JP2007137007A - Dither matrix optimization device, image-forming device and dither matrix optimization method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To prevent a degradation in image quality generated in an electrophotographic image-forming means using multibeams or in an electrophotographic image-forming means using a scanning optical system by means of a polygon mirror. <P>SOLUTION: The dither matrix optimization device evaluates by an assessment means 4, image data generated according to the dither matrix on the basis of the frequency of appearance of dots generated in each beam of the multibeams or the frequency of appearance of dots formed by scanning each reflecting surface of the polygon mirror, executes by a hereditary algorithm processing means 2, a hereditary algorithm processing on the basis of the fitness of the dither matrix determined on the basis of the assessment result, and in a dither data output means 5, outputs the dither matrix from the gene information formed. This can alleviate an influence to the image quality even when there is a difference of light amount in each beam of the multibeams or when there is any unevenness in the reflecting surface of the polygon mirror, so that the degradation of image quality can be prevented. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a dither matrix optimization device that performs optimization processing of a dither matrix used for gradation expression, a printing system including the dither matrix optimization device, an image forming apparatus such as a digital copying machine, and a dither matrix optimization method. .

  In image forming apparatuses such as laser printers, digital copiers, color laser printers, and digital color copiers, new problems have arisen as the output of electrophotographic engines increases in speed and costs.

  In order to further increase the speed of an image forming apparatus such as this type of laser printer, it is necessary to rotate a rotating polygon mirror that scans a laser beam at a high speed. However, it is technically difficult to rotate the polygon mirror at a high speed, and as the polygon mirror increases in speed, uneven rotation, lower durability, and increased noise occur, resulting in a stable high-quality image. There was a problem that could not be obtained.

  In order to solve such problems, there is a so-called multi-beam type laser beam printer that simultaneously scans a plurality of laser beams from a laser on a photosensitive member by a rotary polygon mirror such as a polygon mirror in the main scanning direction. Proposed. Such a multi-beam type laser beam printer can be increased in speed, and the rotational speed of the polygon mirror can be reduced according to the number of laser beams. For example, when the number of laser beams is eight, the polygon rotation speed can be reduced to 1/8 compared to a laser beam printer having one laser beam. Therefore, the polygon mirror can reduce rotation unevenness, and can realize improvement in durability and reduction in noise.

  By the way, the light amounts of the plurality of beams on the photosensitive member may not be the same due to the influence of a lens or the like. In addition, the light amounts of the respective beams are not necessarily the same, and are not the same due to variations in circuits and semiconductor laser elements. Further, the reflection surfaces constituting the polygons do not necessarily have the same amount of light on the photosensitive member because the reflection rates of the respective reflection surfaces are not necessarily the same and there are variations in the manufacturing process. This causes a problem that the image quality is significantly impaired. In other words, the difference in the light amount of the beam appears as a potential difference on the photosensitive member, and finally becomes a difference in the toner adhesion amount.

  In addition, when the output engine (printer engine) is a color laser printer or the like, for example, an image is formed with each plate of cyan, magenta, yellow, and black (a latent image is formed and developed), so there is a slight difference in light quantity. This results in a potential difference between the photoconductors, a toner adhesion amount difference, and a color difference. When full-color continuous printing is performed, a color difference occurs even with the same write data value.

  For such a problem, as described in Patent Documents 1 and 2, it is common to define the positional relationship between the polygon mirror and the halftone dot of the dither. In Patent Document 3, the positional relationship between the number of beams and halftone dots is defined.

  On the other hand, conventionally, an image forming method relating to halftone processing that can be applied to an image forming apparatus or a display device such as a laser printer, a digital copying machine, a color laser printer, and a digital color copying machine has been used. In the conventional halftone process, a gradation reproduction process using a dither method is often used, and it is possible to express gradations and colors even with a binary printer or the like.

  By the way, when performing gradation expression processing by the dither method, it is necessary to use a dither matrix suitable for the characteristics of the output means in order to obtain a high quality output. For this purpose, it is necessary to optimize the dither matrix in accordance with the characteristics of the output means. In order to optimize the dither matrix, image output corresponding to the dither matrix is performed using an actual printer engine, and the dither matrix is corrected so that the output image quality approaches the target level of the product. There is a need.

  However, such a dither matrix optimization work greatly increases the degree of freedom of correction in proportion to the size of the dither matrix, or when the printer engine is a color laser printer, for example, cyan, magenta, yellow, Since correction according to each black version and correction according to color superposition are required, many repetitive correction operations are required, and efficient correction is difficult.

  Therefore, various dither matrix optimization methods have been proposed to solve such problems. In particular, optimization of the dither matrix using a computer is suitable for efficient dither correction.

  For example, Patent Document 4 discloses a dither matrix optimization device using a genetic algorithm. According to the dither matrix optimization device using the genetic algorithm disclosed in Patent Document 4, multi-value dither matrix information using a multi-value dither output level obtained according to a given multi-value dither matrix And the dither matrix is optimized by a genetic algorithm based on the fitness. By using the genetic algorithm in this way, even when the multi-value dither matrix size is large and the total number of multi-value dither matrices cannot be checked, the multi-value dither due to the effects of the mating, mutation, and selection of the genetic algorithm. An efficient optimum value search from the total number of matrices becomes possible.

