JPH0221714B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention is a document reading device that uses a plurality of imaging systems to share and read document image information that should be read as a line of information, and is generally used as a reading system for facsimile machines. This invention relates to a general-purpose reading device.

2. Description of the Related Art Conventionally, document reading devices such as facsimile machines have been of the type shown in FIGS. 1 and 2. In FIG. 1, 1 is a document, 2 is a fluorescent lamp for illuminating the document, 3 is a lens, and 4 is a solid-state sensor such as a CCD. A document 1 is illuminated with a fluorescent lamp 2, an image is created on a solid-state sensor 4 using a lens 3, one line is read, and the information on the entire surface of the document is sequentially read by conveying the document in the direction of the arrow. It is something. However, in such a device, since there is an upper limit to the angle of view of the lens 3, it is necessary to maintain a large distance between the document 1 and the sensor 4, and as a result, the optical path must be turned back using a mirror or the like. However, the disadvantage was that the device became larger. In addition, the amount of light from the periphery of the document decreases with respect to the center of the document, making it impossible to obtain uniform output and requiring the provision of various correction means. An example of a configuration considered as a method for solving these problems is shown in FIG. 2. In other words, as shown in FIG. 2, the imaging optical system is composed of a compound eye optical system in which a plurality of lenses are arranged in parallel, and a plurality of solid-state image pickup devices corresponding to each lens are arranged. In this method, one direction is divided into a plurality of regions, and each lens and each solid-state image sensor are assigned to read each region. FIG. 2 will be explained in more detail. In Figure 2, 5 is the manuscript, 6
₁ , 6 ₂ , and 6 ₃ are part of image information such as calligraphy and drawings drawn on the manuscript. 7 ₁ , 7 ₂ , 7 ₃ are lenses, 8 ₁ , 8
₂ and 8 ₃ are images of parts 6 ₁ , 6 ₂ , and 6 ₃ of calligraphy and paintings in the manuscript, respectively. 9 ₁ , 9 ₂ , 9 ₃ are solid sensors,
10 ₁ -1, _{10 1} -2, ..., 10 ₁ -n, 10 ₂
-1, 10 ₂ -2, ..., 10 ₂ -n, 10 ₃ -1,
10 ₃ -2,..., 10 ₃ -n are solid state sensors 9 ₁ ,
9 ₂ and 9 ₃ are shown for each 1-bit light amount conversion unit. The operation of this reading device is as follows. That is, the original 5 is being conveyed in a direction perpendicular to the paper surface,
The direction in which the large number of lenses 7 ₁ , 7 ₂ , 7 ₃ . . . and the solid-state sensors 9 ₁ , 9 ₂ , 9 ₃ , etc. are arranged is the main scanning direction. Lenses 7 ₁ , 7 ₂ , 7 ₃ create reduced images 8 ₁ , 8 ₂ , _{8 3} of calligraphy images 6 ₁ , 6 2 , 6 ₃ on the original on solid-state sensors 9 ₁ , 9 ₂ , 9 _{3 ,} respectively. Image 8 ₁ , 8
Since ₂ and 8 ₃ are inverted images, solid-state sensors 9 ₁ and 9
The output of each bit _{of 2,93 is 101-1,1} from the end.
0 ₁ -2, ..., 10 ₁ -n, 10 ₂ -1, 10 ₂ -
2,...,10 ₂ -n, 10 ₃ -1, _{10 3} -2,
Even _if they are arranged in the order . Therefore, for example, the output of each bit is 10 ₁ -n..., 10 ₁ -2,1
0 ₁ -1, 10 ₂ -n, ..., 10 ₂ -2, 10 ₂ -
It is necessary to extract them in the order of 1, 10 ₃ -n, . . . 10 _{3 -2} , 10 3 -1. Or you need to take the output one by one, store it in memory, and then retrieve it in the correct order. Alternatively, if the head of the recording system is multi-head, each bit of one solid-state sensor and each bit of the recording head corresponding to that solid-state sensor are inverted, so that the read signal of the solid-state sensor can be changed. are sent to the recording system sequentially or in parallel from the end to record, and correct recording is performed.
That is, each bit of the portion of the document 6 ₁ of the recording head is related to each bit of the solid state sensor in the same geometrical relationship as the portion 6 ₁ of the document and its image 8 ₁ . Solid sensor 9 ₂ , 9 ₃
The same holds true for the relationship between the recording head and the corresponding recording head.

