JPH05161003A - Original reader - Google Patents

Abstract

PURPOSE:To attain precise picture reproduction by detecting a border between an original and an original platen when reading the original so as to detect an askew positioning of an original thereby correcting picture information in response to the result. CONSTITUTION:An original read section reads an original 1 to obtain picture information, a comparator detects a border between the original 1 and an original platen 10 to detect a precise brightness distribution thereby detecting the askew original and correcting the picture information according to the result.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a document reading device for photographing a document from above and for detecting and correcting the degree of bending of a document that is spatially curved.

[0002]

2. Description of the Related Art Conventionally, in this type of device, for example, as disclosed in Japanese Patent Laid-Open No. 3-117965,
There is a method in which a line beam light is irradiated onto the surface of the original at an appropriate angle, the degree of bending of the original is detected from the projected image, and the image information is corrected based on the detection result of the bending.

[0003]

However, in the case of the conventional apparatus that detects the degree of bending of the original from the projected image by light irradiation, the boundary between the original and the original table cannot be detected. Cannot detect the brightness of the part without letters and figures. Therefore, the correct luminance distribution is not read, and correct image information may not be corrected in some cases. The present invention solves the above problem, and detects the degree of bending of a document by detecting the boundary between the document and the document table at the time of reading the document without using a distance sensor. An object of the present invention is to provide a document reading device that corrects information and enables correct image reproduction.

[0004]

In order to achieve the above-mentioned object, the present invention provides a document table on which a document is placed, and a document table for reading a document placed on the document table. Original reading means arranged at intervals, boundary detecting means for detecting the boundary between the original and the platen based on the output of the original reading means, and a predetermined distance from the boundary within the original based on the boundary detecting means. The detection means detects the output of the document reading means at a certain point, and the correction means that corrects the output of the document reading means based on the output of this detection means.

[0005]

According to the above structure, the original reading unit reads the original, obtains the image information, and the boundary detecting unit detects the boundary between the original and the original table to detect the degree of bending of the original. The image data read by the reading means can be corrected according to the detection result.

[0006]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 shows an external configuration of a document reading apparatus according to an embodiment of the present invention. A document 1 is placed facing upward on a document table of a desk-shaped apparatus body, and a reading unit 2 is provided at a position where the document 1 can be read from above. The reading unit 2 is supported by the supporting member 3, and a predetermined space is provided between the document 1 and the reading unit 2.
This distance is such that at least the readable area on the surface of the original 1 is always visible, and this space forms a working space between the surface of the original 1 and the reading unit 2. Further, the device is provided with operation switches 4, a bend correction circuit unit 5, and a bend warning buzzer 20.

2 and 3 show the configurations of the document table 10 on which the document 1 is placed and the reading unit 2. The image information of the document 1 is transferred to the image sensor 6 through the imaging lens 7 of the reading unit 2.
Read in. The image sensor 6 is a semiconductor photoelectric conversion element such as a CCD and may be an area sensor or a line sensor. 2 shows an area sensor, and FIG. 3 shows a line sensor.

FIG. 4 shows a series of copying operation processes using the document reading apparatus according to this embodiment. Hereinafter, the method and process for correcting the bend will be described with reference to FIG. Here, an example using a line sensor as the image sensor is shown. The user sets the document to be copied at a predetermined position within the imageable area of the image sensor (#
1). When the original is set, the copy button is turned on (# 2). When the copy button is turned on, the document on the platen is photographed (# 3). The platen is made of a color other than white, such as black, so that the shape of the boundary between the document 1 and the platen 10 (hereinafter referred to as the edge part) can be seen as shown in FIG.
Shoot the original slightly larger than the original size. In this shooting, edge detection and document height calculation are also performed, as will be described later in detail. Next, after performing line bending correction (# 4), brightness unevenness correction (# 5), blurred image correction (# 6), and image expansion (# 7) for all pixels, the photographed image is printed out. (# 8), the copy operation is completed.

Each of these processes will be described below. First, the edge detection process will be described. The image forming position on the CCD is different between the edge portion of the original when the original is bent like a bound original and the edge portion of the original when the original is flat. This is shown in FIGS. 6 to 10.
c, Δc1 and Δc2 indicate the differences in the image forming positions. In FIG. 9, c is the distance from the center position of the CCD to the image forming position of the edge portion of the document, H is the distance from the document table 10 to the lens 7, and e1 is the image forming position of the edge portion when the document is flat. Reference numeral e2 denotes an image forming position on the edge portion of the document, and E denotes the edge portion of the document. If the values of c, Δc, and H are known, the height h of the document can be obtained by the following equation. h = (Δc / c) · H (1)

As shown in FIG. 10, the value of c can be obtained by calculating the number of pixels included in the pixel range of the edge portion of the document from the pixel at the center position of the CCD. The value of Δc can be obtained by calculating the number of pixels included in the range of pixels of the edge portion of the document from the pixels of the edge portion of the document where the document height h = 0. The value of H is obtained from the variation amount of the zoom of the lens. As described above, it is possible to obtain the same height data of the document as the number of pixels of one line imaged from the CCD.

