JPH06303416A - Image reader - Google Patents

Abstract

PURPOSE:To generate an image signal with resolution higher than the arrangement density of photoelectric converting elements faithfully to various kinds of image. CONSTITUTION:A luminance classifier 6 decides which of three stages of classification ranges of 'high luminance', 'intermediate luminance', and 'low luminance' the respective outputs of many photoelectric converting elements belong to based on the level. A high-luminance processing block 7, an intermediate-luminance processing block 8, and a low-luminance processing block 9 receive the decision result of the luminance classifier 6 and determine black-and-white patterns of two corresponding pixels as to the respective outputs of the photoelectric converting elements as patterns corresponding to patterns determined for the classification ranges that the output levels of the corresponding photoelectric converting elements belong to and the outputs of adjacent photoelectric converting elements.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image reading apparatus which is applied to a facsimile apparatus, an image scanner apparatus or the like and which generates an image signal corresponding to a document.

[0002]

2. Description of the Related Art Generally, an image reading apparatus of this type forms a reflected light image from a document when the light emitted from a light source is applied to the document on a CCD line sensor, and the CCD line sensor outputs a corresponding electric signal. Is generated.

The CCD line sensor is, as shown in FIG.
A large number of photoelectric conversion elements 52 are arranged in a row on a substrate 51 at equal intervals, and the output of each photoelectric conversion element 52 is serially connected to output an electric signal. The output of this CCD line sensor is an analog signal, but since a binary image is generally handled in a facsimile machine or the like, each photoelectric conversion element 5
The two outputs are binarized into white or black with a predetermined threshold value to generate an image signal.

Now, let us consider a case where the positional relationship between the CCD line sensor and the black image to be read is in the state shown in FIG. 5 (a). In FIG. 5A, 61 is a CCD line sensor, and a hatched figure shows a black image to be read.

In this state, the output signal of the CCD line sensor 61 becomes the signal shown in FIG. 5B, and this signal is binarized by the threshold value shown in FIG.
The image signal shown in (c) is obtained. Here, in FIG.
Focusing on the portion indicated by, since a black image is applied to about half of the corresponding photoelectric conversion element 52, the output level thereof is an intermediate level between the black level and the white level. However, since this level is higher than the threshold, it is regarded as a black level after binarization.

That is, although the edge of the actual black image exists at a position corresponding to the vicinity of the center of the photoelectric conversion element 52 in the portion indicated by A, the photoelectric signal in the portion indicated by A is generated in the generated image signal. An edge of the black image will exist between the conversion element 52 and the next photoelectric conversion element 52.
That is, the position of the edge is displaced, and the image quality is deteriorated.

Such deterioration in image quality can be reduced by increasing the arrangement density of photoelectric conversion elements of the CCD line sensor. That is, as shown in FIG. 6A, if a CCD line sensor 61 in which photoelectric conversion elements 52 are arranged at a higher density than that shown in FIG. The image signal after binarization becomes a signal that is more faithful to the original image, as shown in FIGS. 6B and 6C.

However, the CCD line sensor in which the photoelectric conversion elements are arranged at a high density as described above becomes expensive because the number of required elements increases and the manufacturing becomes difficult.

Therefore, a low-resolution and inexpensive CCD line sensor is used to generate an image signal having a higher resolution than the resolution of the CCD line sensor by image processing. This image processing is similar to the enlargement processing in digital copying, and predicts and interpolates the luminance between photoelectric conversion elements by linear interpolation as shown in FIG. As a result, based on the signal shown in FIG.
Two image signals as shown in FIG. 7B are obtained, and an image signal similar to the image signal obtained by using the CCD line sensor with double the resolution shown in FIG. 7C is obtained.

By the way, such image processing by linear interpolation is effective for improving the image quality of the edge portion of a black image having a certain size, but when reading a striped pattern as shown in FIG. It cannot prevent deterioration of image quality.

