JPH0253384A - Picture forming device equipped with high frequency noise removing device - Google Patents

Abstract

PURPOSE:To improve printing quality by discriminating a high frequency noise and an original picture according to the size of a black picture element data continuing part and reloading a picture element part corresponding to a high frequency noise component to a picture element component not to include the high frequency noise component. CONSTITUTION:When an original P surface is divided into fine parts in a matrix shape by the picture element and white and black levels are detected in each picture element, a part, in which 'a' (Japanese syllabary-KANA) and 'o' (KANA) are printed, and a part, in which a black point B exists, are detected as the black level. An area to be composed of the plural picture elements is specified and when the white and black levels are detected concerning this area, the area of the black level goes to be large in comparison with the black point B. Namely, the part, in which the continuing part of black picture element data largely appears and the 'a' (KANA) - 'o' (KANA) are printed, gets out of the area and the black point B is received in the area. When logical arithmetic is executed on the basis of this area, printing (the original picture) and the black point B are discriminated. Thus, the high frequency noise is removed before printing and the quality of the printing original can be improved.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention provides an image forming apparatus equipped with a high-frequency noise removal device that can prevent spot-like dirt, etc., present on a document from being printed as high-frequency noise during printing. Regarding.

Current image forming devices such as facsimile machines and copying machines remove spot-like dirt and scratches that exist on the document when printing (for facsimile machines, when sending, and when copying machines). When copying), black dots B are printed as shown in FIG. 4, resulting in a problem in that the quality of the transmitted (copied) document is significantly impaired. Therefore, there is a strong demand for an image forming apparatus that can prevent such problems from occurring and improve the quality of transmitted (copied) originals.

This problem occurs because the document reading device detects (picks up) the dirt and scratches as high frequency noise components as shown in FIG. 4, and prints them as they are.

The present invention has been made in response to such demands, and provides an image forming apparatus equipped with a high-frequency noise removal device that can significantly improve the quality of printed originals by removing high-frequency noise before printing. The purpose is to

The present invention also provides an image forming apparatus equipped with a high frequency noise removal device that prevents stains in the form of grooves and grooves present on a document from being printed as high frequency noise during printing. ,
A one-dimensional image sensor that sequentially reads a series of pixel data in the main scanning direction of the document as line data in the sub-scanning direction one line at a time; and associating the line data read by the one-dimensional image sensor with the sub-scanning direction. A line memory stores the line data as a series of line data groups, one line at a time is read out sequentially from the line data group stored in the line memory, the read line data is compared with the pixel data of the adjacent read line, and multiple lines are read out. Logic calculation means detects a continuous part of black pixel data that extends across the reading line, and distinguishes between high frequency noise and a document image based on the size of the continuous part of black pixel data, and the logic calculation means determines that it is high frequency noise. In this case, the pixel portion corresponding to the high-frequency noise component of the corresponding line data is rewritten to a pixel component that does not include the high-frequency noise component, and the rewritten line data is transmitted to the printing unit. It is characterized by

As shown in Figure 4, page P of the original is subdivided into pixels into a matrix, and the white and black levels (white and black pixel data) of each pixel are detected. The portions to which "a", "iJ, ruJ,""eJ,roJ" are applied and the portions where the black dot B exists are detected as black levels.

However, if we specify an area consisting of multiple pixels and detect the white and black levels for this area, the black level area will be larger than the black point, and the continuous part of black pixel data will be large. The characters that appear are rAJ, “I”, “
The parts marked with Uj, ``Ej, and O'' will protrude from the area, and the black point B will fit within the area. Therefore, if a region is specified and a logical operation described later is performed based on this region, it becomes possible to distinguish between print (original image) and black dot B.

Disaster - one ship in flames Embodiments of the present invention will be specifically described below based on the drawings.

FIG. 1 is a block diagram showing a system configuration when the present invention is applied to a facsimile machine.

First, the system configuration of this facsimile machine will be explained. When a document P is set in the facsimile body (not shown), as the document P moves in the feeding direction (hereinafter referred to as the sub-scanning direction), the CCD 1 reads the image information on the document P as shown in the figure ■, ■, ■. One line at a time is read as shown by (1), (2), etc. (hereinafter, this reading direction will be referred to as the main scanning direction). Then, the A/D converter 2 binarizes each pixel data in the main scanning direction that has been taken by the CCD 1 once, and the processed data is manually input to the CPU 10.

The CPU 10 has a logic operation unit 11 and a small-capacity memory 12, and binarized line data input from the A/D converter 2 (hereinafter simply referred to as line data ■, ■, ■).
, ■, ■, etc.) are taken into one address of the memory 12. The memory 12 stores addresses corresponding to five lines of line data, namely, first, second, third, fourth, and fifth lines.
, and performs a logical operation as described below based on the line data to be imported to the fifth address and the four line data that have been imported to the first to fourth addresses. Execute and extract pixel data corresponding to high frequency noise components in each line data,
In order to rewrite this extracted portion to pixel data that does not include high frequency noise components, a predetermined rewriting execution command signal is issued to the high frequency noise removal circuit 14.