Further, instead of detecting only one edge portion of the document, both edge portions of the document are detected as shown in FIG. 7, and both heights h ₁ and h ₂ are represented by h ₁ = ( Δc1 / c) H h ₂ = (Δc2 / c) H, and h may be calculated based on the values of h ₁ and h ₂ by using an average or a weighted average. Further, as shown in FIG. 8, if the value of | h ₁ −h ₂ | becomes larger than a certain value, correct bend correction cannot be performed, and therefore the bend alarm buzzer 20 of FIG. 1 sounds or the LED lights up. This prompts the user to replace the original.

Next, the line curl correction (# 4) based on the height data of the original will be described with reference to FIGS. 11 to 13. The line bending occurs because the image forming position on the CCD is deviated between when the document is bent as shown in FIG. 11 and when the document is flat. FIG. 12 is an example of curving, and (a) is an image taken when the original is bent,
(B) shows imaging when the document is flat. From FIG. 11, the shift amount Δb of the image forming position is obtained by the following equation. Δb = (h / H) · b (2) Based on the result of Δb in the above equation, the image information a in the i-th row and the j-th column
By performing the coordinate conversion of (i, j) by the following equation, the curving is corrected. FIG. 13 is a diagram for explaining coordinate conversion. a (i + Δb, j) = a (i, j) (where i ≦ I / 2) a (i−Δb, j) = a (i, j) (where i ≧ I / 2) (3)

Next, the image expansion rate α (j), which is the extent to which the image of the pixel in the j-th column is expanded, is calculated. column j,
(J + 1) of the document in pixels th column height data respectively h _j, When h _{j + 1,} the image expansion rate than FIG 14 alpha (j) is determined by the following equation.

[Equation 1]

Next, the brightness unevenness correction (# 5) will be described with reference to FIG.
It will be explained based on item 5. As shown in FIG. 15A, the edge portion E1 of the original document and the line data of one line of the original document of several pixels (for example, i-th pixel) from the edge portion E1 are read. E
Reference numeral 2 indicates a read line of the brightness distribution. Since this line is a portion where there is no character in a normal document, it can be used as luminance unevenness distribution data. FIG. 16 shows a luminance distribution curve of one line of a document several pixels from the edge portion. Assuming that the row data is P (1), ... P (j) from the first column in order, the image data obtained by correcting the unevenness of brightness of the image information data a (i, j) is obtained by the following equation.

[Equation 2]

Further, instead of reading the luminance distribution of only one edge portion of the document as described above, the luminance distribution of both edge portions of the document is read as shown in FIG. 15B, and the average of the two luminances is read. Alternatively, the luminance distribution of the document may be obtained using a weighted average.

Next, Laplacian processing and gamma correction (# 6) will be described as blur image correction. In the vicinity of the fold line of the document, the document is largely tilted and the height difference of the document is large, so that the out-of-focus is large. Laplacian processing and γ correction are performed to remove this out-of-focus blur.

Next, the image expansion processing (# 7) based on the image expansion rate α (j) will be described. The number k to be expanded for the pixel in the j-th column is calculated by the following equation. k = α (j) · j−j−n (6) n: number of expanded pixels However, k is a value obtained by rounding off the right side. For example, 10
The image data of the column is α (j) = 1.4 (j = 1, ... 1
When it is extended in 0), it becomes as shown in FIG. When j = 1 k = 1.4 × 1-1-0 = 0.4 =
0 From k = 0, when j = 1, the number of expanded pixels is 0 (n = 0) When j = 2, k = 1.4 × 2-2-0 = 0.8 =
1 From k = 1, when j = 2, the number of expanded pixels is 1 (n = 1) When j = 3, k = 1.4 × 3-3-1 = 0.2 =
When j = 3 from 0 k = 0, the number of expanded pixels is 0 (n = 1). Similarly, when j = 4, 5, ... 10 is calculated, it becomes as shown in FIG.

When the number of expanded pixels is obtained in this way, density reproduction interpolation is then performed as expansion processing. FIG. 19 shows the relationship between the density and the expanded pixel when the pixel on the j-th column is expanded to k pixels. The density reproduction interpolation is shown in the following equation. When k ≧ 1, a (i, j + 1) = a (i, j) + (l / k) · {a (i, j + k) −a (i, j)} (l = 0,1, ... k) (7) When k = 0 a (i, j + 1) = a (i, j) (8) By the above-described processes, the bending correction can be performed.