Specifically, as shown in FIG. 8A, CC
If the resolution of the D line sensor 61 is higher than the density of the striped pattern, the output signal of the CCD and the image signal after binarization are
As shown in FIGS. 8B and 8C, a level change corresponding to the striped pattern occurs.

However, if the resolution of the CCD line sensor 61 is lower than the density of the stripe pattern as shown in FIG.
The output signal of the CD line sensor 61 does not change in level corresponding to the striped pattern as shown in FIG. 9B. Therefore, it is impossible to obtain the level change corresponding to the striped pattern by linear interpolation, and the image signal after binarization is shown in FIG.
As shown in, the part corresponding to the striped pattern becomes a signal corresponding to the solid black image.

[0013]

As described above, in the conventional image reading apparatus, if the arrangement density of the photoelectric conversion elements is increased in order to realize high resolution reading, the apparatus cost will increase. On the other hand, if the apparent resolution is increased by performing linear interpolation, it is difficult to generate a corresponding image signal when an image having a minute change such as a stripe pattern is read. there were.

The present invention has been made in consideration of such circumstances, and an object thereof is to faithfully reproduce an image signal having a resolution higher than the arrangement density of photoelectric conversion elements to various kinds of images. An object of the present invention is to provide an image reading device that can generate the image.

[0015]

In order to achieve the above object, the present invention is directed to a large number of photoelectric conversion elements arranged at predetermined intervals and the level of each output of the large number of photoelectric conversion elements. Is preset to at least n + 1
A determination means such as a luminance classifier for determining which one of the classification ranges of stages (for example, three stages) and each output of the plurality of photoelectric conversion elements correspond to n.
The black-and-white pattern of the number of pixels (for example, two) is determined according to the classification range determined by the determination means to which the output level of the corresponding photoelectric conversion element belongs and the pattern determined for the output of the adjacent photoelectric conversion element. For example, a determining unit that includes a high-intensity processing block 7, a medium-intensity processing block 8, and a low-intensity processing block 9 is provided for determining a pattern.

[0016]

By taking such a measure, the level of each output of a large number of photoelectric conversion elements arranged at a predetermined interval is preset to at least n +.
It is determined which one of the one-level (eg, three-level) classification range belongs to. Then, for each output of the plurality of photoelectric conversion elements, the classification range determined by the determination means that the black and white pattern of the corresponding n (for example, two) pixels belongs to the output level of the corresponding photoelectric conversion element. And a pattern corresponding to the pattern determined for the output of the adjacent photoelectric conversion element.

[0017]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a diagram showing a main configuration of an image reading apparatus according to this embodiment. In the figure, reference numeral 1 denotes a document table made of a transparent glass plate, on which a document 2 is placed. The original 2 is placed on the original table 1 as needed by a transport system (not shown).
Be transported over.

A light source 3 such as an LED irradiates the original 2 with light from the lower side of the original table 1. The reflected light from the original 2 at this time is imaged on the CCD line sensor 5 by the optical system 4, and is converted into an electric signal by the CCD line sensor 5.

The CCD line sensor 5 is composed of a large number of photoelectric conversion elements arranged at equal intervals on a substrate like the general one shown in FIG. 4, and the density of the photoelectric conversion elements is to be obtained. The density corresponds to half the resolution (nominal resolution) that Specifically, for example, when the nominal resolution is 16 dots / mm, the photoelectric conversion elements are arranged with a density corresponding to 8 dots / mm.