Note that the line data taken into the first to fourth addresses of the memory 12 is input data from the A/D converter 2 when the first logical operation is executed, as will be described later, but when the second logical operation is performed, the line data is input from the A/D converter 2. After the logical operation, line data is obtained via the memory 13 and high frequency noise removal circuit 14, which will be described next.

CPU10 also stores line data input from A/D converter 2 in memory 13 (external memory). The memory 13 consists of a line memory that stores five line data at each address, that is, the first to fifth addresses,
Each time the logical operation is completed, each address is shifted,
The line data stored at the first address is transmitted via the high frequency noise removal circuit 14 to the printer section (none of which is shown) of the destination facsimile machine, and the second
- The line data stored at the fifth address is input to the memory 12 via the high frequency noise removal circuit 14.

The high-frequency noise removal circuit 14 is equipped with a shift register, an address counter, etc., and sequentially reads line data inputted from the memory 13 as a series of pixel data in the main scanning direction, and stores these pixel data at respective addresses. Once the rewriting execution command signal is input from CPU10, the pixel data at the corresponding address is changed to pixel data that does not contain high-frequency noise components, that is, black level pixel data is changed to white level pixel data. rewrite. The rewritten line data will be input to the memory 12 as described above.

Next, the content of the calculation control of the CP tJ 10 will be explained in detail based on the flowchart shown in FIG. 2 and the logical calculation diagram conceptually shown in FIG. When the starting edge detection signal of the original P is input from a detector (not shown), the CPU 10 causes the CCD 1 to read the original P at a predetermined pitch, and outputs line data ■, ■, ■ through the A/D converter 2. , ■ into the memory 12 (Sl).

Next, the next line data ■ is stored in the memory 12 in the same manner.
(S2), line data ■, ■ to memory 12
,■,■,■ is determined whether or not S3),
After confirming that it has been imported, first take out the line data ■, ■, ■ and send it to the logic operation section 11 as shown in Fig. 3 (a).
The logical operation shown in is executed (S4).

That is, in the line data table (showing line data groups) shown on the left side of FIG. 3(a), first line data ■ and ■
, and the vertical direction (sub-scanning direction) pixel data at the corresponding positions of ■ and ■, respectively, and find the black level pixel data indicated by double diagonal lines in the figure for either of ■ and ■ or ■ and ■. If it exists in isolation, it is determined that a high-frequency noise component exists, and the corresponding part is at the black level as shown on the line data in the center of the line data table located in the middle of Figure 3 (a). I recognize that. Incidentally, in this case, the line data 2.3.4.6.7.9.14.15.16.
17. Recognize that the 18th pixel data is the black level.

The reason why high frequency noise components can be distinguished from printing by the above logical operation is as follows. In other words, according to the principle described in the operation section above, in the case of printing, the black level is continuous between multiple adjacent line data, so by detecting isolated black level pixel data, high frequency noise can be eliminated. It is possible to distinguish between the component and the print.

Next, take out the line data ■ and ■, compare these data with the above recognition data by the method shown below,
The final discrimination between printing and high-frequency noise components (more fine-grained discrimination performed by further enlarging the area) is performed to determine whether or not high-frequency noise components ultimately exist, that is, to determine whether or not high-frequency noise components are present at the time of transmission. It is determined whether the corresponding high-frequency noise component should be deleted as invalid data so as not to be captured (S5). If invalid data exists, the position of this invalid data on the line data is extracted as shown by double diagonal lines on the line data on the right side of FIG. 3(a) (S6).

Here, the above comparison is based on the pixel data of the recognition data to be determined (limited to pixel data determined to be black level by the recognition process) in the line data table in the middle of FIG. 3(a). Line data at positions corresponding to diagonally in front of this, diagonally behind, vertically, and horizontally■
and the pixel data of the line data ■, and if there is no continuous black level pixel data in these areas (
However, within three consecutive pixels in the horizontal direction), this is determined to be invalid data.

Then, the position of this invalid data is extracted as the corresponding position of the erased portion, as indicated by double diagonal lines on the line data shown on the right side of FIG. 3(a). Incidentally, in FIG. 3(a), the extracted portion is the 6th, 7th, and 9th pixel data.

Next, when the first logical operation is completed, the line data (1), (2), (2), (2), (2) are stored in the corresponding address of the memory 13. That is, the line data ■, ■, ■, ■, ■ are stored in the first, second, third, fourth, and fifth addresses, respectively (S7).

