 (5l-5), and the black dust counter is After counting the number of black dust, reset the black dust counter (Sl-6), and if the white counter count > 20 (YES at SL-7), (
The CPU 3 calculates the count number of the bit counter - the count number of the white counter - the count number of the black dust counter.
1 (SL-
8). This value becomes the edge-cutting start bit of the document 24.

On the other hand, if it is black (No in SL-4), the black dust counter is set to +1 (5L-9), and if the count number of the black dust counter is >5 (YES in SL-10), the previously detected white determines that the original 24 is not white and resets the white counter (0) (S1-11). If the bit counter is at the last bit (S1-12), an error is determined. As described above, erroneous detection due to white/black dust is eliminated and an accurate document reading start position is detected.

Next, the end of the document is detected. This is to detect a black belt, which is the opposite of detecting the reading start position described above. First, reset the white counter, black counter, white dust counter, and black dust counter. 5L-13) Read 1 bit from the line buffer 30 (S
1-14b) and increments the bit counter by 1 (S
1-14a). Here, check whether it is white or black, and if it is black (YES in SL-15).

After counting the black counter +IL (S1-16) and the number of white dust, the white dust counter is reset (0) (SL-17). And the count number of the black counter>20
If (YES in Sl-18), calculate (count number of bit counter - count number of black counter - count number of white counter) 5l-19), end bit.
Store in register. On the other hand, if it is white (NO at SL-15), the white dust counter is set to +1 (5l-20), and if the counter number of the white dust counter is >5 (YES at SL-21), the black determines that the black member 23 is not black and resets the black counter (S1-22);
Here, if the bit counter is the final bit, (SL~
23: YES), the final bit is stored in the end bit register (SL-24), and the process ends.

By the way, if this continues, if a black band portion is detected as in the document 24 shown in FIG. 5, the point A in the figure will also be detected as the final end of the document 24.

Therefore, following step (SL-19), the operations from step (SL-1) to step (SL-12) described above are repeated, and the white belt is detected again. If the cycle of performing the black band detection operation is repeated up to the final bit, the final edge of the document 24 can be detected without fail. Further, when the original 24 has up to the final bit, black belt detection cannot be performed, but in that case, if the final bit is taken as the final end of the original 24, there is no problem.

Here, as a condition for the document 24 to be detectable, there must be a white portion of at least about 2 to 3 nn at both ends (main scanning direction) of the document 24.

In the manner described above, the position and width of the document 24 placed at an arbitrary position can be accurately detected in bit units.

Next, a method for detecting the skew of the original 24 will be explained.

First, in FIG. 1, the reading position of the reading means 20 is moved to the B position, and the original 24 is moved slightly from the B position by 0.5~
About 1.0 strokes, reading level r! Convey in the direction of iA. In this state, the reading start position and end position of the original 24 are detected by the operation described in the flowchart of FIG.

Here, when the skew amount of the original 24 is zero and ideal conveyance is performed, when the reading means 20 detects the line Q portion on the original 24 at the position of the original 24 indicated by the broken line in FIG. 8(a). , the reading start position A and the end position B are detected, and then the document 24 is moved again in the direction of the B position by 0.5 to 1
．． When the reading means 20 detects the start position [A' and the end position B' again, the first position is A4A', B'': B''.
'.

However, if the document 24 is conveyed in a skewed manner in either direction, as shown in FIGS. 8(b) and (C),
Before and after the movement of the document 24, the reading start positions C and C' and the end position and D' are detected to be shifted. In the case of FIG. 8(b), C>C' and D>D'. Here, if c-c'=α and D'-D=β, if α>β, the manuscript 24 is tilted diagonally to the right as shown in FIG. 8(c), and if α<β. For example, it is determined that the document 24 is tilted diagonally to the left as shown in FIG. 8(b).

Therefore, if α is β, tan(1””/α) is the 8th
The skew angle θ in FIG. 8(b) is determined, and if α>β, the skew angle θ' in FIG. 8(c) is determined by jan (1 cape 1)/β). The larger the angles θ and θ', the larger the amount of skew.

Therefore, if the angles θ and θ' are larger than the specified values, it is determined as an error that the conveyance of the original 24 is stopped to prevent reading errors, jamming of the original, or damage to the original.

Also, the image rotation calculation element 36 and page memory 37 in FIG.
Using this method, skew correction can be performed by performing image rotation processing on the image data of the corresponding page based on the skew angles θ and θ' and rotating one image.

By the way, in conventional image reading devices, data on a white tape serving as a reference is read using a CCD or the like, and the data is stored in a memory element (RAM) as shading data.
) and obtain correction data from the data. However, when reading the shading data by COD, for example, if the white tape is
If there is black dust with a size of 2 to 3 pixels, the waveform of white data will be such that a portion corresponding to the position of the black dust appears at a low output as shown in FIG. 9(a). When the waveform of this white data is corrected and a uniform density chart is read, the waveform becomes as shown in FIG. 9(b), and white spots occur in areas where black dust existed.