A luminance classifier 6 samples the output signal of the CCD line sensor 5 pixel by pixel and classifies the levels into three levels of "high luminance", "medium luminance" and "low luminance". Then, the luminance classifier 6 gives the activation signal S1 to the high luminance processing block 7 when the level of the output signal of the CCD line sensor 5 is "high luminance". When the level of the output signal of the CCD line sensor 5 is "medium luminance", the activation signal S2 is given to the medium luminance processing block 8. When the level of the output signal of the CCD line sensor 5 is "low brightness", the low brightness processing block 9 sends the start signal S
Give 3

The high-luminance processing block 7, the medium-luminance processing block 8 and the low-luminance processing block 9 have activation signals S1 and S, respectively.
2 and S3 are activated, and the data D _i and D _{i + 1} for two pixels are output. The high-luminance processing block 7, the medium-luminance processing block 8, and the low-luminance processing block 9 are output from the previous pixel data holding block 2
The contents of the data D _i and D _{i + 1} are determined by referring to the reference pixel data D _i-1 and D _i-2 for the pixels.

The previous pixel data holding block 10 holds the data output from the high-luminance processing block 7, the medium-luminance processing block 8 and the low-luminance processing block 9 one timing before, and stores the reference pixel data D _i-1 , Output as D _i-2 .

Next, the operation of the image reading apparatus configured as described above will be described. First, it is assumed that the positional relationship between the CCD line sensor 5 and the black image to be read is as shown in FIG. It should be noted that in FIG. 2A, a hatched figure shows a black image to be read.
In this state, the output signal of the CCD line sensor 5 is as shown in FIG.
The signal shown in (b) is obtained. This signal is input to the luminance classifier 6.

The luminance classifier 6 samples the input signal at the position corresponding to the photoelectric conversion element 11 (the position indicated by x in FIG. 2B), and the levels thereof are classified into high luminance, medium luminance and low luminance. Classify as either. Here, high brightness is a range equal to or higher than the first threshold value shown in FIG. Further, the middle brightness is in a range equal to or higher than the second threshold value shown in FIG. 2B and lower than the first threshold value. The low brightness is a range lower than the second threshold.

Now, in the portion indicated by B in FIG. 2B, C
The level of the output signal of the CD line sensor 5 is equal to or higher than the first threshold value. Therefore, the brightness classifier 6 determines that the brightness is high. Then, the luminance classifier 6 determines the high luminance processing block 7
To the start signal S1.

When the activation signal S1 is given, the high-intensity processing block 7 outputs the data D _i , D _{i + 1} for two pixels according to a preset pattern. Here, in the high brightness processing block 7, a part of the pattern shown in FIG. 3 in which "classification of brightness" is "high brightness" is set. Here, regarding “high brightness”, the two pixels are set to white pixels regardless of the state of the reference pixels. Therefore, the high-intensity processing block 7 uses the reference pixel data D _i-1 ,
The "white pixel" data is output as the data D _i and D _{i + 1} regardless of D _i-2 .

On the other hand, in the portion indicated by C in FIG.
The level of the output signal of the CD line sensor 5 is lower than the second threshold value. Therefore, the brightness classifier 6 determines that the brightness is low. Then, the luminance classifier 6 causes the low luminance processing block 9 to
To the start signal S3.

When the activation signal S3 is given, the low luminance processing block 9 outputs data D _i , D _{i + 1} for two pixels according to a preset pattern. Here, in the low-luminance processing block 9, a portion whose "classification of luminance" is "low luminance" in the pattern shown in FIG. 3 is set. Here, regarding "low brightness", the two pixels are set to black pixels regardless of the state of the reference pixels. Therefore, the low-luminance processing block 9 uses the reference pixel data D _i-1 ,
Regardless of D _i−2 , the data of “black pixel” is output as the data D _i and D _{i + 1} .

In the portion indicated by D and the portion indicated by E in FIG. 2B, the level of the output signal of the CCD line sensor 5 is equal to or higher than the second threshold value and lower than the first threshold value. Therefore, the brightness classifier 6 determines that the brightness is medium. Then, the luminance classifier 6 sends the activation signal S2 to the medium luminance processing block 8.
give.