Next, an example of the processor for correcting the bend will be described with reference to the block diagrams of FIGS. However,
A line sensor is used as the CCD. CCD 21
Output from the CCD 21 is sampled and held, and the output from the CCD 21 and the sampled and held are differential amplifier 23.
The dark output of the CCD 21 is removed by differential amplification. The output from the differential amplifier 23 is the A / D converter 2
A / D conversion is performed according to 4. ROM using counter 27
The data of the shading correction coefficient of the CCD 21 in the main scanning direction is sent from 26 to the multiplication circuit 25, and the A / D converter 24
The shading correction of the CCD 21 in the main scanning direction can be performed by performing the multiplication with the output of. The output of the multiplication circuit 25 is latched by the latch circuit 28. The ROM 29 is a ROM that converts the next data latched by the latch circuit 28 into a negative value. Latch circuit 28 and ROM 29
Output is sent to the adder circuit 32 to be added.

By inputting the output of the adding circuit 32 to the comparator 33, the level difference between the black and white edges is calculated. That is, the edge portion of FIG. 5 is detected here. When the level difference between the black and white edges exceeds a certain value, the latch timing is controlled by the comparator 33 so that the value of the counter 38 is latched by the latch circuit 37. Counter 38
The latched value of becomes the value of Δc. By sending the output of the latch circuit 37 and the H / C data to the multiplication circuit 40 and performing multiplication, the height data h of the document can be calculated. The output h ₁ of the multiplication circuit 40 is the output h ₁ of the upper edge portion of the document shown in FIG.
_1, the value of h ₁ is latched by the latch circuit 73.
Furthermore, the output h ₂ of the multiplier circuit 40 is h ₂ of the lower edge portion of the document of FIG. When h ₁ and h ₂ are put in the comparator 75, and when the difference between h ₁ and h ₂ exceeds a certain value, the buzzer 76 sounds or the LED lights up for a warning. h ₁
And the values of h ₂ are sent to the adder circuit 74 to be added. The output of the adding circuit 74 is sent to the multiplying circuit 77 and multiplied by 1/2 to obtain the average of h ₁ and h ₂ , and the result h is the height of the document (terminal A).

The output of the multiplication circuit 25 enters the shift register 30. The shift register 30 is for delaying data. The output of the shift register 30 is R
Converted to the inverse by OM34. Since the output of the ROM 34 is the original data of the shedding correction coefficient in the sub-scanning direction, the multiplication circuit 36 outputs the correct white level after the correction.
Sent to and performs a multiplication by some value. The output of the multiplying circuit 36 is controlled by the comparator 33 so as to be latched at the end of the document, and is latched by the latch circuit 39. The output becomes the shading correction coefficient of the CCD in the sub-scanning direction. The output of the multiplication circuit 25 is the line buffer 3
1, the output of the line buffer 31 and the latch circuit 39 are entered.
The output of is sent to the multiplication circuit 35 and is multiplied,
Shading correction in the sub-scanning direction of the CCD can be performed (terminal B).

The height data of the document at the terminal A is latched by the latch circuit 4.
The data is latched by 1 and the next data is converted into a negative value by the ROM 42. The outputs of the latch circuit 41 and the ROM 42 are sent to the adder circuit 43 to perform addition. This is equivalent to the value of h _{j + 1} -h _j of equation (4). The output of the adder circuit 43 is sent to the multiplier circuit 44 performs the square of the multiplication of h _{j + 1} -h _j, square value of output and line spacing t of the multiplier circuit 44 is sent to the addition circuit 45 adds. The output of the adder circuit 45 is R
The image expansion rate α (j) in the equation (4) can be calculated by converting the square root by the OM 50 and converting it into an integer by the integer conversion circuit 51 (terminal C).

The original height data of the latch 41 is sent to the multiplication circuit 46 and multiplied by the value of the counter 49, that is, the value of b, and the output of the multiplication circuit 46 is multiplied by the multiplication circuit 47.
To (h / H) · b by multiplying by 1 / H
The value of −Δb can be calculated. This Δb and the value b of the counter 49 are sent to the adder circuit 48 to add (b + Δ
The value of b) can be calculated. The value of (b + Δb) is converted into an integer by the integer conversion circuit 52 and latched by the latch circuit 53. The output of the latch circuit 53 and the next data of (b + Δb) are sent to the comparator 54 for comparison. By controlling the write clock of the line buffer 55 according to the output of the comparator 54, the shading-corrected output from the CCD from the terminal B is subjected to the coordinate conversion processing of the formula (3) in the line buffer 55. Enter. By the counter 56,
It controls the read clock of the line buffer 55 and the selector 57. The selector 57 selects white data to be input to the data missing by the line curving correction process of the formula (3). The output of the selector 57 enters the line buffer 58 (terminal D).

