JP2569667B2 - Image reading device - Google Patents

Description

The present invention relates to an image reading apparatus that reads an image using a linear CCD image sensor.

(Prior Art) In an image reading apparatus that reads an image of a document, a CCD linear image sensor that is a photoelectric conversion element is used. The photosensitive area of the photosensitive portion of one pixel of the current CCD linear image sensor is, for example, 5 μm ×
It is as small as 7 μm (see FIG. 10. In this figure, the hatched portion indicates the photosensitive portion of each pixel. The size of one pixel is 7 μm both vertically and horizontally). In order to obtain an image with high resolution, the photosensitive portion must be small.
However, when the size of the photosensitive portion is reduced, the photosensitive area is correspondingly reduced, resulting in insufficient light quantity.

For reading an image, it is necessary to illuminate the original. For the illumination, a light source such as a halogen lamp or a fluorescent lamp is used.

The reflected light of the original due to the illumination is detected by the CCD linear image sensor and converted into an electric signal. Then, a binary signal digitized by A / D conversion is obtained.

(Problems to be Solved by the Invention) As described above, the CCD linear image sensor is
Only reflected light having a width of about 7 μm is received. However, when using a halogen lamp, the original can be narrowed to a width of only about 2 to 3 mm at most due to the positional accuracy of the filament of the halogen lamp and the manufacturing accuracy of the reflecting mirror. difficult. On the other hand, even when a fluorescent lamp is used, since illumination is performed by light emitted from the fluorescent substance on the tube wall, condensing illumination cannot be performed.

Therefore, the lighting efficiency of both halogen lamps and fluorescent lamps is very poor, and it has been necessary to strongly illuminate with a large wattage lamp.

An object of the present invention is to provide an image reading apparatus that can efficiently use illumination light.

(Means for Solving the Problems) An image reading apparatus according to the present invention illuminates an original, projects an image of the original on a CCD linear image sensor, and mainly arranges CCD pixels of the CCD linear image sensor. In an image reading apparatus that performs mechanical scanning of a document in a sub-scanning direction perpendicular to the scanning direction to read an image, a CCD linear image sensor means including CCD pixels having a photosensitive portion longer in the sub-scanning direction than in the main scanning direction; A / D conversion means for digitizing an electric signal detected by each CCD pixel of the linear image sensor means, image storage means for storing image data of one page of a document digitized by the A / D conversion means, One line of image data in the sub-scanning direction is read out from the image storage means and subjected to Fourier transform. Image quality compensating means for compensating for image quality degradation of image data by multiplying the Fourier transform value by the reciprocal of the wave number characteristic or its approximate function and performing Fourier inverse transform of the obtained product to obtain image read data. It is characterized by the following.

(Operation) In general, as shown in FIG.
It is assumed that an input of (y) is input and an output of g (y) is output. Further, as an example, as shown in FIG. 2 (b), it is assumed that a pulse having no width (pointwise brightness in terms of image) is input and an output h (y) appears. This h (y)
Indicates the characteristics of the system.

The frequency characteristics of f (y), g (y) and h (y) are f
Fourier transform of (y), g (y), h (y)
If (w), G (w), and H (w) are used, they are expressed as follows.

Using this frequency characteristic, the output G (w) becomes the input F
(W) and the characteristic H (w) of the system.

G (w) = F (w) · H (w) (2) Inverse Fourier transform for transforming based on frequency characteristics is also established at the same time.

The general description of f, g, h above is, for example,
It is described on page 211 of Hiroshi Kubota, "Applied Optics" (Iwanami Zensho, 1970).

FIG. 3 shows an example of the shape of a pixel formed by the CCD linear image sensor 22. The shaded area is the photosensitive area 24, 24,
... is shown. The length b of the photosensitive portion 24 in the y direction is made larger than the pitch a in the x direction (for example, b = 14 μm, a = 7 μ)
m, b / a = 2). The width c (= 5 μm) of the photosensitive portion 24 is shorter than the pitch a to separate the pixels.

By increasing the shape of the photosensitive section 24 in the sub-scanning direction,
The amount of received light increases.

However, the contrast and the resolving power in the sub-scanning direction decrease, and the quality of the image deteriorates. Therefore, the following arithmetic processing is performed.

First, the response characteristics of the CCD linear image sensor 22 will be described. The characteristic of the light intensity I of the photosensitive portion 24 of the CCD pixel constituting this sensor is that which should be read as a point,
Due to the effect of mechanical movement of the optical system and the shape of the photosensitive section 24, the shape shown in FIG. 4 is obtained. That is, if a coordinate axis is set at the center for analysis, I (y) = 0 y ≦ −b / 2 (4) = I ₀ b / 2 ≧ y ≧ −b / 2 = 0 y ≧ b / 2 where , B is the length of the photosensitive section 24 in the y direction. CCD
More precisely, one of the pixels is mechanically scanned even during light reception. However, the time during which one pixel operates and scans is about 0.1 microsecond, and the mechanical scanning speed as sub-scanning is usually about 100 mm / sec. Only a degree is scanned. Since this length is sufficiently smaller than the element size of 14 μm, it is ignored here.