  Further, according to Patent Document 5, for the generation of dither when used for offset printing or the like, the dither matrix information adaptability is obtained using the dither output level obtained according to the given dither matrix, and this adaptability is obtained. The dither matrix is optimized by the genetic algorithm based on it. By using the genetic algorithm in this way, even when the dither matrix size is large and the total number of dither matrices cannot be checked, the total number of multi-value dither matrices can be reduced by the effects of genetic algorithm mating, mutation, and selection. Efficient optimum value search is possible from the inside.

  The above-mentioned two cases are searches using a genetic algorithm, but the optimization method is not limited to the genetic algorithm, and other optimization methods can be used.

Japanese Patent Publication No. 05-74983 Japanese Patent No. 3530538 JP 2001-341356 A JP 2000-152004 A International Publication No. 2002/071738 Pamphlet

  However, according to Patent Documents 1 and 2 described above, since the number of polygon mirror surfaces is determined in accordance with the speed of the output engine, there is a problem that the degree of freedom of arrangement of halftone dots is limited. Similarly, in Patent Document 3 described above, the degree of freedom of arrangement of halftone dots is limited.

  The present invention has been made in view of the above, and the image quality generated in an electrophotographic image forming means using a multi-beam or an electrophotographic image forming means using a scanning optical system by a rotating polygon mirror is achieved. An object of the present invention is to provide a dither matrix optimization device, an image forming apparatus, and a dither matrix optimization method that can prevent deterioration.

  In order to solve the above-described problems and achieve the object, the invention according to claim 1 is a dither matrix optimization that performs a dither matrix optimization process in accordance with the characteristics of an electrophotographic image forming means using a multi-beam. And an evaluation unit that evaluates image data generated according to the dither matrix based on the appearance frequency of dots generated in each beam of the multi-beam, and the dither based on the evaluation result by the evaluation unit. Obtaining the fitness of the matrix, deciphering the genetic information describing the dither matrix based on the obtained fitness, and performing the mating process and the mutation process on the remaining gene information to obtain new gene information A genetic algorithm processing means to be generated, and the genetic information generated by the genetic algorithm processing means Ri comprises a dither data output means for describing the dither matrix, a.

  According to a second aspect of the present invention, in the dither matrix optimizing device according to the first aspect, the genetic algorithm processing means has the appearance of dots generated in each beam of the multi-beam, which is an evaluation result of the evaluation means. The smaller the frequency deviation, the better the fitness.

  The invention according to claim 3 is the dither matrix optimizing device according to claim 1 or 2, wherein the genetic algorithm processing means performs mating processing and mutation processing on the genetic information in each beam of the multi-beam. Information on the frequency of occurrence of generated dots is used.

  According to a fourth aspect of the present invention, there is provided a dither matrix optimizing apparatus for performing a dither matrix optimization process in accordance with characteristics of an electrophotographic image forming means using a scanning optical system using a rotary polygon mirror. Evaluation means for evaluating image data generated according to the dither matrix based on the appearance frequency of dots formed by scanning on each reflecting surface of the mirror, and adaptation of the dither matrix based on the evaluation result by the evaluation means Genes that generate new gene information by calculating the degree, deducting the genetic information describing the dither matrix based on the obtained fitness, and performing the mating process and the mutation process on the remaining gene information A dithermato using the genetic information generated by the genetic algorithm processing means and the genetic algorithm processing means Comprising a dither data output means for describing the box, the.

  According to a fifth aspect of the present invention, in the dither matrix optimizing apparatus according to the fourth aspect, the genetic algorithm processing means is formed by scanning with each reflecting surface of the rotary polygon mirror as an evaluation result in the evaluation means. It is judged that the fitness is better as the deviation in the appearance frequency of the dots is smaller.

  The invention according to claim 6 is the dither matrix optimizing device according to claim 4 or 5, wherein the genetic algorithm processing means performs mating processing and mutation processing on the genetic information in each reflection of the rotating polygon mirror. Information on the frequency of appearance of dots scanned and formed by the surface is used.

  According to a seventh aspect of the present invention, in the image forming apparatus including an electrophotographic image forming unit using a multi-beam, the dither matrix is added to the dither matrix based on the appearance frequency of dots generated in each beam of the multi-beam. An evaluation unit that evaluates the image data generated in response, and obtains the fitness of the dither matrix based on the evaluation result of the evaluation unit, and hesitates genetic information describing the dither matrix based on the obtained fitness A genetic algorithm processing means for generating new genetic information by performing a mating process and a mutation process on the remaining gene information, and a dither matrix based on the genetic information generated by the genetic algorithm processing means. The dither data output means for describing and outputting, and the dither data output means Is provided with an image processing means for performing gradation expression using the dither matrix, the image forming unit forms an image in accordance with the gradation expression in the image processing unit.

