JPH04288774A - Read processing method for image data - Google Patents

Abstract

PURPOSE:To transmit images with improved picture quality by removing noise signals, which are mixed into image data when reading image information, on the transmission side. CONSTITUTION:When reading black-and-white binarized image data from a transmitting original for each line, the same line is repeatedly read twice, and the respective correspondent picture elements of the both read image data are successively compared each other as aimed picture elements. When the both compared aimed picture elements are in black-and-white inverting relation each other as the result of this comparison, the respective aimed picture elements in the black-and-white inverting relation are further compared with picture elements positioning before and behind the image data to which the respective aimed picture elements belong. Then, when the aimed picture elements are in the black-and-white inverting relation to the picture elements positioning before and behind the image data as the result of this comparison, among these aimed picture elements, the pixel data having the same black-and-white relation with the preceding and following picture elements are transmitted.

Description

[Detailed description of the invention]

0001

[Field of Industrial Application] The present invention relates to a method for reading and processing image data in a facsimile machine, and more specifically,
The present invention relates to a device characterized in that speckles caused by noise mixed into image data are removed at the stage of reading a transmission document.

0002

[Prior Art] Recently, facsimile machines that transmit and receive image information via telephone lines have been developed and used. Image data is obtained by scanning with an image pickup device such as a CCD and converting it into white pixels or black pixels, and the obtained image data is encoded and transmitted to the called party via a telephone line.

By the way, as a method for reading the image information of the transmitted original document, as shown in FIG.
is transported in the sub-scanning direction (in the direction of the arrow) by transport rollers R, while irradiating the transmission document P with a line light source 100, and the reflected light is reflected by a mirror M and condensed by a lens L to be displayed on a CCD.
In addition to the solid-state image sensor 101 such as
The electrical signal obtained is processed by a binarization circuit (not shown).
The pixel data is output as "H" level black pixel data or "L" level white pixel data with a predetermined threshold as the boundary.

However, when a facsimile machine is reading a document to be sent, if a large noise signal leaks from other electrical equipment to the AC power supply side, or if external noise arrives from outside the facsimile machine, the noise signal may be added to the original image data. For example, as shown in Fig. 7(a), noise may be mixed in at the beginning of a continuous black run part and the original white pixel may be inverted to a black pixel, or as shown in Fig. 7(b). As shown in (c), noise is mixed in at the end of a continuous black run part and the original black pixel is reversed to a white pixel, and as shown in (c), noise is mixed in at the end of a continuous white run part and the original black pixel is reversed. White pixels are reversed to black pixels,
Alternatively, as shown in (d), various disturbances may occur depending on the timing of the noise, such as when noise is mixed into a continuous black run part and the original black pixel is reversed to a white pixel. In particular, if noise is mixed in at the timings shown in (c) and (d), the noise signal will be printed as spots on the receiving record paper of the called side, resulting in a significant deterioration of the image quality, which cannot be improved. It was wanted.

[0005]

SUMMARY OF THE INVENTION In view of the above circumstances, the present invention has been proposed by removing noise signals mixed into image data at the time of reading image information on the transmitting side.
It is an object of the present invention to provide an image data reading processing method that allows image transmission with improved image quality.

[0006]

[Means for Solving the Problems] The present invention proposed to achieve the above object repeatedly reads black and white binarized image data for each line from a transmission original twice, and
Each corresponding pixel of both read image data is sequentially compared as a pixel of interest, and as a result, if the pixel of interest is in a black-and-white inversion relationship with each other, each pixel of interest is further compared with the pixels located before and after it. As a result, when the pixel of interest is in a black-and-white inverted relationship with the pixels located before and after it,
Among these pixels of interest, the pixel data of the one that has the same black and white relationship with the preceding and succeeding pixels is transmitted.