Each solid-state sensor must read image information only in a predetermined area of the document, but in order to avoid reading information in other areas at the same time, light shielding means 11 ₁ , 11 ₂ , 11 ₃ , 11 are provided. _Four …
...is provided. For example, the light beam 12 from the part of the image ₆₃ is blocked by the light shielding means ₁₁₂ , so that the photoelectric conversion output of the solid-state sensor ₉₁ is correct when only part of the calligraphy ₆₁ of the manuscript is read. It becomes a signal. Regarding the length of each solid-state sensor, in order to read with high resolution, the length of the photoelectric conversion section array must be equal to the length of each subframe (for example, 6 ₁ ), that is, the lenses 7 ₁ , 7 ₂ , 7 ₃ It is desirable that ... be a same-magnification system, but in the configuration of each solid-state sensor itself, there is a non-conversion area outside the photoelectric conversion part and a light shielding means is provided, so each lens is made to have a same-magnification system. It is not a system but a reduced image. By the way, consider the positional accuracy of an imaging system using a plurality of lens systems as shown in FIG. When the imaging lens (for example, 7 ₁ ) is parallel decentered by △ in the direction perpendicular to the optical axis, and the imaging magnification of the lens is β, then
The amount of movement δ of the field of view in the direction perpendicular to the optical axis on the document surface that has an imaging relationship with a specific fixed sensor has the following relationship.

δ=Δ(1-1/β)...Here, when β>0, erect imaging is formed, and when β<0, inverted imaging is formed.

The fact that the field of view on the document surface that has an imaging relationship with a specific solid-state sensor moves in a direction perpendicular to the optical axis means that the images obtained by the solid-state sensors 9 ₁ , 9 ₂ , 9 ₃ . . . The correspondence relationship between the solid-state sensors 9 ₁ , 9 ₂ , 9 ₃ ...
Jumping and doubling occur at the seams. Therefore, the amount of movement δ of the field of view on the document surface is 1
It is desirable to make it smaller than a certain amount determined by the size of the bit.

For example, if the size on the document surface corresponding to the size of 1 bit of the solid-state sensor is 83.3 μ × 1/β, then the amount of movement δ of the field of view on the document surface is equivalent to 1/4 or less of 1 bit. In order to make the size smaller than 20.8μ×1/β, the imaging magnification β=−0.8 (×0.8 inverted reduction imaging) and δ=△(1+1/0.8)≦26.0, that is, the imaging The amount of parallel eccentricity Δ of the lens (for example, 7 ₁ ) in the direction perpendicular to the optical axis must be 11.55 μ or less. Similarly, if the size on the document surface corresponding to the size of 1 bit of the solid-state sensor is 125 μ × 1/β, then the amount of movement δ of the field of view on the document surface is 1/4 of 1 bit.
In order to make the size on the document surface equivalent to the following, that is, 31.25μ × 1/β or less, as an inverted reduction image of 0.8 times, δ = △ (1 + 1 / 0.8 ≦ 39.06μ, that is, the parallel eccentricity of the lens △ must be 17.4μ or less.
As can be seen from the above, in order to obtain accurate image information with no jumps or overlaps in the original image, it is necessary to The amount of parallel eccentricity △ in the direction perpendicular to must be suppressed to 10μ to 20μ or less. However, the imaging lenses 7 ₁ , 7 ₂ , 7 ₃ ...
If the lens is made of a material that is relatively stable against temperature and humidity, such as glass, each lens must be manufactured separately and then assembled, and the solid-state sensors 9 ₁ , 9 ₂ , 9 ₃ must be assembled. ……
Performing alignment with a high precision of 10 to 20 microns requires a lot of effort since there are a large number of lenses, and this becomes a factor that hinders mass production. Furthermore, if the imaging lenses 7 ₁ , 7 ₂ , 7 ₃ . . . are molded with plastic such as acrylic to integrate a plurality of lenses, the solid state sensors 9 ₁ , 9 ₂ , 9 ₃ . . .
Although the time required for alignment during assembly is greatly reduced, moldable optical plastics such as acrylic have large changes in size or refractive index depending on environmental conditions such as temperature and humidity. Therefore, the parallel eccentricity accuracy as shown earlier is
It is difficult to suppress the thickness to 10μ to 20μ or less. For example, looking at the expansion of acrylic due to heat, the linear thermal expansion coefficient is ~6×10 ^-5 cm/cm/℃, so
The amount of linear expansion of acrylic with a length of 200 mm is 480 μm, assuming that the amount of temperature change is 40 degrees (for example, 20 degrees Celsius ± 20 degrees). On the other hand, the coefficient of linear thermal expansion of silicon (Si) and glass (SiO ₂ ), which are the substrates of solid-state sensors, is about an order of magnitude smaller than that of acrylic, so the amount of linear expansion is smaller under the same conditions as in the case of acrylic. , several tens of micrometers, and when the compound eye lens is integrated with acrylic, the amount of parallel eccentricity of the lens with respect to the sensor is several hundred micrometers at most, making it difficult to put it into practical use.

The present invention provides a compact reading device that satisfies the above-mentioned strict alignment accuracy of the sensor and imaging system, has an excellent signal-to-noise ratio (S/N ratio), and is suitable for mass production. It is something to do.

In the reading device according to the present invention, as an imaging system that divides and forms an image of a document into each predetermined region, a compound eye imaging system is used, which is composed of a plurality of rows of zone plates provided on a common substrate. The above object is achieved by using a zone plate and providing a means for effectively blocking noise light beams different from that of a conventional compound eye optical system generated by using the zone plate. That is, in a reading device that divides the document surface from which the image information is read into a plurality of areas and reads the image information of each divided area with a sensor,
Zone plates are disposed on a common substrate corresponding to each area of the document surface, and the zone plates image information on each of the divided document surfaces onto a sensor. Furthermore, a first light shielding means is provided between the document surface and the zone plate so that a light beam from a region other than the region of the document surface on which each zone plate is to be imaged is not incident on each zone plate; , in order to prevent diffracted light other than the predetermined diffracted light to be detected by the sensor from entering the sensor among the light flux diffracted by the zone plate, a second light shielding means is provided between the zone plate and the sensor. It was established.

Hereinafter, when the present invention is described in detail with reference to the drawings, a Fresnel type zone plate will be used, but it goes without saying that the present invention may also be applied to a Gabor type zone plate.

FIG. 3 is a diagram for explaining a zone plate applied as an imaging system to the reading device of the present invention.
In Figure 3, 13 is the manuscript, 14 ₁ , 14 ₂ , 1
4 ₃ is a part of the calligraphy and painting drawn on the manuscript, 15 ₁ , 15 ₂ ,
15 ₃ is an imaging optical system called a so-called zone plate; 16 ₁ , 16 ₂ , 16 ₃ are images of portions 14 ₁ , 14 ₂ , 14 ₃ of the calligraphy and painting of the manuscript; 17 ₁ , 1
7 ₂ and 17 ₃ are thin film sensors typified by amorphous silicon, and 18 and 19 are both optically transparent media such as glass.