  According to an eighth aspect of the present invention, in an image forming apparatus including an electrophotographic image forming unit using a scanning optical system using a rotating polygon mirror, the appearance of dots scanned and formed by each reflecting surface of the rotating polygon mirror An evaluation unit that evaluates image data generated according to the dither matrix based on the frequency; an adaptability of the dither matrix is obtained based on an evaluation result by the evaluation unit; and the dither matrix is calculated based on the obtained fitness Genetic algorithm processing means for generating new genetic information by performing a mating process and a mutation process on the remaining genetic information by deceiving genetic information describing a matrix, and the genetic algorithm processing means A dither data output means for describing and outputting a dither matrix based on the gene information generated; And an image processing means for performing gradation expression using the dither matrix that is output by the data output means, said image forming means forms an image according to the gradation representation in the image processing unit.

  According to a ninth aspect of the present invention, there is provided a dither matrix optimization method for performing a dither matrix optimization process in accordance with characteristics of an electrophotographic image forming means using a multi-beam. An evaluation step for evaluating the image data generated according to the dither matrix based on the appearance frequency of the dot to be calculated, and the fitness of the dither matrix is determined based on the evaluation result of the evaluation step. A genetic algorithm processing step for generating new genetic information by performing a mating process and a mutation process on the remaining genetic information based on the genetic information describing the dither matrix based on the genetic information; Describe and output a dither matrix using the gene information generated by the algorithm processing step It includes a Izadeta output step.

  The invention according to claim 10 is the dither matrix optimization method for optimizing the dither matrix in accordance with the characteristics of the electrophotographic image forming means using the scanning optical system by the rotary polygon mirror. An evaluation process for evaluating image data generated according to the dither matrix based on the appearance frequency of dots scanned and formed by each reflecting surface of the mirror, and the adaptation of the dither matrix based on the evaluation result of the evaluation process Genes that generate new gene information by calculating the degree, deducting the genetic information describing the dither matrix based on the obtained fitness, and performing the mating process and the mutation process on the remaining gene information Dithering using a genetic algorithm processing step and the genetic information generated by the genetic algorithm processing step The Rikusu describes including a dither data output step of outputting.

  According to the present invention, it is possible to reduce the influence on the image quality even when there is a difference in the amount of light among the beams of the multi-beams, so that it is possible to prevent the deterioration of the image quality.

  In addition, according to the present invention, even when there are variations in the reflecting surface of the rotary polygon mirror, it is possible to reduce the influence on the image quality, so that the image quality can be prevented from being deteriorated.

  Exemplary embodiments of a dither matrix optimizing apparatus, an image forming apparatus, and a dither matrix optimizing method according to the present invention will be explained below in detail with reference to the accompanying drawings.

[First Embodiment]
A first embodiment of the present invention will be described with reference to FIGS. This embodiment is an example in which a laser printer that forms a monochrome image is applied as an image forming apparatus including a dither matrix optimization apparatus.

[1. Schematic configuration of laser printer 100]
FIG. 1 is a cross-sectional view showing a schematic configuration of a laser printer 100 according to the first embodiment of the present invention. As shown in FIG. 1, in the laser printer 100, a sheet on which an image is to be formed is set on the main body tray 101 or the manual feed tray 102, and conveyance of the sheet is started from the tray 101 or 102 by the paper feed roller 103. . Prior to the conveyance of the sheet by the sheet feeding roller 103, the photosensitive member (photosensitive drum) 104 rotates, the surface of the photosensitive member 104 is cleaned by the cleaning blade 105, and then uniformly charged by the charging roller 106. The Here, the laser beam modulated in accordance with the image signal is exposed from the laser optical system unit 107, developed by the developing roller 108, and the toner adheres. Made. The paper fed from the paper feed roller 103 is conveyed while being sandwiched between the photosensitive member 104 and the transfer roller 109, and at the same time, a toner image is transferred to the paper. The toner on the photosensitive member 104 remaining without being transferred is scraped off again by the cleaning blade 105. A toner density sensor 110 is provided in front of the cleaning blade 105, and the density of the toner image formed on the photoreceptor 104 can be measured by the toner density sensor 110. The paper on which the toner image is placed is transported to the fixing unit 111 along the transport path, and the toner image is fixed on the paper in the fixing unit 111. The printed sheets finally pass through the paper discharge roller 112 and are discharged in page order with the recording surface down.

  Here, the laser optical system unit 107 will be described in detail with reference to FIG.

  In order to reduce the size of the unit, the laser optical system unit 107 uses two folding mirrors 121 and 122 as shown in FIG. 1, but in FIG. 2, for the sake of simplicity, the folding mirror is used. Illustrations 121 and 122 are omitted.

  As shown in FIG. 2, the laser optical system unit 107 includes a light source unit 61, a cylindrical lens 62, a polygon mirror (rotating polygonal mirror) 63 as a deflecting unit, an fθ lens 64, and a toroidal lens 65. ing.

  In the example shown in FIG. 2, the light source unit 61 includes four semiconductor lasers 85, 86, 87, 88, four collimator lenses 81, 82, 83, 84, and a prism 89. The lights (four light beams) from the lasers 85, 86, 87, 88 are made substantially parallel light by the corresponding collimator lenses 81, 82, 83, 84, respectively, and then the four light beams are approximately vertical by the prism 89. This is a four-beam optical system arranged in one row.

