JPH0630267A - Image reader - Google Patents

Abstract

PURPOSE:To improve picture quality by preventing lateral fine lines from being cut off by MTF correction at an image processing part. CONSTITUTION:An image input part 1 is composed of a photoelectric conversion element such as a CCD and a driving part to scan the photoelectric conversion element, analog image information from the image input part 1 is converted into digital information, and an image processing part 2 performs MTF correction, edge detection and white/black binarizing or the like. The read image information is outputted to an output part 3 by a CPU 5, which controls an entire system, and transmitted through a communication control part 4. When a minimum lateral fine line output is larger than a minimum longitudinal fine line output in a minimum output characteristic to line width at the image input part 1 of a device provided with the image input part 1 of a facsimile or the like, a weight coefficient to a picture element in a sub scanning direction is made larger than the weight coefficient of a picture element in a main scanning direction by the MTF correction at the image processing part 2.

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, and more particularly, to an image reading apparatus which prevents breakage of thin horizontal lines and improves image quality. For example, it is applied to a facsimile or a scanner.

[0002]

2. Description of the Related Art An image read by an image input device (scanner) such as a facsimile has deteriorated spatial frequency characteristics. There is a signal correction method using a convolution filter as a means for restoring this deteriorated image, that is, MTF (Modulation Transfer Function) correction. What obtains the blur correction image by the real-time MTF correction circuit by obtaining the numerical value of this filter by the minimum error method is
"Correction of blurred image by real-time MTF correction circuit" (Koichi Ejiri and 2 others Ricoh Technical Report No.6, NOVEMB
ER, 1981).

The "edge image processing method" proposed in Japanese Patent Laid-Open No. 2-226378 is for performing feature analysis of a digitized grayscale image including many edges such as a scene image and a human image. , When calculating an edge image from a grayscale image, a strong (large value) edge is extracted from a portion with a large difference in density value, and conversely, an edge is hardly extracted from a portion with a small difference in density value. In order to do so, an edge detection operator suitable for each case is used, and the edge images obtained by the respective operators are combined.

On the other hand, the output distribution when reading vertical thin lines and horizontal thin lines having the same line width does not change when the document is stopped. However, when the original is moving, the horizontal thin line minimum output is higher than the vertical thin line minimum output in the small character mode as shown in FIG. 4 because the original is moving. Therefore, even with the conventional MTF correction (E ′ = 3 * E- (B + D + F + H) / 2), the output is higher when the horizontal thin line is read than when the vertical thin line is read. In the case of binarization, the correction is weak in the sub-scanning direction, so image quality deterioration such as thin horizontal line breakage occurs.

The output characteristic of the horizontal thin line minimum output characteristic is different in each reading mode. Covering all the reading modes with one conventional MTF correction causes a deviation in the correction, and the line becomes unclear in a certain mode. Image deterioration such as chipping occurred. Furthermore, since the horizontal thin line minimum output characteristics are different in each reading mode, covering all reading modes with one edge detection causes a deviation in processing, and an image such as a line being cut in a certain mode. Deterioration had occurred.

[0006]

The present invention has been made in view of the above circumstances, and in the minimum output characteristic with respect to the line width in the image input unit, when the horizontal fine line minimum output is higher than the vertical fine line minimum output,
In the MTF correction in the image processing unit, the weighting coefficient for the pixel in the sub-scanning direction is made larger than the weighting coefficient for the pixel in the main-scanning direction, so that the thin horizontal line is prevented from being broken and the image quality is improved. The purpose of the present invention is to provide an image reading device.

[0007]

In order to achieve the above object, the present invention provides (1)
It comprises a photoelectric conversion element such as a CCD and an image input section having a drive section for driving the photoelectric conversion element, and an image processing section for converting analog image information from the image input section into digital image information and performing MTF correction. In the image reading apparatus, when the horizontal thin line minimum output is higher than the vertical thin line minimum output in the minimum output characteristic with respect to the line width in the image input unit, the pixel in the sub-scanning direction is weighted by the MTF correction in the image processing unit. The coefficient is made larger than the weighting coefficient of the pixel in the main scanning direction, and further, (2) in the minimum output characteristic with respect to the line width in the image input section, the minimum output of vertical fine lines and the minimum output of horizontal fine lines in each reading mode. According to the above relationship, the weighting coefficient for the pixel in the sub-scanning direction and the weighting coefficient for the pixel in the main-scanning direction are optimized by MTF correction in the image processing unit.
(3) In the minimum output characteristic with respect to the line width in the image input unit, optimization is performed by edge detection in the image processing unit depending on the relationship between the minimum vertical thin line output and the minimum horizontal thin line output in each reading mode. It is a feature. Hereinafter, description will be given based on examples of the present invention.