When the activation signal S2 is given, the medium luminance processing block 9 outputs the data D _i , D _{i + 1} for two pixels according to a preset pattern as follows. First, in the medium luminance processing block 8, a portion of the pattern shown in FIG. 3 in which the "brightness classification" is "medium luminance" is set. Note that, here, two kinds of “medium luminance” are set for two pixels depending on the state of the reference pixel. That is, when the reference pixel is any of “white, white”, “white, black” and “black, white”, “white, black” is displayed, and when the reference pixel is “black, black”, “black, black” is displayed. "White" is set.

The data output by any of the high-intensity processing block 7, the intermediate-intensity processing block 8 and the low-intensity processing block 9 at the immediately preceding timing is fetched and held in the previous pixel data holding block 10. ing. Then, it is output as reference pixel data D _i-1 and D _i-2 from the previous pixel data holding block 10.

Here, in the portion indicated by D, the low luminance processing block 9 has the data D _i ,
Data of "black pixel" is output as D _{i + 1} . Therefore, the reference pixel data D _i-1 and D _i-2 are both "black pixel" data. Medium luminance processing unit 8 where the data of "black pixels" as the data D _i, D _{i + 1}
To output the "white pixel" data.

On the other hand, in the portion indicated by E, the high-intensity processing block 7 has the data D _i , D at the immediately preceding timing.
Either _{i + 1} outputs "white pixel" data, or the intermediate luminance processing block 8 outputs data D _i "white pixel" data and D _{i + 1} outputs "black pixel" data. . Therefore, the reference pixel data D _i-1 and D _i-2 are
Both are either "white pixel" data, or the reference pixel data D _i-1 is "white pixel" data and the reference pixel data D _i-2 is "black pixel" data. Therefore, the medium luminance processing unit 8 outputs the data of “white pixel” as the data D _i and the data of “black pixel” as D _{i + 1} .

That is, in the medium luminance processing block 8, 1
When the two pixels at the immediately preceding timing are both “white pixels”, it is determined that the photoelectric conversion element 11 being processed is located at the edge where the white image changes to the black image, and such an edge is detected. Correspondingly, the front side of the two pixels is a "white pixel" and the rear side is a "black pixel".

On the contrary, when the two pixels at the immediately preceding timing are both "black pixels", it is determined that the photoelectric conversion element 11 being processed is located at the edge where the black image changes to the white image. In order to correspond to such an edge, the front side of the two pixels is “black pixel” and the rear side is “white pixel”.
And

On the other hand, if there are both "white pixels" and "black pixels" in the two pixels at the immediately preceding timing, the photoelectric conversion element 11 being processed has an image with a minute change such as a stripe pattern. It is determined that the two pixels are located at, and the front side of the two pixels is a “white pixel” and the rear side is a “black pixel”.

As described above, the data of the pixel array as shown in FIG. 2C is obtained for the signal shown in FIG. Here, the pixel array shown in FIG.
A comparison with the black image shown in (a) shows that they are very similar.

As described above, according to this embodiment, the output levels of the photoelectric conversion elements 11 are set to "high brightness" and "medium brightness".
It is classified into three stages of "low brightness", and when it is "high brightness", both pixels are "white pixels", and when it is "low brightness", both pixels are "black pixels". Further, when the brightness is "medium brightness", the front side of the two pixels is "white pixel" and the back side is "black" only when both of the two pixels corresponding to the previous photoelectric conversion element are "white pixels". Pixel "
Otherwise, the front side of the two pixels is the “black pixel” and the rear side is the “white pixel”.

In this way, the black / white of two pixels is determined for the output of one photoelectric conversion element 11, so the resolution is twice the resolution corresponding to the density of the photoelectric conversion element 11. That is, in this embodiment, the photoelectric conversion element 11 has 8 dots
Since they are arranged at a density corresponding to / mm, the obtained resolution is 16 dots / mm. Further, even for an image portion having a minute change such as a striped pattern, a signal corresponding to the image can be obtained.