  The operation of the laser optical system unit 107 having such a configuration will be described. In the laser optical system unit 107, the four light beams emitted from the four semiconductor lasers 85, 86, 87, 88 as described above in the light source unit 61 are collimated by the collimator lenses 81, 82, 83, 84, respectively. The light is converted into substantially parallel light, and is synthesized by the prism 89 so that the four light beams are arranged in a substantially vertical row. These four beam bundles enter the polygon mirror 63 through the cylindrical lens 62.

  The polygon mirror 63 rotates in the direction of the arrow R, and scans the four incident beam bundles in the horizontal direction (main scanning direction).

  The four beam bundles scanned in the main scanning direction pass through the fθ lens 64 and the toroidal lens 65 and scan the photosensitive member (photosensitive drum) 104 at a constant speed. In the example of FIG. 1, optical path folding mirrors 121 and 122 are provided in the middle of these optical systems.

  As shown in FIG. 2, a horizontal synchronization sensor 69 is disposed on the start side of the light beam scanning in the main scanning direction, and the horizontal synchronization sensor 69 synchronizes in the main scanning direction. The horizontal synchronization sensor 69 is thus installed at the scanning start end outside the image forming area 104a.

  The photosensitive drum 104 rotates in the direction of arrow Q, and the latent image formed in the image forming area 104a on the photosensitive drum 104 is visualized by a developing device (developing roller) 108.

  The laser printer 100 is provided with an operation panel 74 as shown in FIG. The operation panel 74 is used to display the operation state of the laser printer 100, or to set an operation mode and data during operation.

  The laser optical system unit 107 is connected to a video control unit (video control circuit) 113 and an LD drive circuit 114 which are image processing means including a dither matrix optimization device that executes a dither matrix optimization process to be described later. The video control unit 113 controls an image signal from a personal computer or a workstation, or generates an evaluation chart (test pattern) signal held inside. Data to be printed on the paper is transferred from the interface 75 to the video control unit (video control circuit) 113 and converted into bitmap data by the video control unit 113. Bitmap data from the video control unit 113 is given to the LD drive circuit 114. The LD drive circuit 114 synchronizes with the horizontal synchronization signal from the phase synchronization signal generating means 70, and the four semiconductor lasers 85, 86, 87, and 88 are modulated.

  The developing roller 108 is applied with a high voltage bias by a bias circuit 115. By controlling the bias in the bias circuit 115, the overall density of the image can be controlled. Yes.

  In such a laser printer 100, an image signal from a personal computer or a workstation is binarized by the dither method in the video control unit 113, and the LD drive circuit 114 is driven with the binarized signal of the original image, and the above-described operation is performed. Generally, an image is formed by an output engine by an electrophotographic process.

[2. Explanation of basic items of dither method]
Here, the basic items of the dither method will be briefly described. In an image forming apparatus that performs printing with the presence or absence of dots (binary data) like the laser printer 100, the dither method performs binarization by comparing the size of the original image with a threshold value arranged in a rectangular (matrix) area. This is a technique for performing binary image generation at a relatively high speed. In the dither method, N × N pixels are considered as one unit for gradation reproduction, and a corresponding N × N threshold matrix (this is called a dither matrix) is created, and this dither matrix is overlaid on the original image as a kind of mask. In addition, the density of each pixel and the corresponding threshold value are compared and binarized. That is, binarization is performed by repeatedly using the same dither matrix for all pixels. The threshold arrangement of the matrix follows a certain rule, and the image quality greatly depends on the shape of the dither matrix to be used.

  By the way, it is generally known that the number of gradations that can be expressed increases as the pattern size increases as a feature of the dither method. For example, when the pattern size of the dither matrix pattern is 4 × 4, 17 gradations can be expressed, and when it is 8 × 8, 65 gradations can be expressed. In the case of 16 × 16, up to 257 gradations can be expressed.

  Further, when multi-value conversion such as quaternary processing is performed, a plurality of dither threshold values and input pixel values are compared in the same manner as in the case of binarization, and multi-value data is output. In other words, each dither matrix becomes a threshold for multi-value conversion, such as “1” if the input pixel value is higher than the first dither threshold and “2” if the input pixel value is higher than the second dither threshold. .

[3. Dither Matrix Optimization in Video Control Unit 113]
In the video control unit 113, binarization by the dither method is performed on an image signal from a personal computer or the like. As described above, when performing gradation reproduction processing by the dither method, it is necessary to use a dither matrix that matches the characteristics of the output engine, which is an image forming means, in order to obtain high quality output. For this purpose, it is necessary to optimize the dither matrix according to the characteristics of the output engine. Here, dither matrix optimization in the video control unit (dither matrix optimization device) 113 will be described.

As shown in FIG. 2, the part relating to the dither matrix optimization of the video control unit (dither matrix optimization device) 113 includes a CPU unit 1 and a genetic algorithm which is a genetic algorithm processing means for executing genetic algorithm processing. An output image formed by the processing unit 2, a simulation processing unit 3 which is a simulation unit for obtaining a dither image output based on image data generated according to the dither matrix and output engine characteristic data, and a simulation processing unit 3 An evaluation value calculation unit 4 that is an evaluation unit that evaluates the electronic image of the image, and a dither data output unit 5 that is a dither data output unit that describes and outputs a dither matrix based on genetic information. The unit 1 is configured to control each unit.