In the above-described general theory of the response of the system, I (y) corresponds to h (y). Therefore, a function H (w) representing the frequency characteristic of the system is obtained by Fourier transform of I (y). That is, H (w) = I ₀ sinπw / πw (5) where w = b · N.

Where b is the actual length in mm and N is the spatial frequency, expressed in units (books / mm). FIG. 5 shows a frequency characteristic of the function H (w). As a matter of course, the contrast at a high frequency, that is, at a fine location is reduced.

The function H (w) will be specifically described by applying the actual spatial frequency N. FIG. 6 shows the frequency characteristics of H (N) when the length b of the photosensitive section in the sub-scanning direction is 14 μm and 7 μm. H (w) becomes 0 when w = bN = 1.
The value of N at this time is 71 lines / mm when b = 14 μm, and 143 lines / mm when b = 7 μm. By the way, a fine type image reading apparatus usually has a reading density of 400 dots per inch. Expressing this as a spatial frequency N,
8 lines / mm. CCD linear image sensor with 50 pixels
If the number is 00 and the pitch is 7 μm, the photosensitive size of the CCD linear image sensor is 35 mm. When A3 size 297mm is reduced and projected on this CCD linear image sensor, The spatial frequency of 8 lines / mm is 68 lines / mm. When the reciprocal (1 / H (w)) becomes infinite, an appropriate maximum value may be used.

Since the size of the curve A is twice as large as that of the curve B, the value of cos (0.022
N) It is twice. Therefore, the reciprocal 1 / cos (0.
022N) is multiplied by a curve A to obtain a curve B.
Despite having increased the size of the photosensitive section, the characteristics have returned to the original.

In the present invention, a function H representing the frequency characteristic of this system
Using (w), the output value g (y) of the optical characteristic and the input value f
(Y) is obtained. As a result, the frequency characteristics lowered by increasing the size of the photosensitive portion of the CCD are restored. At this time, the following equation is used.

G (w) × (1 / H (w)) = F (w) · H (w) · (1 / H
(W)) = F (w) (6) That is, G is obtained by Fourier transform of the output value g (y).
(W) is obtained (Equation (1b)), then multiplied by the reciprocal of H (w) to obtain F (w) (Equation (6), and then inverse Fourier transform of F (w) is performed ((3a) ), The original image input f (y) is obtained.

 Specifically, the processing procedure shown in FIG. 7 is used.

First, the whole image of the document is sequentially scanned by the optical system (step S1), and the two-dimensional information of the image for one page is stored in the image memory (step S2).

Next, x is specified (step S3), read data g (y) on one line in the sub-scanning direction is read from the image memory (step S4), and G (w) is obtained by Fourier transform (step S5). .

Next, the reciprocal (1 / H) of the frequency characteristic of the photosensitive portion in the sub-scanning direction
(W)) is multiplied by G (w) to obtain F (w) (step S6).

Next, image data f is obtained by inverse Fourier transform of F (w).
(Y) is obtained and stored in the image memory (step S7).

Next, when it is determined that the value is not the final value of x (step S
8) Returning to step S3, the same process is performed on the line in the sub-scanning direction by sequentially changing x.

When processing of all x is completed, f (y) is obtained as an image reading signal.
Is output from the image memory (step S9). After that, the image reading signal is subjected to a process such as binarization and output to an external printer or the like.

Even if the shape of the CCD pixel is not a rectangle but a parallelogram in order to avoid a moiré pattern, arithmetic processing can be performed similarly if the frequency of the photosensitive portion determined by the shape of the CCD pixel is obtained.

By performing the process of multiplying by the reciprocal, it is possible to further improve the characteristics of the signal read by the original CCD having the small photosensitive area.

As a characteristic improvement, it is needless to say that an operation for approximately increasing the value of the high frequency is effective even if the inverse number (1 / H (w)) is not multiplied accurately.

Hereinafter, an embodiment of the present invention will be described with reference to the accompanying drawings.

FIG. 8 shows a schematic arrangement of a mirror scanning type optical system which is an example of the image reading apparatus. The light emitted from the halogen lamp 2 illuminates the original 10 placed on the original glass 8 via the reflecting mirrors 4 and 6. The original image is sequentially reflected by mirrors 12, 14, and 16, passes through a lens 18, passes through a filter 20, and passes through a CCD.
An image is formed on the linear image sensor 22. The halogen lamp 2, the reflecting mirrors 4, 6 and the mirror 12 move together to
0 is scanned, and at the same time, the mirrors 14 and 16 also move at half the speed, and the optical path length to the lens 9 is made constant. The CCD linear image sensor 22 has the shape shown in FIG.

FIG. 9 shows a method of reading the entire image by rewriting the configuration of FIG. 8 as a model. The sub-scanning direction is
6, the direction in which 12 to 16 move to scan the original 10; CC
The D linear image sensor 22 outputs a signal of each pixel in the direction perpendicular to the sub-scanning direction (referred to as a main scanning direction).
The entire image of the document 10 is read by both the main scanning and the sub-scanning. Hereinafter, the main scanning direction is defined as the x direction, and the sub scanning direction is defined as the y direction.

