JPH05294018A - Printing device - Google Patents

Abstract

PURPOSE:To constitute a printer having a function for eliminating noise which causes deterioration of image quality called mosquito noise occurring when JPEG compression using standard DCT processing is used. CONSTITUTION:In a printing device which prints in response to printing instruction from a computer as an external device, an interface 101 to which command group based on various references, data group based on reference of JPEG, and area separation data group of character and image sections are to be inputted, an area information detect means 103, a demodulation means 104 of JPEG data based on DCT compression, a character area correction means 105, and a printer engine 108 are provided. As a result, mosquito noise can be eliminated and image with high quality can be obtained.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a printing apparatus for outputting character information or the like created or processed by a personal computer, a workstation or the like on paper.

[0002]

2. Description of the Related Art Printing by a printer is performed by transferring data created by a personal computer, a workstation or the like, data converted into digital data from a scanner, a video device, etc. It is processed on the inside according to the print engine and printed on paper. As the performance of personal computers and workstations has improved, it has become necessary for these display screens in the computer to match what is printed. It has begun to combine processing for printing and processing for printing by a printer. Therefore, unless the processing capacity of the printer is matched with the processing capacity of the computer, the processing cannot be performed at a reasonable speed. Therefore, in order to obtain the same print result without increasing the processing capacity of the printer, it is necessary to convert the data into data that is as close as possible to the printing state of the printer in the computer and then transfer the data to the printer. However, in the case of print processing in which the resolution of the printer is high and gradation processing, color processing, etc. are included, both transfer time and processing time increase. Therefore, when data transfer is performed based on the conventional interface rule of the printer or the rule of the page description language, data compression processing is performed based on each rule, and the transfer to the printer is performed. It As a method that has been generalized as an effective method for compressing gradation data, a compression method using discrete cosine transform (DCT) has begun to be widely used. As an example of standardization, ISO (International Organization for Standardization) IEC (International Electrotechnical Commission) and CCITT (International Telegraph and Telephone Consultative Committee) are jointly conducted by JPEG (Photographic Experts Joint Committee) It is standardized. Among them, the "adaptive DCT method" is the basic method for compressing an image.

FIG. 7 shows a block diagram of a JPEG demodulation section. A data transmission unit such as a computer performs compression based on the standard, and an output device such as a printer performs demodulation based on the standardization. First, the outline of FIG. 7 will be described. The input of 701 is JPEG-compressed data, and the information from the Huffman table 703 is used to first demodulate the first Huffman-compressed data, and to separate the AC and DC components separately processed for further compression, respectively. DC component 705 and AC
The component 707 is processed and decompressed, and the high frequency component removal process is performed to increase the overall compression rate. Therefore, the dequantization process is performed by the quantizer 708, and the inverse DCT operation is performed by 710. The block-processed data is expanded in 711 and restored to the original data 712.

When the compression and decompression based on JPEG are applied to gradation data such as an image, the compression rate is excellent, and even if the compression of about 1/30 to 1/40 is performed, the original data and the restored data are restored. No difference is seen. (Assuming a data resolution of 300 to 400 dots / inch) But 2
When the value includes character data, the data contains high-frequency components. Therefore, if the compression ratio of the above degree is obtained, characters, lines, etc. for the quantization processing that is performed to remove the high-frequency components during quantization are calculated. Image quality is deteriorated in the portion where the gradation change is large. This is generally called mosquito noise.

FIG. 6 is a diagram showing an example of the mosquito noise. Reference numeral 601 is the original image, and reference numeral 602 is the image that has been JPEG-compressed and then restored. The original image is like a character 2
Even a value image is treated and processed as a gradation image when DCT compression is performed. In the restored image 602,
In particular, low gradation level mosquito noise 603 is generated around the black data, and the character data and the like are deteriorated.

Various proposals have been made as measures against this. As an example, in one encoding method of a mixed text / photograph document (Journal of the Institute of Image Electronics of Japan, Volume 20, No. 6), etc., a character portion and a photograph portion are separated, and a compression method is set to a character portion and a photograph portion. A method has been proposed in which is encoded using a different method. Since these methods use the optimum compression method for both, effective results can be obtained at the compression rate. However, in a device such as a printer in which the compression unit and the decompression unit are provided in different devices, this method requires that each processing unit be configured in the compression unit and the decompression unit, resulting in a heavy load on the printer and the like. Become.