FIG. 1 is a block diagram for explaining an embodiment of an image reading apparatus according to the present invention. In the figure, 1 is an image input section, 2 is an image processing section, 3 is an output section, and 4 is communication control. Division 5,
Is a CPU (Central Processing Unit), 6 is a RAM (Random Access)
Memory) and 7 are operation units. The image input unit 1 is a CCD
And the like, and a driving unit that scans the photoelectric conversion device. The analog image information from the image input unit 1 is converted into digital information, and MTF correction, edge detection, black and white binarization, etc. are performed. The part to be performed is the image processing unit 2. The read image information is output to the output unit 3 by the CPU 5 controlling the entire system, or transmitted via the communication control unit 4. In addition, it is composed of an operation unit 7 that detects a key input and the like, a RAM 6 that functions as a buffer for image information, and the like.

The invention according to claim 1 will be described below. A CCD or the like is used as a photoelectric conversion element in the image input section. FIG. 4 described above shows the minimum output characteristics of the CCD when a thin line is read. The horizontal axis shows the line width of the thin line, and the vertical axis shows the CCD output. The data can be roughly divided into the minimum output of vertical fine lines (main scanning: 8 lines / m
m) and minimum output of horizontal fine line. The minimum output of horizontal fine line is normal character of each reading mode (1 line feed width in sub-scanning direction = 1 / 3.85mm), small character (secondary: 1 / 7.7mm). , The data of fine characters (sub: 1 / 15.4mm ≒ 65μm) are shown.

In the small character mode, the reading densities in the main scanning direction and the sub scanning direction are almost the same, but as shown in FIG. 4, the fine line reading output characteristic is higher in the output in the sub scanning direction. As can be seen from FIG. 2C, the reading fine line output in the sub-scanning direction is high because the amount of information within the storage time changes due to the movement of the document itself. is there. When the 3 × 3 matrix shown in FIG. 3 is used, the conventional MTF correction E ′ = 3E− (B + D +
With F + H) / 2, the output is higher when reading horizontal fine lines than when reading vertical fine lines, and when binarized, the lines are cut off and the image quality deteriorates. In the embodiment described in claim 1, MTF correction (small character mode) E '= 3E- (1.5
B + 0.5D + 0.5F + 1.5H) / 2 such that the weight for B and H pixels located above and below in the sub-scanning direction with respect to the pixel of interest E is larger than that for D and F pixels. Prevents line breakage when reading horizontal thin lines.

The invention according to claim 2 will be described below. In FIG. 4, the horizontal thin line minimum output characteristic is different in each reading mode.
Covering all the reading modes with one correction causes a fine deviation, and in a certain mode, image deterioration such as a line drop occurs. In the second embodiment, the MTF correction to be used is selected according to each reading mode, and appropriate processing is performed. As an embodiment, in the fine character mode, the vertical fine line minimum output and the horizontal fine line minimum output are substantially equal, so MTF correction (fine character mode) E '= 3E- (0.5B + 1.5D + 1.5F + 0.5H ) / 2 is selected. Also, in the normal character mode, the horizontal fine line minimum output is higher than the vertical fine line minimum output, so MTF correction (normal character mode) E ′ = 3E = (1.8B + 0.2D + 0.2F + 1.8H) / Select 2. The weighting factor for the pixel in the sub-scanning direction is made larger than that in the small character mode.