Next, the line data ■ stored in the first address of the memory 13 is transmitted to the printer section of the destination facsimile body, and the line data ■, ■, ■, ■ are sequentially input to the high frequency noise removal circuit 14 (S8). .

At this time, if invalid data obtained by the first logical operation is present in the line data ■, ■, ■, it is determined that the pixel data is being rewritten (S9), and the rewriting execution command signal is generated by high-frequency noise. It is applied to the removal circuit 14. Then, the high frequency noise removal circuit 1 that received this
4 rewrites the corresponding pixel data of the line data ■, ■, ■ to the white level (S10).

Incidentally, in this case, the portions not shown in FIGS. 3(a) and 3(b) and surrounded by circles will be rewritten. After the rewriting process, the line data (1), (2), (2), and (2) are respectively taken into the first, second, third, and fourth addresses of the memory 12 (Sll). Note that, if invalid data does not exist in the line data, this line data is taken into the memory 12 as is.

In this way, the line data from which high-frequency noise components have been removed is transmitted to the printer section of the destination facsimile machine, resulting in high-quality printing that does not include high-frequency noise components.

Next, CPU10 stores the new line data ■ in memory 1.
2, and executes the same processing as 83 to Sll described above based on these line data (S
12). Note that FIG. 3(b) shows the logical operation results in this case. As shown in the figure, in this case, invalid data, that is, no erased extracted portion exists.

Thereafter, the process similar to S12 above is executed a necessary number of times (313), and when the end detection signal of the document P is input from the detector, this is confirmed (514), and the final process is executed for the first page of the document P. Execute (S15). Then, the rewriting process will be performed on all other line data (2), (2), (2), etc. except for the last line data. After this last process, the memory 13
The line data for the final five lines will be stored in , but since there is no new line data for the first page anymore, the line data for the final five lines is sent to the printer section (S16).

Then, it is determined whether the processing for the first page has been completed (517), and when it is confirmed that the processing has been completed, the processing for the second page is executed, and the processing for all pages is completed. When this is confirmed (S18), the process ends.

Note that FIG. 3(C) and FIG. 3(d) show the results of the third and fourth logical operations, respectively.

In the above embodiment, as a logical operation for distinguishing between printing and high-frequency noise components, three adjacent pixels are first compared in the sub-scanning direction (vertical line), and then, using the pixel of the recognition data as a reference, A total of eight pixels in the diagonal front, top and bottom (vertical), diagonal back, and horizontal directions of adjacent line data were compared at 4 degrees, but it is also possible to match eight or more pixels, or the above Instead of a two-step process, an embodiment may be adopted in which multiple pixels are compared at one time. In short, it is sufficient to perform the same processing as described above in a region where printing and high-frequency noise components can be substantially distinguished. Further, in the above embodiment, the present invention is applied to a facsimile machine, but it can be similarly applied to a copying machine. However, in this case, a transmission signal will be emitted from the high frequency noise removal circuit to the printing section of the copying machine.

According to the present invention as described above, it is possible to reliably distinguish between printing on a document and high-frequency noise components caused by spot-like dirt, scratches, etc. present on the document. Since it can be removed in advance before sending, copying, etc., printing quality can be significantly improved.

[Brief explanation of the drawing]

Fig. 1 is a block diagram showing the system configuration when the present invention is applied to a facsimile machine, Fig. 2 is a flowchart showing the content of CPU calculation control, and Fig. 3 is a conceptual explanation showing the content of logical calculations by the CPU. 4 are explanatory views showing the principle of the present invention. 1 ・ ・ ・ CCD, 10 ・ ・ ・ CPU
, 11...Logic operation section, 12...Register,
13...Memory, 14...High frequency noise removal circuit. P...Manuscript.

Claims (1)

[Claims] (1) An image forming apparatus equipped with a high-frequency noise removal device that prevents spot-like dirt, etc. present on a document from being printed as high-frequency noise during printing, the image forming apparatus comprising: A one-dimensional image sensor that sequentially reads pixel data line by line in the sub-scanning direction, and stores the line data read by the one-dimensional image sensor as a series of line data groups associated with the sub-scanning direction. 1 from the line memory and the line data group stored in the line memory.
Line by line is read out sequentially, and the read line data is compared with the pixel data of adjacent read lines to detect a continuous part of black pixel data that spans multiple read lines, and the size of the continuous part of black pixel data is detected. According to
A logic operation means that discriminates between high frequency noise and a document image, and when the logic operation means determines that it is high frequency noise,
It is characterized by comprising a high frequency noise removal circuit that rewrites the pixel portion corresponding to the high frequency noise component of the corresponding line data to a pixel component that does not include the high frequency noise component and transmits the rewritten line data to the printing unit. An image forming apparatus equipped with a high frequency noise removal device.