[0007]

Therefore, particularly when dealing with gradation data, it is effective for the entire system to reduce the load on the printer which is a peripheral device, and a printer that prints under that condition is also effective. In order to match the output with the contents displayed on the computer, the image of the contents to be printed on the printer in the computer is expanded as gradation data, and standard compression is performed during transfer, There is a need for a method in which the decompression process is performed in order to obtain a high compression rate and a high quality restored image.

As a compression method for this, a standardized D
Although JPEG compression and the like using CT processing are effective, mosquito noise generated at this time particularly causes image quality deterioration. The problem is to eliminate this mosquito noise and construct a printing system that achieves high quality and high compression rate.

[0009]

In a printing device for printing according to a printing instruction from a computer or the like which is an external device, a command group based on various standards, a data group based on standards such as JPEG, and an area of a character part image part This is realized by including an interface unit for inputting a group of separated data, a region information detecting unit, a demodulating unit for data such as JPEG based on DCT compression, a character region correcting unit, and a print engine unit.

[0010]

According to the present invention, after the data transferred by the compression method based on DCT such as JPEG is demodulated, the character information part is corrected based on the area separation data added at the time of transfer, and the image quality is improved. Improvements are made.

[0011]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 shows an embodiment of the present invention, in which 101 is an interface unit, 102 is a command and data detecting means, 103 is area information detecting means, 104 is data demodulating means such as JPEG, and 105 is A character area correction means for correcting only the character area, 106 a print data control means for print processing, 107 a printer engine control means,
Reference numeral 108 is a print engine. An embodiment will be described according to the flow of data. Data, commands, etc. from the outside are input to the apparatus through the IF unit or means 101. The data group input to this device is, for example, JPE.
This is a data group standardized by G, and when a picture and a character are mixed, information for separating the two is added as described later. In the command / data detection means 102, a group of data to be printed, a group of commands for printer control from among a mixture of commands or additional information,
The data is classified into data groups related to print data. The printer control information is transmitted to the printer engine control means 107, and the data group is transmitted to the area information detection means 103. The area information detection unit 103 separates the actual print data from the information designating the character area, and the character area information is transmitted to the character area correction unit 105. The print data, that is, JPEG data, in this embodiment is decompressed by the JPEG data demodulation means 104 and restored to data to be printed. The restored data is the character area correction means 1
In 05, only the character area is subjected to the correction filter processing described later to correct the image quality. The print data control means 106 performs conversion processing depending on the engine on the data for which the correction processing has been completed, and the printer engine 10
Sent to 8.

FIG. 2 shows an example of character area information to be input by the IF means 101 and further detected by the area information detecting means 103 in the present invention. The rectangle in FIG. 2 indicates the print area of one page, and the inside rectangle is a picture, photograph, or graphic area, and the empty portion is a character area. FIG. 2A shows an example in which the areas A and B of the picture are designated, and areas other than A and B are sucked up as information by the area information detecting means 103, and the character area correcting means 105 is acquired.
Then, a correction filter is added to the parts other than A and B. In addition, FIG. 2B is an example in which the character areas themselves of C, D, E, F, G, H, and I are designated, and similarly, a correction filter is added to the designated portion. This character and picture discrimination information can be easily added as area data because the character and picture are generally processed and managed separately when the print data is expanded into pixels.