The invention according to claim 3 will be described below. In FIG. 4, the output characteristics of the horizontal thin line minimum output characteristics are different in each reading mode, so that covering all reading modes with one edge detection causes deviation in processing, and lines are missing in a certain mode. The image deterioration will occur. When (black side) 16 ≥ B, D, E, F, H ≥ 0 (white side), the OR operation in the conventional case; (EB, ED, E)
-F, E-H) E is black when ≧ 4 and E when not ≧ 4
Is white. In the embodiment according to claim 3, as a specific edge detection, as can be seen from the characteristic diagram in the small character mode, the horizontal thin line minimum output is higher than the vertical thin line minimum output, so edge detection (EB ≧ 3, EH ≧ 3, ED ≧ 4, EF ≧ 4
And E is black when at least one of Similarly, in fine character mode, edge detection (when at least one of EB ≧ 4, EH ≧ 4, ED ≧ 4, EF ≧ 4 is satisfied, E is black), and in normal character mode, edge detection (EB ≧ 2) is performed. , EH ≧ 2, ED ≧ 4, and EF ≧ 4 are satisfied, E is black).
By optimally detecting the edge according to each reading mode, it is possible to prevent the image quality from being deteriorated such as a broken thin line. The above is summarized in Table 1.

[0013]

[Table 1]

[0014]

As is apparent from the above description, the present invention has the following effects. (1) Effect to claim 1: In a device including an image input unit such as a facsimile, when the minimum output characteristic with respect to the line width in the image input unit has a horizontal thin line minimum output higher than a vertical thin line minimum output, an image processing unit Since the weighting coefficient for the pixels in the sub-scanning direction is made larger than the weighting coefficient for the pixels in the main-scanning direction by the MTF correction in (1), thin horizontal lines can be prevented from being broken and the image quality is improved. (2) Effect on claim 2: In a device having an image input unit such as a facsimile, in the minimum output characteristic with respect to the line width in the image input unit, the relationship between the vertical fine line minimum output and the horizontal fine line minimum output in each reading mode. Therefore, M in the image processing unit
Since the weighting coefficient for pixels in the sub-scanning direction and the weighting coefficient for pixels in the main scanning direction are optimized by the TF correction, thin horizontal lines can be prevented from being broken and the image quality can be improved. (3) Effect to claim 3: In a device having an image input unit such as a facsimile, in the minimum output characteristic with respect to the line width in the image input unit, the relationship between the vertical fine line minimum output and the horizontal fine line minimum output in each reading mode. By this, since the optimization is performed by the edge detection in the image processing unit, it is possible to prevent the horizontal thin lines from being broken and to improve the image.

[Brief description of drawings]

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

FIG. 2 is a diagram showing a CCD reading movement distance in each reading mode within one storage time according to the present invention.

FIG. 3 is a diagram showing a 3 × 3 matrix used in the present invention.

FIG. 4 is a diagram showing a minimum output with respect to a line width for explaining a conventional image reading apparatus.

[Explanation of symbols]

1 ... Image input unit, 2 ... Image processing unit, 3 ... Output unit, 4 ... Communication control unit, 5 ... CPU (central processing unit), 6 ... RAM
(Random Access Memory), 7 ... Operation part.

Claims (3)

[Claims] 1. An image input section having a photoelectric conversion element such as a CCD and a drive section for driving the photoelectric conversion element, and an image for converting analog image information from the image input section into digital image information and performing MTF correction. In the image reading apparatus including a processing unit, in the minimum output characteristic with respect to the line width in the image input unit, when the horizontal thin line minimum output is higher than the vertical thin line minimum output, the image processing unit performs MTF correction in the sub-scanning direction. The image reading apparatus is characterized in that the weighting coefficient for each pixel is set to be larger than the weighting coefficient for the pixel in the main scanning direction. 2. In the minimum output characteristic with respect to the line width in the image input unit, M in the image processing unit is determined by the relationship between the minimum output of vertical fine lines and the minimum output of horizontal fine lines in each reading mode.
The image reading apparatus according to claim 1, wherein the weighting coefficient for pixels in the sub-scanning direction and the weighting coefficient for pixels in the main-scanning direction are optimized by TF correction. 3. The minimum output characteristic with respect to the line width in the image input unit is optimized by edge detection in the image processing unit according to the relationship between the vertical thin line minimum output and the horizontal thin line minimum output in each reading mode. The image reading apparatus according to claim 1, wherein: