JPH06164931A - Picture signal processor - Google Patents

Abstract

PURPOSE:To attain proper binarization processing to a picture signal. CONSTITUTION:A picture signal is converted into a digital signal at an A/D converter 1 and the digital signal is binarized by plural binarization circuits 2 whose threshold level differs from each other, Binarization data from the binarization circuits 2 are stored in line buffers 3 to form a matrix output of 3X3 picture elements. Then each pattern of a matrix output and a character line pattern are compared by a pattern detection circuit 5 and binarized data from the binarization circuit 2 whose patterns are coincident are outputted through a selection circuit 6.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image signal processing device which is applied to an image reading device such as a facsimile or a copying machine and which binarizes an image signal.

[0002]

2. Description of the Related Art In the conventional image signal processing apparatus described above, as shown in JP-A-60-74775, the image signal is compared with a predetermined threshold value, and if it is large, it is black or if it is small. If it is white, it is binarized. The threshold value is usually fixed, but there is also a method in which the threshold value is changed according to the shading waveform and binarized.

[0003]

In the binarization method in the above-mentioned conventional image signal processing apparatus, the threshold value is fixed or changed only in accordance with the shading waveform. Due to deterioration of MTF (modulation transfer function), the level difference between black and white becomes small, and the black level is output as an intermediate level.
There was a problem that the characters were faint, broken lines, and crushed.

An object of the present invention is to provide an image signal processing device which can appropriately perform binarization processing on an image signal.

[0005]

In order to achieve the above object, the present invention comprises a plurality of binarizing units for comparing image signals with different threshold values and binarizing the image signals. A pattern detection unit that compares the output binarized data for a plurality of pixels with a predetermined character / line pattern, and selects one of the binarized data from the binarization unit based on the comparison result of the pattern detection unit. And a selecting unit for outputting the output.

Further, the pattern detection unit is characterized in that the pattern comparison is performed by a matrix of 3 × 3 pixels in which a central pixel is a target pixel.

Further, the selected and determined binarized data is used for the reference pixel processed before the target pixel, and a plurality of binarized data is used for the target pixel and the reference pixel processed after the target pixel. It is characterized by using data.

[0008]

According to the above means, the image signal is binarized with different threshold values, and each pattern consisting of a plurality of pixels of binarized data with different threshold values is converted into
For example, by comparing with a pattern appearing in the outline of a character or a thin line, if the pattern matches, it is determined that it is a character or a line portion, and the corresponding binarized data is selected and output, thereby relating to a thin character or a thin line. Even for the image signal, the black level binarized data is surely output, and proper binarization processing is performed.

Further, by comparing the patterns with a 3 × 3 matrix, it is possible to perform efficient pattern comparison.

Further, as the pattern detection data of each binarized data, the binarized data relating to the previous pixel (previous line) which is already selected and decided is used in common to store the binarized data. It is possible to reduce the number of installations, and since at least one data is already selected and determined, the detection accuracy is improved.

[0011]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram showing a configuration of a first embodiment of the present invention, in which 1 is an A / D (analog / digital) conversion for converting an image signal from an image reading section (not shown) into a digital signal. In this embodiment, all white is "0", all black is "15", and the intermediate level between them is "1-14".

Further, 2 is a plurality (three are used in this embodiment).
In the binarization circuit of No. 2, each of the binarization circuits 2 has a different threshold value. In this embodiment, the threshold value is set to "8" or "7".
It is set to "6".

Reference numeral 3 denotes a line buffer, which is a storage unit provided in each of the two binarization circuits 2. Two line buffers 3 and one output line 4 will be described in detail later. 3x
3 matrix data are output.

Reference numeral 5 denotes a pattern detection circuit provided corresponding to each binarization circuit 2. The pattern detection circuit 5 is supplied with the 3 × 3 matrix data of binarized data. Reference numeral 6 is a selection circuit that selects and outputs appropriate binarized data of the binarization circuit 2 based on the comparison result in each pattern detection circuit 5 described later.

Next, the operation of the first embodiment will be described.

The image signal digitized by the A / D converter 1 is binarized by entering three binarizing circuits 2 having different threshold values. The binarized data in each binarization circuit 2 is stored in the line buffer 3, and as shown in FIG. 2, the previous line, the current line, the next line, the next pixel, the current pixel, and the previous pixel. Thus, a matrix of 3 × 3 pixels consisting of and is formed.

In FIG. 2, the central pixel is the pixel of interest (pixel thereof), and each numeral indicates an example of the output level of the binarized data. The threshold value of each binarization circuit 2 is “8” or “7”
Since it is "6", the states of the matrix input to each pattern detection circuit 5 are as shown in FIGS. 3A to 3C at the output level of FIG. In FIGS. 3 (a) to 3 (c), the shaded portions indicate black pixels and the white background portions indicate white pixels.

That is, in the binarization circuit 2 having a threshold value of "8", there is one black pixel (FIG. 3A), and in the binarization circuit 2 having a threshold value of "7", there is one black pixel. In the binarization circuit 2 having three (FIG. 3B) and a threshold value of “6”, there are four black pixels (FIG. 3C).

The pattern detection circuit 5 stores typical patterns appearing in thin characters and character outlines shown in FIGS. 4 (a) to 4 (d). In FIGS. 4 (a) to 4 (d), the shaded portions indicate black pixels and the white background portions indicate white pixels. In the pattern detection circuit 5, the patterns of FIGS. 4 (a) to 4 (d) and FIG.
It is compared and detected whether or not the matrix output, which is the comparison target pattern of (c), matches, and the binarized data of the binarization circuit 2 relating to the matched matrix output is selected from the selection circuit 6 by the comparison result. Is output.

In the present embodiment, since the patterns of FIG. 3 (c) and FIG. 4 (c) match, the binarized data from the binarization circuit 2 having the threshold value “6” is output. , If there is no matching pattern, or if there are multiple matching patterns, the default value will be selected.

In this way, the image signal is binarized with a plurality of threshold values, and a pattern that matches the pattern such as the outline of a character or a thin line is selected from the patterns of the binarized data according to each threshold value. Therefore, the correspondence of the character and line parts to the binarized data is properly determined, and even if the level difference between the black level and the white level in the binarized data such as thin characters and lines is small, the black level Since the binarized data corresponding to is accurately output, it is possible to prevent the occurrence of character blurring, line breakage, crushing, and the like in the output image.

Further, since the pattern detection circuit 5 detects the coincidence between the binarized data and the pattern in a matrix, the pattern detection circuit 5 can detect the pattern efficiently and accurately.

FIG. 5 is a block diagram showing the structure of the second embodiment of the present invention. The basic structure is the same as that of the first embodiment, but in this second embodiment, the line buffer 13 is provided with two lines of the previous line. The binarized data is stored and commonly used for the matrix output of FIG. 2, and the number of line buffers provided corresponding to each binarization circuit 2 is reduced. Furthermore, since the binarized data on the previous line is the data that has already been selected and determined, the detection accuracy can be improved.

The threshold value of the binarization circuit 2 can be appropriately changed to broaden the applicable range. Further, the number of binarization circuits 2 installed is not limited to three in the above-described embodiment. Similarly, the pattern and matrix outputs are 3
The matrix is not limited to × 3.

[0026]

As described above, according to the image signal processing device of the present invention, according to the structure of claim 1, the image signal is binarized by a plurality of different threshold values and different threshold values are set. By comparing each pattern consisting of a plurality of pixels of the binarized data according to the above with a predetermined character / line pattern, it is possible to reliably judge the characters and line parts of the binarized data, that is, the black data, and it is fine. It is possible to reliably output the black level binarized data in the image signal related to the character or the thin line, and it is possible to prevent the occurrence of the character blurring, line breakage, crushing, or the like in the output image.

According to the second aspect of the present invention, efficient pattern comparison can be performed by comparing the patterns with a 3 × 3 matrix.

Further, according to the third aspect of the invention, the binarized data storage unit for forming the pattern is used by using the binarized data relating to the previous pixel (previous line) as one of the pattern detection data. The number of installations such as
Moreover, since the binarized data has already been selected and determined, the detection accuracy is improved.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of an image signal processing device according to a first embodiment of the present invention.

FIG. 2 is an explanatory diagram of matrix output.

FIG. 3 is an explanatory diagram of each matrix output binarized with different threshold values.

FIG. 4 is an explanatory diagram of character / line patterns.

FIG. 5 is a block diagram showing a configuration of an image signal processing device according to a second embodiment of the present invention.

[Explanation of symbols]

1 ... A / D converter, 2 ... Binarization circuit, 3, 13 ... Line buffer, 4 ... Output line, 5 ... Pattern detection circuit,
6 ... Selection circuit.

Claims (3)

[Claims] 1. A plurality of binarization units for comparing image signals with different threshold values to binarize them, and binarized data for a plurality of pixels output from each binarization unit into predetermined characters. An image signal, comprising: a pattern detecting unit for comparing with a line pattern; and a selecting unit for selecting and outputting one of the binarized data from the binarizing unit according to a comparison result of the pattern detecting unit. Processing equipment. 2. The image signal processing apparatus according to claim 1, wherein the pattern detection unit is configured to perform pattern comparison using a matrix of 3 × 3 pixels in which a central pixel is a target pixel. 3. The binarized data that has already been selected and used is used for a reference pixel processed before the target pixel, and a plurality of binary values are used for the target pixel and the reference pixel processed after the target pixel. 3. The image signal processing device according to claim 1, wherein the image data is used.