  An outline of the operation of the video control unit (dither matrix optimization device) 113 will be described. The video control unit (dither matrix optimization device) 113 is generally a device that optimizes a dither matrix using a genetic algorithm. The genetic algorithm is a method of searching for individuals (genes) having high fitness by repeating three kinds of genetic operations, ie, selection, mating, and mutation. Such genetic algorithm processing is executed in the genetic algorithm processing unit 2. The fitness is an evaluation index of each individual (gene), and is generally obtained for each individual using an evaluation function for assigning superiority or inferiority of each gene. Individuals (genes) with low fitness will be culled.

[3-1. Explanation of Genetic Algorithm Processing Unit 2]
Here, the genetic algorithm processing executed in the genetic algorithm processing unit 2 will be described.

  First, the conversion relationship between gene information and a dither matrix will be described. For example, in the case of halftone dither, the dither mask is composed of a combination of cells. FIG. 4 shows an example of a cell in the dither, which is a simple schematic representation of a roughly rhombus cell. In practice, the number of pixels constituting the image may be much larger than the number shown in FIG. The cell is composed of a plurality of subcells, and a, b, c,. . . , P have the same shape. Here, as shown in FIG. 4, in describing the conversion relationship between the gene information and the dither matrix, such rhombus cells are rearranged according to the rules in the cells in which the cells are arranged in a line. A certain rule may be in order from the top, for example, or may be spiral from the center. Hereinafter, the description will be made according to the information arranged in this line. A, b, c,. . . , P are illustrated in the same shape, but may be in different shapes.

  FIG. 5A is a schematic diagram illustrating cells arranged in a row. The cells arranged in a line shown in FIG. 5A are prepared in 256 columns from # 0 to # 255, and using this cell, the state (entity) of dots from 0 to 255 gradations is provided. It is shown. In FIG. 5A, it is shown that a dot is hit at a position painted black. The number of pixels to which dots are applied increases in accordance with the order of gradation, and is finally filled with 255 gradations. For gradation 0, normal dots are not hit, but it may be better to hit depending on the output engine.

  On the other hand, FIG. 5-2 shows information (gene information) paying attention to the difference between gradations in FIG. Since one dot is added between the actual gradations 0 and 1 in FIG. 5A, “1” is written at a position corresponding to that as shown by # 1 in FIG. Similarly, between gradation 1 and gradation 2, “1” is set as shown in # 2 in FIG. This is repeated up to a point between gradation 254 and gradation 255 in order. In addition, although it is # 0 in FIG. 5B, information between gradation (-1) and gradation 0 is set as an actual gradation, but virtual gradation (-1) is assumed to be virtual. As a matter of fact, the data of # 0 in FIG. 5-2 is created on the assumption that no dot is hit. Looking at the collection of 1 and 0 in FIG. 5-2 in this state, it is clear that the horizontal direction becomes 1 only at some position from # 0 to # 255. In other words, this indicates that dots are to be placed at the position of the cell corresponding to the position where “1” exists in the # 1 column when the gradation is 1. When the gradation 2 is expressed, dots are shot at the positions of cells corresponding to the position where “1” exists in the # 1 column and the position where “1” exists in the # 2 column.

  As described above, all 256 columns from # 0 to # 255 in FIG. 5-2 can represent one individual gene in the genetic algorithm. However, in FIG. 5B, most positions in each column (# 0, # 1...) Are filled with “0”. Therefore, if this is used as it is, the mating and mutation, which are the operations of the genetic algorithm, are not performed effectively, and most mating / mutation processes are wasted.

  Therefore, in this embodiment, the interpretation of the collection of 1 and 0 in FIG. 5-2 described above is expanded as follows. Here, FIG. 6 shows information (gene information) focusing on the difference between gradations in FIG. 5A in correspondence with expansion of interpretation. In the gene shown in FIG. 6, the position of each column (# 0, # 1...) Is read sequentially from the left column (# 0) to the right side (# 255), and “1” appears first. In the gradation corresponding to each row, it is shown that a dot is hit at that position. That is, in the present embodiment, when there are a plurality of 1s in the horizontal direction, the leftmost “1” (the changed pixel that first appears in the same pixel of the dither matrix) is validated. Therefore, even if “1” occurs in the middle like the chromosome K1 in FIG. 6, based on the expansion of the interpretation of the present embodiment, the chromosome K1 in FIG. It will be ignored. However, in future generations, it can be expected that this chromosome works effectively by mating. Therefore, based on the expanded interpretation of the present embodiment, it is assumed that all chromosomes for # 255 are always “1”. By doing so, it is guaranteed that all dots are always filled in the 255th gradation of the actual image even if mutation or mating occurs. Therefore, by treating the group of 1 and 0 in FIG. 6 as a gene, the dither matrix can be optimized effectively by a genetic algorithm.

  Further, by treating such a group of 1 and 0 as a gene, gradation inversion can be prevented. That is, gradation skipping can be prevented by treating such a group of 1 and 0 as a gene. More specifically, since the number of dots always increases (0 or more) from gradation 0 to gradation 255, the density is the same or the density increases only. As a result, when performing mating and mutation using a genetic algorithm, it is possible to make sure that the number of dots is always increased regardless of the configuration, and it is efficient because it does not generate a lethal gene that inverts the gradation. Good evolution is possible. That is, there is an advantage that the gradation is never inverted with respect to the completed dither.