[0007] According to the present invention, when the pixel of interest is in a black-and-white reversal relationship with each other, and furthermore, the pixel of interest is in a black-and-white reversal relationship with the pixels located before and after it, The data may be inverted and transmitted, thereby obtaining the same result as when transmitting pixel data of a pixel of interest that has the same monochrome relationship with the preceding and succeeding pixels.

[0008]

[Operation] In the method of the present invention, for each line from the transmitted original,
When reading black and white binarized image data, the same line is read twice, each corresponding pixel of both image data is sequentially compared as the pixel of interest, and the result of this comparison is When both of the pixels of interest are in a black-and-white inversion relationship with each other, each of the pixels of interest in the black-and-white inversion relationship is further compared with pixels located before and after the image data to which each pixel of interest belongs. As a result of this comparison, if the pixel of interest has a black-and-white inverted relationship with the pixels located before and after it, the pixel data of the pixel of interest that has the same black-and-white relationship with the pixels before and after it is transmitted. Therefore, if the image data read twice is different due to the inclusion of noise, the pixel data is corrected so that isolated pixel points do not occur.

[0009]

Embodiments Hereinafter, embodiments of the present invention will be described with reference to the drawings. FIG. 1 shows a flowchart of the image data reading processing method of the present invention, and FIGS.
The operation will be explained with reference to the schematic diagram in (c). Furthermore, 1
The image data read the first time is referred to as D, the image data read the second time is referred to as E, and the image data transmitted based on these image data D and E is referred to as S. ■The image sensor reads one line of image data from the original to be sent,
A white pixel signal (the unhatched part in FIG. 2(a)) and a black pixel signal ((a) in FIG. 2) binarized using a predetermined threshold value
) is obtained (see step 100 in FIG. 1). (2) Repeatedly reading the same line of image data to obtain image data (see (b) in FIG. 2) E (see step 101 in FIG. 1). ■Initialize the parameter k indicating the pixel of interest to 1, and set the 0th bit pixel data D(0) and E(0), which are outside the effective reading range of the transmitted document, to 0 (white pixel). (See step 102 in FIG. 1). ■ Compare the pixel data D(k) and E(k) of the pixel of interest k of each image data D and E, and if these D(k) and E(k) have the same black and white relationship, then D(k) or A process is performed in which the pixel data of E(k) is directly assigned to the pixel data of S(k) (see steps 103 and 112 in FIG. 1). ■However, attention bit k
When pixel data D(k) and E(k) have a black and white inversion relationship, first, D(k) and D(k-1
) are the same. As a result of the discrimination, in (a) of FIG. 2, D(k) and D(k-1
) have the same black and white relationship, so next, it is determined whether the pixel data of E(k) and E(k-1) one bit before are the same. As a result of the discrimination, in FIG. 2(b), E(k) and E
Since the pixel data of (k-1) is in a black-white inversion relationship, it is then determined whether the pixel data of E(k) and the pixel data of E(k+1) 1 bit later are the same. As a result of this determination,
In FIG. 2(b), since the pixel data of E(k) and E(k+1) also have a black-and-white inversion relationship, the white pixel data of D(k) is replaced with S() without taking the pixel data of E(k). k) is performed (steps 103, 104, 1 in FIG. 1).
09-111). ■While incrementing the parameter of the pixel of interest k, repeat the operations of ■ and ■ above. If the value of k exceeds the specified number of bits for one line (1728 for A4 standard size paper), the pixel generated by the above process is 1 line of image data S is transmitted (Fig. 1
(See steps 103-113). (2) In this way, the image data for each line of the transmitted document is sequentially read and transmitted, and when the reading of the image data is completed, the process is stopped (see steps 100 to 114 in FIG. 1).

In the above explanation, FIGS. 2(a) to 2(c) were taken as examples, but when the image data D and E shown in FIGS. 3(a) and (b) are read, , pixel data D
Attention bit k where (k) and E(k) have a black-and-white inversion relationship
First, it is determined whether the pixel data of D(k) and D(k-1) one bit before are the same, and then, D(k) and 1
It is determined whether the pixel data of D(k+1) after the bit is the same or not, and since they are both black and white inverted, D(
The pixel data of E(k) is converted to S without taking the pixel data of k).
(k) is performed (see steps 103 to 106 in FIG. 1).