As described above, since the CCD line sensor 6 in which the arrangement density of the photoelectric conversion elements 11 is low can be used to perform high-resolution and faithful reading, the cost of the device can be reduced.

The present invention is not limited to the above embodiment. For example, the relationship between the output of one photoelectric conversion element and the white / black pattern of the two pixels corresponding to this photoelectric conversion element 11 is not limited to that described in the above embodiment.

In the above embodiment, the output level of the photoelectric conversion element 11 is classified into three levels, and white / black of two pixels is determined based on the classification result. White / black of three or more pixels may be determined. In addition, various modifications can be made without departing from the scope of the present invention.

[0043]

According to the present invention, a large number of photoelectric conversion elements arranged at a predetermined interval and at least n + 1 levels (the levels of which are preset for the respective outputs of the large number of photoelectric conversion elements) ( For example, a determination unit such as a luminance classifier that determines which one of the three (3) levels of classification range it belongs to, and the output of each of the plurality of photoelectric conversion elements corresponds to n (for example, two) pixels of black and white. The pattern is determined to be a pattern according to the classification range determined by the determination means to which the output level of the corresponding photoelectric conversion element belongs and the pattern determined for the output of the adjacent photoelectric conversion element, for example, a high brightness processing block 7. Since the determination means including the medium brightness processing block 8 and the low brightness processing block 9 is provided, a resolution higher than the arrangement density of photoelectric conversion elements is provided. Signal, the image reading apparatus can faithfully generated in various types of images.

[Brief description of drawings]

FIG. 1 is a diagram showing a main configuration of an image reading apparatus according to an embodiment of the present invention.

FIG. 2 is a diagram showing a positional relationship between a CCD line sensor 5 and a black image to be read, and a correspondence relationship between an output signal of the CCD line sensor 5 and a generated image signal.

FIG. 3 is a high luminance processing block 7 and a medium luminance processing block 8
9A and 9B are diagrams schematically showing patterns set in a low-luminance processing block 9.

FIG. 4 is a diagram showing a configuration of a CCD line sensor.

FIG. 5 is a positional relationship between a CCD line sensor and a black image to be read in conventional basic image signal generation,
FIG. 3 is a diagram showing a correspondence relationship between an output signal of a CCD line sensor and a generated image signal.

FIG. 6 is a positional relationship between a CCD line sensor and a black image to be read in conventional basic image signal generation,
FIG. 3 is a diagram showing a correspondence relationship between an output signal of a CCD line sensor and a generated image signal.

FIG. 7 is a diagram illustrating generation of a high-resolution image signal using conventional linear interpolation.

FIG. 8 is a positional relationship between a CCD line sensor and a black image to be read when a stripe pattern having a density lower than the resolution of the CCD line sensor is read by a conventional basic reading method; The figure which shows the correspondence of an output signal and the produced | generated image signal.

FIG. 9 is a positional relationship between a CCD line sensor and a black image to be read when reading a stripe pattern having a density higher than the resolution of the CCD line sensor by a conventional reading method using linear interpolation; The figure which shows the correspondence of the output signal of a sensor and the produced | generated image signal.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Original platen 2 ... Original 3 ... Light source 4 ... Optical system 5 ... CCD line sensor 6 ... Luminance classifier 7 ... High-luminance processing block 8 ... Medium-luminance processing block 9 ... Low-luminance processing block 10 ... All pixel data holding block 11 ... Photoelectric conversion element

Claims (1)

[Claims] 1. A large number of photoelectric conversion elements arranged at predetermined intervals, and the output of each of the large number of photoelectric conversion elements belongs to any of at least n + 1-level classification ranges whose levels are preset. A determination unit that determines whether or not the output level of the corresponding photoelectric conversion element belongs to the black-and-white pattern of the corresponding n pixels for each output of the plurality of photoelectric conversion elements. An image reading apparatus comprising: a determining unit that determines a pattern according to a range and a pattern determined for an output of an adjacent photoelectric conversion element.