  Conversely, it is easy to create a dither mask based on genes. Here, FIG. 7 is an explanatory diagram showing a method of creating a dither mask based on genes. As shown in FIG. 7, the position of each column (# 0, # 1...) Is read sequentially from the left column (# 0) to the right side (# 255), and “1” appears first. A value corresponding to is set as a threshold value for the position. That is, the threshold value 0 is set for the place where “0” is set to # 0 (S0). Next, for the place where “1” is set in # 1, a threshold value 1 is set (S1). Next, the threshold value 2 is set for the place where “1” is set in # 2 (S2). Hereinafter, the threshold value is set in the same manner. However, when the threshold value of the corresponding portion is already set as in the case of the chromosome K1, the chromosome K1 is ignored. Finally, in S255, all dither threshold values are set. Note that 255 is sufficient as a dither threshold necessary for generating 256 gradations, and a 256 level threshold from 0 to 255 is not required as described above. Actually, the threshold value 0 is a special state and is handled specially. Usually, a threshold value of 255 level from 1 to 255 is set.

  Next, a method for initializing this gene will be described. There are two methods for gene initialization: initialization using random numbers and initialization using input initial values. It is also possible to input a dither matrix created in advance as an initial value to be input, and to proceed with optimization from there.

  Subsequently, a method of mating in the genetic algorithm will be described. Crossing in the genetic algorithm is one of genetic operations on individuals, and refers to an operation of selecting a specific gene pair from an individual group and exchanging that specific portion.

FIG. 8 is an explanatory diagram showing an example of a mating method in the genetic algorithm. The chromosome has a value of either 0 or 1, which is omitted in FIG. In the example shown in FIG. 8, it is assumed that there are two individuals A0 and A1. When mating the two individuals A0 and A1, the mating position is first determined. The method for determining the mating position is to randomly generate a random number and set the position. The example shown in FIG. 8 shows a case where the mating position (position corresponding to the gradation of the dither matrix) is determined at one place and mated. Once the mating position is determined, the individual A0's chromosomes P01 and P02 are mated to the individual A1's chromosomes P11 and P12.
A0 '= P01 + P12
A1 '= P11 + P02
Is generated. Even if the evaluation of the individuals A0 and A1 is not so good, if the information of the P01 portion of the individual A0 and the information of the P12 portion of the individual A1 are both excellent, the generated individual A0 ′ is An excellent evaluation value can be obtained. By performing the mating in this way, an effect is achieved that the evolution by the genetic algorithm can be advanced at high speed.

Here, another method of mating will be described. FIG. 9 is an explanatory diagram showing another example of the method of mating in the genetic algorithm. In the example shown in FIG. 9, mating on a cell is performed. In addition, although the value of either 0 or 1 is contained in the chromosome, it is omitted in FIG. In the example shown in FIG. 9, it is assumed that there are two individuals A0 and A1. When mating the two individuals A0 and A1, the mating position is first determined. The method for determining the mating position is to randomly generate a random number and set the position. The example shown in FIG. 9 shows a case where two mating positions (dither matrix pixel positions) are determined and mated. Once the mating position is determined, the new individuals A0 ′, A1 ′ shown below are obtained by mating the chromosomes P01, P02, P03 of the individual A0 with the chromosomes P11, P12, P13 of the individual A1.
A0 '= P01 + P12 + P03
A1 ′ = P11 + P02 + P13
Is generated. Even if the evaluation of the individuals A0 and A1 is not so good, if the information on the P01 and P03 portions of the individual A0 and the information on the P12 portion of the individual A1 are both excellent, the generated individual A0 ′ Will get a great rating. By performing the mating in this way, the effect of being able to advance the evolution by the genetic algorithm at high speed is produced.

  Next, the mutation method in the genetic algorithm will be described. Mutation in the genetic algorithm is one of genetic operations on individuals, and changes a part of genetic information with a certain probability. By causing mutations in this way, the population is less likely to fall into local stability.

  Here, FIG. 10 is an explanatory diagram showing an example of a mutation method in the genetic algorithm. The example shown in FIG. 10 is a method of selecting a location to be mutated using, for example, a random number and exchanging the position information with 1 and 0. Such mutations produce mutant individuals and produce the effect of changing the evolution by genetic algorithms.

  Here, another method of mutation will be described. FIG. 11 is an explanatory diagram showing another example of a mutation method in the genetic algorithm. The example shown in FIG. 11 is a method of selecting a chromosome of information “1” to be mutated using, for example, a random number and moving the information in the horizontal direction. Such mutations produce mutant individuals and produce the effect of changing the evolution by genetic algorithms.