On the other hand, when the image data D and E shown in FIGS. 4(a) and 4(b) are read, the pixel data D(
k) and D(k+1) have a black-and-white inverted relationship, but pixel data D(k) and D(k-1) have the same black-and-white relationship, and pixel data E(k) and E(k-1 ) have a black-and-white inversion relationship, but pixel data E(k) and E(k+1) have the same black-and-white relationship, so either the white pixel of D(k) or the black pixel of E(k) is directly converted to S( k) (in (c) of FIG. 4, the white pixel of D(k) is assigned to S(k)) (steps 103 and 10 in FIG. 1).
4,109,112).

Furthermore, when the image data D and E shown in FIGS. 5A and 5B are read, the pixel data D(k) and D( k+1
), but the pixel data D(k) and D
(k-1) has the same relationship between black and white, and the pixel data E
(k) and E(k-1) have a black and white inverted relationship, but pixel data E(k) and E(k+1) have the same black and white relationship, so the white pixel data of D(k) or E(k) Processing is performed to allocate any of the black pixel data to S(k) as is (in (c) of FIG. 5, D() is assigned to S(k)).
k) (step 103 in FIG. 1)
, 104, 109, 112).

As described above, according to the image data reading processing method of the present invention, based on the image data read twice for each line of the transmitted original, the pixel of interest with different pixel data is prevented from becoming an isolated point. However, for a pixel of interest where the image data read twice matches, the original data is processed as is, even if it is an isolated point.

In the method of the present invention described above, the image data of each line of the transmission document is read twice, but unlike this method, for example, the image data of each line is read three times. It is also possible to perform a processing method in which the above-mentioned reading is repeated, and when a pixel of interest whose pixel data differs from each other is generated, the data containing more white pixels or more black pixels is transmitted as the data of the pixel of interest.

[0015]

Effects of the Invention As can be understood from the above explanation, according to the image data reading processing method of the present invention, one
Based on the image data read twice for the line,
Only for pixels of interest with different pixel data, transmission processing is performed in accordance with the preceding and following pixels to prevent them from becoming isolated points, and for pixels of interest whose pixel data match, the read data is transmitted even if it is an isolated point. Since the transmission process is performed as is, it is possible to effectively remove only isolated points caused by noise, etc., and it is possible to improve the print quality of copies and the print quality of images on the receiving side.

[Brief explanation of the drawing]

FIG. 1 is a flowchart illustrating the method of the present invention.

FIGS. 2(a) to 2(c) are schematic diagrams illustrating the method of the present invention.

FIGS. 3(a) to 3(c) are schematic diagrams illustrating the method of the present invention.

FIGS. 4(a) to 4(c) are schematic diagrams illustrating the method of the present invention.

FIGS. 5(a) to 5(c) are schematic diagrams illustrating the method of the present invention.

FIG. 6 is a diagram illustrating a configuration example of reading an image of a transmission document.

FIGS. 7(a) to 7(d) are explanatory diagrams when noise is mixed into image data.

[Explanation of symbols]

E, D Image data k Noteworthy bits k-1 previous bit k+1 bit after

Claims (1)

[Claims] Claim 1: Repeatedly reading black and white binarized image data for each line from a transmitted original twice, comparing each corresponding pixel of both read image data sequentially as a pixel of interest, and comparing the results. When the pixel of interest is in a black-and-white reversal relationship with each other, each pixel of interest is further compared with the pixels located before and after it, and as a result, when the pixel of interest is in a black-and-white reversal relationship with the pixels located before and after it, these pixels are An image data reading processing method that transmits pixel data of a pixel of interest that has the same black and white relationship with the preceding and succeeding pixels.