The thin film sensor is provided on a transparent substrate 19, and the zone plate is provided on a transparent substrate 18. FIG. 4a is an enlarged view of the zone plate, showing a state in which a Cr metal film 21 is provided on the transparent substrate 18. Figure 4b shows the fourth
A view of figure a viewed from the optical axis 22 direction is shown. As is well known, the Fresnel (or Soret) zone plate has an imaging effect similar to a lens, and as shown in Figure 4 a and b, it alternates between light-transmitting zones and opaque zones. and rm and rm+1
If the size of the belt whose radius is (m=1, 2, 3...) is given by rm=√×+, then the focal length f
Provides the same imaging effect as a lens with Here, λ is the wavelength of light. The first-order diffracted light diffracted by each band in FIG. 4 produces an imaging effect with a focal length f. Zone plates usually use monochromatic or quasi-monochromatic light in order to reduce so-called chromatic aberration. FIGS. 4a and 4b show an example in which m=10, and the opaque film 21 for shielding light is obtained by depositing a metal film such as Cr. For example, consider the case where the focal length f=10 mm, the wavelength of light λ=546.07 nm (monochromatic light), and the number of bands m=200. At this time, the radius of the minimum circle is r ₁ = 0.07389mm
and the radius of the outermost band is r ₂₀₀ = 1.04499mm
Or r ₁₉₉ = 1.020966mm. That is, the width of the outermost band is 2.4μ from r ₂₀₀ −r ₁₉₉ . This method of forming an opaque film in a band-like shape is possible by photo-etching, which is well known as a step in the IC manufacturing process, and it is possible to form many zone plates at once on the substrate 18 with high precision. be able to. On the other hand, thin film sensors such as amorphous silicon 17 ₁ , 17 ₂ , 17 ₃
... are integrally provided on the transparent substrate 19 in a desired pattern using the same photo-etching method as the zone plate, and the thin film sensors 17 ₁ , 17 ₂ , 17 ₃ ... and the zone plate 15
₁ , 15 ₂ , 15 ₃ ..., for example, thin film sensors 17 ₁ , 1
7 ₂ , 17 ₃ . . . on the transparent substrate 19 and the zone plates 15 ₁ , 15 ₂ , 15
_3. A reference pattern for alignment obtained from the same mask may be provided on the transparent substrate 18 on which . . . is provided by a photoetching method. This allows individual sensors to
There is no need to adjust individual imaging elements, and it is relatively easy to align the integrated imaging system with respect to the integrated sensor with high precision (e.g., less than a few microns). be able to. In addition, the substrates of both the imaging system and the sensor are made of a material with excellent environmental resistance such as glass, and if the substrates of both the imaging system and the sensor are made of the same material, the amount of thermal expansion will be greater than that of the sensor. Since both imaging systems are equal, expansion and contraction due to temperature changes is zero, that is, the positional deviation between the imaging system and the sensor is zero. Even if they are not the same type of glass, if you select a glass with a small coefficient of thermal expansion, it will be possible to achieve the required accuracy (for example, less than 20 μ) in the positional deviation between the imaging system and the sensor due to expansion and contraction due to temperature changes. It can be suppressed.

The example shown in FIGS. 4a and 4b is an example in which opaque and transparent areas are alternately provided for the belt extending from the center. This is called a so-called amplitude type zone plate and is well known. In this case, an example has been shown in which the smallest circle part is opaque, but generally the smallest circle part may be transparent, and opaque and transparent may be alternately repeated outward. In addition, as shown in Figure 5 a and b, although the zone plate transmits light over the entire area, the phase alternately increases by λ/2 (λ: wavelength of light) in the zone from the center. A transparent film 24 may be provided. This is called a so-called phase type zone plate and is well known. In FIG. 5a, the refractive index of the transparent film 24 providing a phase shift is np, and its thickness is dp (np-1)dp=λ/2 (λ: wavelength of light). The opaque film 23 provided on the transparent film 24 is a metal film such as Cr. This has the purpose of blocking stray light coming from the original, since the surface on the side of the zone plate is an ineffective area that does not contribute to image formation.

As described above, if the zone plates shown in FIGS. 4a, b and 5a, b are used as the imaging systems 15 ₁ , 15 ₂ , 15 ₃ . . .
Sensors 17 ₁ , 17 ₂ , 17 ₃ ... and imaging system 15
It is possible to provide a compact document reading device that satisfies the strict positioning accuracy of ₁ , 15 ₂ , 15 ₃ . . . and is suitable for mass production. In FIG. 3, 20 ₁ , 20 ₂ , 20 ₃ , and 20 ₄ are formed by imaging systems other than the imaging system 15 ₂ , for example, adjacent imaging systems 15 ₁ ,
Since there is a possibility that light beams may enter the sensor 16 ₂ from other areas of the document via the document 15 ₃ , this light shielding means has the purpose of eliminating this possibility. However,
In the present application, since a zone plate is used in the imaging system, noise light is generated which is different from that of the optical system shown in FIG. 2, and this situation will be explained using FIG. 6.