[3-2. Description of Simulation Processing Unit 3]
Next, the simulation processing unit 3 will be described. FIG. 12 is an explanatory diagram showing an outline of processing of the simulation processing unit 3. As shown in FIG. 12, when the simulation processing unit 3 sets image data as a result of image processing and output engine characteristic data, an electronic image (dithered image output) of the output image is created. . For example, in the case of optimization of a dither matrix, image data (dither image output) as a result of image processing includes bitmap information obtained as a result of image processing of gradation patches from gradation 0 to gradation 255. Become. If the simulation processing unit 3 simulates the electrophotographic laser printer 100 including the laser optical system unit 107 using, for example, a polygon scanner, the output engine characteristic information is given to the simulation processing unit 3, for example, a photoconductor An electronic image (dithered image output) of the output image is formed from the potential information formed on 104. Further, the simulation processing unit 3 may simulate the final toner adhesion information. Alternatively, the simulation processing unit 3 may create an electronic image (dithered image output) of an output image by hitting an elliptical dot and filtering it.

[3-3-1. Description of Evaluation Value Calculation Unit 4]
Next, the evaluation value calculation unit 4 will be described. FIG. 13 is an explanatory diagram illustrating an example of processing in the evaluation value calculation unit 4. As shown in FIG. 13, the evaluation value calculation unit 4 performs an electronic image (dither image output) (b) of the output image formed by the simulation processing unit 3 based on the bitmap information (a) for each gradation. Then, individual evaluation and calculation (c) of the comprehensive evaluation value are performed. More specifically, as shown in FIG. 14, the evaluation value calculation unit 4 evaluates the image quality of the electronic image of the output image formed by the simulation processing unit 3 and calculates it as an evaluation value. It is.

  Here, FIG. 15 shows the gamma characteristic, the horizontal axis is the gradation level of 0 to 255, and the vertical axis is the density information of the simulation result. For example, if it is better that the gamma characteristic is a straight line, since there is a deviation from the straight line in FIG. 15, this deviation amount is used as an evaluation value for image quality evaluation for the electronic image of the output image formed by the simulation processing unit 3. Will be calculated.

  FIG. 16 shows the spatial frequency characteristic in a one-dimensional graph. The horizontal axis represents the spatial frequency and the vertical axis represents the power. Since the power is multiplied by the characteristics of the human eye (VTF characteristics), the power is attenuated as the frequency becomes higher. Since the area information in FIG. 16 corresponds to the graininess (graininess) when visually observed, a dither with less graininess (graininess) is searched by using this area information as an evaluation value. be able to.

  On the other hand, FIG. 17 is an explanatory diagram showing another example of processing in the evaluation value calculation unit 4. As illustrated in FIG. 17, the evaluation value calculation unit 4 may perform evaluation on the bitmap information (a) before simulation. More specifically, the digital perimeter (hereinafter referred to as digital perimeter) calculation (b) is applied to the bitmap information (a) before simulation. As shown in FIG. 18, the digital perimeter is obtained by calculating a perimeter for bitmap information (see FIG. 17) that is an original image.

  The point that a dither matrix suitable for the electrophotographic method is generated by including the digital perimeter calculated in this way in the evaluation value will be described. Here, FIG. 19 is an explanatory diagram showing a halftone dot pattern which is a halftone dot expressed by a set of two types of three-pixel dots. If one side of each square pixel has a length 1, (a) is 12, (b) is 8, and the peripheral length of (b) is shorter. That is, the perimeter of the portion where the portions are connected to each other like the dotted line portion of (b) is shortened. Here, in an electrophotographic system to which dither is applied, it has been conventionally known that a dot that is more stable is formed by a combined dot than by a discrete dot. That is, in an electrophotographic output engine such as a resolution of 600 dpi or 1200 dpi, exposure with one isolated dot cannot secure a sufficient potential to stably attach toner, and therefore toner adhesion (that is, The reflection density on the paper becomes unstable. This is not preferable for general users because it is recognized that the image quality is deteriorated. Therefore, if the short circumference is added to the evaluation, it is better to collect dots, so the evaluation of genes that collect dots by simple calculation is good (output of electrophotographic method that does not collect dots) For engines, it is less likely to leave unnecessary genes in the next generation). That is, when the calculation result of the peripheral length is included in the evaluation value, an effect of improving the ease of collecting dots is exhibited.

[3-3-2. Another Description of Evaluation Value Calculation Unit 4]
Next, another example of processing in the evaluation value calculation unit 4 will be described. The evaluation value calculation unit 4 may evaluate the bitmap information before the simulation (see FIG. 13A). More specifically, the digital usage frequency for each of the multi-beam lasers # 0, # 1, # 2, and # 3 to be used is calculated with respect to the bitmap information before the simulation (see FIG. 13A). This method is applied. As shown in FIG. 20, the laser beam is sequentially scanned in the order of LD0, LD1, LD2, and LD3 (semiconductor lasers 85, 86, 87, and 88). Accordingly, it is determined which LD (semiconductor lasers 85, 86, 87, 88) is used to scan the pixels that are the constituent elements of the cell (or subcell).

  Here, the evaluation value calculation will be described in detail with reference to FIG. First, as shown in FIG. 21A, the genes are sorted and divided for each laser beam. This is only sorted for convenience of explanation, and if the following usage frequency can be calculated, it is not necessary to sort in particular.

  Next, as shown in FIG. 21B, after dividing the concentration direction into a plurality of areas, as shown in FIG. 21C, the number of genes 1 contributing to the gradation change is determined. Count and make a table.