The method of mechanically performing sub-scanning includes a method of moving the original while illuminating the original in the main scanning direction as in FIG. There is a method of moving the linear image sensor.

FIG. 1 is a block diagram of a circuit for detecting the density of a document.

The clock generation circuit 40 includes the CCD linear image sensor 22
, And is also connected to a CPU 42 for controlling the reading of the original, and is used as a clock signal. The CCD linear image sensor 22 converts an optical signal into an electric signal. The A / D converter 44 converts an analog output of the CCD linear image sensor 22 into a digital signal. The shading circuit 46 detects the unevenness in the light amount in the main scanning direction and the CCD linear image sensor 2 for the digital signal.
This is for correcting variations between two bits (pixels), and the timing is given by a shading signal from the CPU. The output of the shading circuit 46 is
It is input to the image memory 48. The image memory 48 is a RAM that stores all read data of the document 10.

When the read data of one page is stored in the image memory 48, first, the Fourier transform unit 50 reads the read data of one line in the sub-scanning direction (y direction) (in the response of the general system, g ( (corresponding to y)) is Fourier-transformed into G (w) by equation (1b). Next, the multiplier 52 multiplies G (w) by the reciprocal 1 / H (w) of the frequency characteristic of the response of the system according to the equation (6) to obtain F (w). This 1 / H (w) is given by the shape b of the photosensitive portion 12 of the CCD pixel of the CCD linear image sensor 22 and the scanning frequency N in the sub-scanning direction (see equation (5)). A function may be used. Next, upon receiving F (w) for all the frequencies, the inverse Fourier transformer 54 performs an inverse Fourier transform according to equation (3a),
Send to 56. The above arithmetic processing is repeated at each position y in the sub-scanning direction. Thus, the image read data of one page subjected to the arithmetic processing is stored in the image memory 56. In the above description, two image memories 48 and 56 are used for the sake of convenience, but in practice, arithmetic processing can be performed using only one image memory. The CPU 42 determines a binary or dither attribute for each predetermined area based on the information lacking in the image memory 56, and writes the attribute to the attribute RAM 58.

Next, binarization processing is performed in the same manner as in the related art, and output to the outside. That is, the correction calculation read data is sequentially sent from the image memory 56 to the comparison circuit 60. Pattern generation circuit 62
Is to generate a threshold when dither is selected.
The thresholds are generated in an (mxn) matrix. The selector 64 switches the threshold value between binary and dither based on the information in the attribute RAM 58 at the time of data transfer, and outputs it to the comparison circuit 60. The comparison circuit 60 compares the corrected read data with the threshold value sent from the selector 64, and outputs the result to the output circuit 66 as one bit. The output circuit 66 outputs 1 when an effective image signal (synchronous signal) is sent from the CPU 42.
The bit image signal and the synchronization signal are output to an external device (such as a printer).

Note that the CPU 42 inputs a motor signal, a ramp signal, a fixed position signal, a command signal, and the like, and controls the operation of the image reading apparatus.

(Effect of the Invention) By increasing the size of the photosensitive portion of the CCD linear image sensor in the sub-scanning direction, illumination light can be used efficiently. As a result, a lamp with a small wattage can be used, and the cost can be reduced.

On the other hand, the decrease in image quality due to the enlarged photosensitive area
Since the compensation can be performed by the arithmetic processing, the image contrast and the resolving power do not decrease.

[Brief description of the drawings]

FIG. 1 is a block diagram of a circuit system of the image reading apparatus. FIGS. 2 (a) and 2 (b) are diagrams each showing the response of the system. FIG. 3 is a view of a photosensitive portion of the CCD linear image sensor. FIG. 4 is a graph of the response of a CCD pixel. FIG. 5 and FIG. 6 are graphs of the frequency characteristics of the response of the system, respectively. FIG. 7 is a diagram showing a flow of the arithmetic processing. FIG. 8 is a diagram of an optical system of the image reading apparatus. FIG. 9 is a diagram showing scanning of the entire image. FIG. 10 is a diagram of a photosensitive section of an example of a conventional CCD linear image sensor. 22: CCD linear image sensor, 42: CPU, 48: Image memory, 50: Fourier transformer, 52: Multiplier, 54: Fourier inverse transformer, 56: Image memory.

Claims (1)

(57) [Claims] An original is illuminated, an image of the original is projected on a CCD linear image sensor, and the original is mechanically scanned in a sub-scanning direction perpendicular to a main scanning direction in which CCD pixels of the CCD linear image sensor are arranged. An image reading device that reads an image by performing a CCD linear image sensor comprising CCD pixels having a photosensitive portion longer in the sub-scanning direction than in the main scanning direction, and an electric signal detected by each CCD pixel of the CCD linear image sensor A / D conversion means for digitizing an image, image storage means for storing image data for one page of a document digitized by the A / D conversion means, and image data for one line in the sub-scanning direction from the image storage means Is read out and subjected to Fourier transform. An image reading apparatus, comprising: image quality compensation calculation means for compensating for image quality deterioration of image data by multiplying the transformed value and performing an inverse Fourier transform of the obtained product to obtain image read data.