FIGS. 3 and 4 are characteristic diagrams of a character correction filter which is the center of the present invention. Before describing the filter, mosquito noise, which is a target for image quality improvement, will be described. FIG. 6 shows J in the contour portion appearing on characters, lines, etc.
It shows a state before compression and after demodulation when the PEG compression ratio is increased (for example, 1/30 or more). (A) is the original image before compression, and (b) is the JPEG demodulated image. Noise as indicated by 603 occurs in the demodulated image. This is mosquito noise. Originally, in the representation of characters and lines, there is a sharp change in the white background of the background and the printed part, and when viewed as an image, when viewed in the frequency domain, it has very high frequency components. According to the JPEG standard, DCT calculation, which is frequency axis conversion, is performed, and quantization is performed by a quantization table in order to further increase compression efficiency. When a high compression rate is set for a high-frequency component image, Results in mosquito noise generation indicated by 603 in FIG. When viewed as gradation data, this noise occurs in the low gradation part and the background is white in most cases. Therefore, it is possible to correct the character by adding an effective filter only to the character area. This filter has the characteristics shown in FIGS. 3 and 4, and is a filter arranged in the character area correction unit 105 in 105 of FIG. Figure 3,
In each drawing of FIG. 4, the horizontal axis represents the image input data and the vertical axis represents the corrected output. Both the input and output are treated as gradation data, and the higher the density, the higher the value. Figure 3
The filter of (a) is a conversion that handles only low gradation input data as white data. Low level noise is removed to improve image quality. FIG. 3B is for converting the output into a binary print image. Further, FIG. 4A shows a filter in which low density is white and medium density or higher is raised. Similarly, FIG. 4B shows a steep curve of medium-high density conversion.

FIG. 5 shows another embodiment, which has a CPU configuration.
These are shown by general building blocks such as. 501
Is an interface unit, 502 is a CPU, 503 is a ROM in which all operations are stored or character codes are stored,
Reference numeral 504 is a RAM for calculation, 505 is a data output control unit for transferring data to the engine, and 506 is an engine control unit for operating the engine. In this block, processing similar to that of the embodiment in FIG.
It is done by processing. The correction filter for the character area is configured in the ROM of 503 or as a result of calculation, as shown in FIG. 3 and FIG.

Although FIG. 7 has been described as a conventional example, the same function is located in the JPEG data demodulation means 104 of FIG. By using this standard JPEG data demodulation means, standardized data can be used as printer transfer data.

As for the method of designating the area information transferred from the outside, the same effect can be obtained by various designating methods other than FIG. Based on various instructions, it can be adapted to the character area correction means 105 by confirmation processing in the recognition unit in the printer.

[0017]

In the past, in order to increase the compression rate in filing mixed documents such as characters and photographs, it has been proposed to perform compression processing according to the contents of each, but in the case of adapting to a printer, It is desirable for the entire system to reduce the load on the printer as much as possible, and as a countermeasure, the present invention sacrifices the compression rate to some extent, but simplifies the processing in the printer and makes the printer practical. Therefore, the price can be reduced. That is, with this process, the mosquito noise in the character part can be easily erased with an appropriate filter effect, and the gradation of the character part can be used more effectively. Therefore, the legibility of the grayscale font is improved. Maintained treatment can be performed.

Regarding the reduction of the compression ratio, general characters,
When an image of about 300 to 400 dots / inch is evaluated in a mixed photo document, the limit of the compression rate that can be adopted is 1/32, but it is reduced to about 1/32. However, when considering the transfer process to the printer. Is a sufficient compression rate and is effective as a printer process.

[Brief description of drawings]

FIG. 1 is a diagram showing an embodiment of the present invention.

FIG. 2 is an explanatory diagram of an example.

3A and 3B are diagrams showing filter characteristics of the present invention.

4A and 4B are diagrams showing filter characteristics of the present invention.

FIG. 5 is a diagram showing another embodiment of the present invention.

FIG. 6A illustrates mosquito noise before compression.
FIG. 6B is a diagram showing a state after demodulation.

FIG. 7 is a diagram illustrating a conventional example.

[Explanation of symbols]

 101 Interface Means 102 Command Data Detection Means 103 Area Information Detection Means 104 JPEG Data Demodulation Means 105 Character Area Correction Means 106 Print Data Control Means 107 Printer Engine Control Means 108 Print Engines

─────────────────────────────────────────────────── ─── Continuation of front page (51) Int.Cl. ⁵ Identification code Office reference number FI technical display location H04N 1/41 Z 9070-5C

Claims (1)

[Claims] 1. In a printing device for printing according to a printing instruction from a computer or the like which is an external device, a command group based on various standards, a data group based on standards such as JPEG and a region separation data group of a character part image part, etc. An interface unit for inputting the information, a region information detecting unit, a data demodulating unit such as JPEG based on DCT compression, a character region correcting unit, and a print engine unit are provided. In addition, the printing apparatus is characterized in that the data of only the character area portion is corrected.