Two lines of the output of the terminal D are put into the line buffers 59 and 60. Using the data for three lines output from the terminal D, the 3 × 3 edge enhancement filter 61 performs edge enhancement, and the ROM 62 performs γ conversion (terminal E).

The output α (j) of the terminal C is input to the microcomputer 63. The microcomputer 63 controls the selector 67. According to the output α (j) of the terminal E after the γ conversion, the microcomputer 63 converts the data of one line into the line buffers 64, 65, 6
Put in 6, ... The ROM 69 converts the output of the line buffer selected by the selector 67 into a negative value. The line buffer containing the previous data is selected by the selector 67, and the selected line buffer and the output of the ROM 69 are sent to the adder circuit 70 and added. That is, {a (i, j + k) -a (i,
j)} is calculated. The value of 1 / k in the equation (7) is calculated by the microcomputer 63, and the data and the output of the addition circuit 70 are sent to the multiplication circuit 71 and are multiplied. The output of the multiplication circuit 71 and the output of the shift register 68 are the addition circuit 7
By being sent to 2 and added, the processing of the equation (7) is performed, and the output after the bending correction processing is obtained.

[0026]

As described above, according to the present invention, by detecting the boundary between the original and the original table, the degree of bending of the original is detected, and the original data is corrected based on the result of the detection of the original. High luminance unevenness correction can be performed, and as a result, the compressed and distorted image is expanded and corrected by arithmetic processing, and can be reproduced more beautifully and more easily viewed.

[Brief description of drawings]

FIG. 1 is an external configuration diagram of a document reading apparatus according to an embodiment of the present invention.

FIG. 2 is a perspective view showing a configuration of a reading unit.

FIG. 3 is a perspective view showing a configuration of a reading unit.

FIG. 4 is a flowchart of an image copy operation.

FIG. 5 is a diagram showing a shooting size of a document.

FIG. 6 is a diagram showing a difference in image forming position on a CCD of one edge portion of a document when the document is bent and when the document is flat.

FIG. 7 is a diagram showing a difference in image forming position on a CCD of both edge portions of a document when the document is bent and when the document is flat.

FIG. 8 is a diagram showing a difference in image forming position on a CCD of both edge portions of a document when the document is bent and when the document is flat.

FIG. 9 is an explanatory diagram of a formula representing the height h of a document.

FIG. 10 is a diagram showing a relationship between c and Δc in imaging a document.

FIG. 11 is an explanatory diagram of an expression representing a deviation amount Δb of an image formation position.

12A and 12B are diagrams showing a result of capturing an image of a document, in which FIG. 12A is when the document is bent, and FIG. 12B is when the document is flat.

FIG. 13 is an explanatory diagram of coordinate conversion.

FIG. 14 is an explanatory diagram of an expression representing an image expansion rate α (j).

FIG. 15 is an explanatory diagram of luminance unevenness correction.

FIG. 16 is a diagram showing a luminance distribution.

FIG. 17 is an explanatory diagram of image expansion processing.

FIG. 18 is an explanatory diagram of image expansion processing.

FIG. 19 is an explanatory diagram of density reproduction interpolation.

FIG. 20 is a block diagram showing a bend correction processor.

FIG. 21 is a block diagram showing a bend correction processor.

FIG. 22 is a block diagram showing a bend correction processor.

FIG. 23 is a block diagram showing a bend correction processor.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Original 2 Reading section 5 Bending correction circuit section 6 Image sensor 7 Imaging lens 10 Original plate

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Nobuyuki Okis, 2-13-3 Azuchi-cho, Chuo-ku, Osaka, Osaka International Building Minolta Camera Co., Ltd. (72) Satoshi Nakamura, Azuchi-cho, Chuo-ku, Osaka 3-13-3 Osaka International Building Minolta Camera Co., Ltd. (72) Inventor Toshihiko Karazaki 2-33-1 Azuchi-cho, Chuo-ku, Osaka City Osaka International Building Minolta Camera Co., Ltd.

Claims (1)

[Claims] 1. A manuscript table on which a manuscript is placed, manuscript reading means arranged at a predetermined distance from the manuscript stand for reading the manuscript placed on the manuscript table, and the manuscript reading device. Boundary detection means for detecting the boundary between the original and the platen based on the output of the means; and detection means for detecting the output of the original reading means at a point within the original at a predetermined distance from the boundary based on the boundary detection means. A document reading apparatus comprising: a correction unit that corrects the output of the document reading unit based on the output of the detection unit.