In FIG. 6, when the zone plate is used as a compound eye, after the light from the objects 26 ₁ , 26 ₂ , 26 ₃ enters the zone plate 27 ₁ , 27 ₂ , 27 ₃ , the sensor 28 ₁ , 28 ₂ , 28 The diagram shows the diffracted light that participates in image formation on _{the three} surfaces and the diffracted light that becomes the noise component. For zone plates, images 29 ₁ , 2
The diffracted light used to obtain 9 ₂ and 29 ₃ is the +1st order light as shown in FIG.
±3rd order light, etc. become noise components on the image plane,
This results in a decrease in image contrast. For example, the 6th
As shown in the figure, this is undesirable because it enters the adjacent sensor group as stray light. The present invention prevents a decrease in image contrast caused by diffracted light other than the desired +1st-order diffracted light, and examples thereof will be described below.

FIG. 7 shows a reading device showing one embodiment of the present invention. In FIG. 7, 30 is manuscript paper, 31 ₁ , 3
1 ₂ , 31 ₃ are manuscripts to be read, 33 ₁ to 33 ₃
is a zone plate provided on the glass substrate 34, 32 ₁ to 32 ₄ and 35 ₁ to 35 ₄ are light shielding plates,
37 ₁ to 37 ₃ are images of original documents. In the example shown in FIG. 7, sensors 38 ₁ , 38 ₂ ,
means 32 ₁ to 32 ₄ for blocking light entering 38 ₃ ;
As explained in FIG. 6, out of the diffracted lights 41, 42, 43 other than the diffracted light 40 used for image formation among the diffracted lights of the incident light beam 39 by the zone plate, the diffracted lights 41, 42 that enter the sensor are blocked. Means 3
5 ₁ to 35 ₄ are provided, respectively.

As described above, the present invention provides a means for blocking light from entering a sensor from a portion outside a predetermined area of a document through an adjacent zone plate, and a means for blocking light that is diffracted by the zone plate other than the diffracted light that participates in image formation. The means for blocking light (including zero-order straight light) is achieved by separate means, and both means are provided, resulting in a compact and high-contrast document reading device.

FIG. 8 shows another embodiment of the reading device of the present invention. Figure 8 shows zone plates 47 ₁ , 47 ₂ , 47
₃ provided with the substrate 48 and the sensors 54 ₁ , 54 ₂ ,
A cut groove 4 for light shielding is formed on the substrate 51 provided with 54 ₃ .
This is a case where substrates 9 ₁ to 49 ₄ and 52 ₁ to 52 ₄ are provided, and both substrates are bonded by adhesive 50. Depending on the thickness of the substrate, the light-shielding groove provided on the substrate may be provided on only one of the zone plate and the sensor substrate. 46 ₁ to 46 ₄ are
This is a means for blocking light from entering the sensor from portions outside the predetermined areas 56 ₁ , 56 ₂ , 56 ₃ of the original 55 through the adjacent zone plate. The embodiment of the reading device shown in FIG. 9 has the structure of the embodiment shown in FIG. 8, except that the zone plate substrate and the sensor substrate are not bonded to each other. In FIG. 9, the same numbers as those shown in FIG. 8 indicate the same members. In the embodiment shown in FIG. 9, it is desirable that the distance between both substrates be small.