Therefore, in this embodiment, a shading correction control system shown in FIG. 10 is provided in the block diagram of the control system shown in FIG. 4. In FIG. 10, 25 is a COD constituting the reading section 19 of FIG. 1, 29 is a binarization processing section, 31 is a C'PU, 40
is an S/H (sample/hold) circuit, 41 is an amplifier, 42 is an A/D converter, 43 is a path selector, 44 is a comparison calculation circuit, 45 is a RAM for shooting, 46
is a shooding correction circuit.

The white member 22 in FIG. 1 is read by the reading means 20, and photoelectrically converted by the CCD 25 based on the amount of light.

The obtained data is input to an A/D converter 42 via an S/H circuit 40 and an amplifier 41. At this time, the minus reference level and plus reference level of the A/D converter 42 are the dummy black level output from the CCD 25 and the maximum white level of the white member 22, respectively, and between the two levels there are n bits (for example, 8 bits). A/D conversion is performed to a digital value.

This digital data is stored in RAM2S. The above operation is performed for all pixels of the CCD 25. However, if this continues, white spots will occur as mentioned above, so
In this embodiment, as shown in the flowchart of FIG. 11, the minimum black data is stored in the RAM 45 in advance (S
2-1), select the path selector 43 to shading mode S2-2), input the data of the white member 22 from the C0D 25 to the A/D converter 42 as described above (
S2-3). Then, in order to read the data already stored from the RAM 2S (S2-4), the comparison operation circuit 44 compares the output (white data) of the A/D converter 42 with the RAM 2S.
The data corresponding to that bit of are compared in size, and if white data>RAM (YES in S2-5), the current white data is stored in the RAM 45 (S2-6). In this way CC
All pixels of D25 are compared in size as described above, and the larger one (white) is stored in the RAM 45 (S2-
7).

Next, move the reading means 20 of FIG. 1 by one line (S2).
-9), perform the same process and spread this out to several dozen lines (spread)
The maximum value of the white level during that period is stored in the RAM 45 (S2-8). Therefore, the shading data is updated and stored over several I times.

This means that during the above number I, normal shading data can be obtained as long as several fields of black dust are not on the same bit.

(Effects of the Invention) According to the present invention, by optically detecting the boundary between the black member and the white end of the document, if there is a document at the installation position, the document can be positioned without using a document guide. The width and width of the document can be detected.

An image reading device that can accurately detect width can be provided.

[Brief explanation of the drawing]

FIG. 1 is a configuration diagram showing one embodiment of the holding plate portion of the image reading device of the present invention, FIG. 2 is a perspective view showing the external appearance of the image reading device, and FIG. 3 is a configuration diagram of the image reading device of the present invention. Figure 4 is a block diagram showing the control system for detecting the position and width of the original, Figure 5 is an explanatory diagram showing the reading state of the original, and Figure 6 (a) shows the video waveform. Waveform diagram, Figure 6 (
b) is an explanatory diagram showing the binarized data, Fig. 7 is a flowchart of the detection operation for detecting the position and width of the document, and Fig. 8 (a) is an explanatory diagram showing the binary data.
). (b) and (C) are explanatory diagrams explaining the installation state of the manuscript;
Figure 9(a) is a waveform diagram of white data, Figure 9(b) is a waveform diagram of white data.
FIG. 10 is a block diagram showing a shading correction control system, and FIG. 11 is a flowchart of shading correction operation. 1... Main body of the image reading device, 2... ADF, 5
...Document mounting table, 7a...Separation roller, 7b
... Feed roller, 11... Contact glass, 12... Holding plate, 14... Light source, 15.16.
17...Reflection mirror 18...Lens, 19...
Reading section, 20... Reading means, 22... White member, 23... Black member, 24... Document, 25... C
CD, 29...Black and white binarization processing unit, 31...CPU
, 36... Image rotation calculation element, 37... Page memory, A, B... Reading position.

Claims (3)

[Claims] (1) Separate multiple sheets of original and feed them to the reading position,
In an image reading device that moves a fed document itself and reads image data by a reading means at the reading position,
An image reading device characterized in that a black member is provided at the position where the original is placed, and the boundary between the black member and the white edge of the original is optically detected to detect the position and width of the original. Device. (2) The image reading apparatus according to claim 1, wherein the image reading apparatus is configured to detect the skew amount of the document based on detection data of the position and width of the document. (3) The image reading apparatus according to claim (2), wherein the image reading apparatus is configured to correct the skew amount by performing image rotation processing on the image data based on the detected data of the skew amount of the document. Device.