FIG. 10 shows another embodiment of the reading device of the present invention. In FIG. 10, 69, 72, and 75 are a substrate 69 on which zone plates 68 ₁ , 68 ₂ , and 68 ₃ are provided, respectively, and light shielding masks 71 ₁ to 71 _{4 .}
and a substrate 72 on which 74 ₁ to 74 ₄ are provided in the form of a thin film (for example, a metal vapor deposited film), and a sensor 7
7 ₁ to 77 ₃ , and these three substrates are joined to each other by adhesives 70 and 73 and are integrated. 76 ₁ , 76 ₂ , 76 ₃ are predetermined areas 66 ₁ , 76 3 of the document formed on the sensor.
These are the images of 66 ₂ and 66 ₃ . Light shielding plates 67 ₁ to 67 ₄
are predetermined areas 66 ₁ , 66 ₂ , 66 ₃ of the original 65
It is a means for blocking light that enters the sensor from the outside through the adjacent zone plate, and the light blocking masks 71 ₁ to 71 ₄ and 74 ₁ to 74 ₄ block the diffracted light used for imaging out of the diffracted light by the zone plate. This is a means for blocking diffracted light other than light (including light of a straight component called a so-called zero-order light component). Zone plate and light shielding mask 71 ₁ to 71 ₄ and 74
₁ to 74 ₄ and the sensors 77 ₁ to 77 ₄ , when the zone plate, the light-shielding mask, and the sensor are all made of thin films, the positioning of the sensors 77 1 to 77 4 is determined in advance by adjusting the photomask used for patterning each of them. A positioning pattern may be provided, and positioning operations may be performed during fetetching and bonding.

The embodiment of the reading device shown in FIG. 11 is different from the example shown in FIG. 10 in that the substrate on which the zone plate is provided, the substrate on which the light-shielding mask is provided, and the substrate on which the sensor is provided are separated from each other. This is the case.

In FIGS. 10 and 11, the light shielding mask may be provided on the zone plate substrate or the sensor substrate. In addition, the number of light-shielding masks is two in the optical axis direction of the imaging system in the case of FIGS. The arrangement may also be considered to achieve this goal. still,
The alignment error of the light shielding plate in the direction perpendicular to the optical axis direction may be approximately one order of magnitude rougher than the alignment error of 10 to 20 μ between the zone plate and the sensor. In the embodiments described above, the substrates on which the zone plate, sensor, and light-shielding mask are provided are not limited to glass, but any transparent material with a similar coefficient of linear expansion can be used. Further, the sensor is generally not limited to a thin film sensor such as amorphous silicon, but may also be a solid sensor such as a CCD. The sensor array is not limited to one-dimensional sensors, but may be a staggered array or a general two-dimensional array. In that case, the arrangement of the zone plates may be one-dimensional, staggered, or a general two-dimensional arrangement, and the arrangement may be selected depending on the situation.

As described above, the present invention provides a document reading device in which document image information that should be read as one line of information is shared and read by an imaging system composed of a plurality of zone plates. With regard to (for example, facsimile), it is possible to obtain a compact device that obtains high-contrast images that satisfies strict alignment accuracy and S/N ratio of a sensor and an imaging system, and is suitable for mass production.

[Brief explanation of drawings]

Figures 1 and 2 are diagrams for explaining conventional reading devices, Figures 3, 4 a, b, and 5 a, b
are diagrams for explaining the zone plate applied to the reading device of the present invention, FIG. 6 is a diagram for explaining the state of diffracted light by the zone plate, and FIG. 7 is a diagram for explaining the state of diffracted light by the zone plate.
Figures 8, 9, 10, and 11 are, respectively,
1 is a diagram showing an embodiment of a reading device according to the present invention. 30... Manuscript, 32 ₁ , 32 ₂ , 32 ₃ , 32 ₄ ,
35 ₁ , 35 ₂ , 35 ₃ , 35 ₄ ... light shielding plate, 33 ₁ ,
33 ₂ , 33 ₃ ... zone plate, 38 ₁ , 38
_{2,38 3} ...Sensor.

Claims (1)

[Claims] 1. A compound eye imaging system consisting of a plurality of rows of zone plates provided on a common substrate that divides and forms an image of the document surface into predetermined regions, and receives predetermined diffracted light from the compound eye imaging system. a first optical sensor provided between the document surface and the zone plate so that a light beam emitted from a region other than the region of the document surface to be imaged by each zone plate is not incident on each of the zone plates; a second light shielding means provided between the zone plate and the optical sensor so that, of the luminous flux diffracted by the zone plate, a luminous flux other than the predetermined diffracted light to be detected by the optical sensor does not enter the optical sensor; A reading device characterized by comprising a light shielding means.