  Subsequently, as shown in FIG. 21D, it is determined whether the frequency of use is high or low for each density area, and the degree of bias is calculated for each density area. For the calculation of the bias, a test method generally used in statistics may be used, or the deviation of the number of used LDs may be simply accumulated from the average value for each density area. Then, the calculated bias value calculated for each density area is integrated in all density directions to obtain an evaluation value of the bias of the LD usage frequency (beam appearance frequency). The less this bias, the better the gene is given.

  The point that a dither matrix suitable for the electrophotographic method is generated by including the evaluation value of the bias of the LD usage frequency (beam appearance frequency) calculated in this way in the evaluation value will be described. Here, FIG. 22 shows a case where a dot consisting of five pixels is formed by LD0, LD1, LD2, and LD3. Each of the two dots shown in FIG. 22A is composed of five pixels. Here, assuming that the amount of light of LD1 is larger than that of LD1, LD2, and LD3, the toner adheres as shown in FIG. 22B, and the two dots have different densities microscopically. However, since the evaluation values are introduced so that the LD use frequency (beam appearance frequency) is substantially the same for each density block, the density variation of the toner image as shown in FIG. From the standpoint of viewing, the image quality is almost uniform, and the image quality can be significantly prevented from degrading compared to the case where the frequency of laser use is not taken into consideration.

  In this embodiment, since only the presence / absence of uneven usage frequency (appearance frequency of beam) is considered for each LD, in FIG. 20, LD is LD0, LD1, LD2, LD3, LD0. Obviously, even if used in the order of LD2, LD3, LD0, LD1, LD2,..., The effect of the invention is not affected.

[3-4. Flow of dither matrix optimization process]
Next, a series of processing operations related to dither matrix optimization in the video control unit (dither matrix optimization device) 113 will be described. FIG. 23 is a flowchart schematically showing the flow of dither matrix optimization processing.

  As shown in FIG. 23, first, gene information is initialized to give a 0th generation gene (step S1). This gene may be generated by a random number or may be generated based on a conventional dither mask. Generating initial values with random numbers has the effect of broadening the initial gene search space, and generating it based on the conventional dither mask narrows the search space, but has good characteristics. The effect is that it is possible to obtain the above at high speed.

  Next, the genetic information is converted into a dither mask (step S2), and a bitmap image of a patch of dither output at each level, that is, gradation 0 to gradation 255 is created (step S3). Here, first, evaluation values that can be evaluated in a digital state are evaluated (step S4). For example, evaluation values are calculated at this stage for the above-mentioned digital perimeter and bias in the usage frequency of LDs.

  In the subsequent step S5, a simulation image is created. In this embodiment, since an electrophotographic output engine is simulated, the toner adhesion state on the paper is obtained as an electronic image.

  An evaluation value is calculated for the electronic image (simulation image) of each patch created in this way (step S6). For example, the brightness of one patch and the granularity (roughness) of one patch are expressed by numerical values. Thereafter, the evaluation values are put together to calculate a comprehensive evaluation value (step S7).

  The above steps S2 to S7 are repeated until it is determined that the processing for all genes has been completed (Y in step S8) because there are a plurality of genes in one generation.

  If it is determined that the processing for all genes has been completed (Y in step S8), the process proceeds to step S9 to determine whether or not evolution has been completed. This is determined based on the value of the comprehensive evaluation value calculated in step S7. When the initial target value is reached, the evolution ends, or the evolution may be terminated due to time over or generation over.

  If it is determined that the evolution has ended (Y in step S9), a dither table is output (step S15), and the process ends. In the present embodiment, the dither table from the best to three types is output. It is possible to actually output with three types of dither, and select a dither from among them, or just apply the best one as it is to the output engine.

  On the other hand, when it is determined that the evolution has not been completed yet (N in step S9), the process proceeds to step S10 to determine the sputum gene. In the present embodiment, the fitness is obtained based on the comprehensive evaluation value obtained in step S7, and 20% of the lower fitness is selected from the whole gene.

  In the following step S11, actual dredging is performed. That is, of the plurality of genes, 20% of all genes are deleted from the memory.

  Next, in order to compensate for the deleted individual, a mating process (step S12) and a mutation process (step S13) are performed. In the present embodiment, half of the lost 20% is generated by the mating process, and the remaining half is generated by the mutation process.

  Thereafter, it is checked whether or not a lethal gene has been generated (Y in step S14). If a lethal gene has been generated (Y in step S14), the mutation process (step S13) is performed again.

  Now, the use of the bias of the LD usage frequency (beam appearance frequency) as the evaluation function has already been described. At this time, depending on the LD0, LD1, LD2, LD3 and the density block of each individual, Since the use frequency (appearance frequency of the beam) is calculated, using this information enables more efficient mutation and mating.

  First, the use of LD usage frequency (beam appearance frequency) information for mutation will be described. In FIG. 11 described above, the dot generation level is shifted, but information on the use frequency of the LD (appearance frequency of the beam) is used in this portion. That is, an efficient mutation is given by moving “1” where the LD usage frequency (beam appearance frequency) is high to a location where the LD usage frequency (beam appearance frequency) is low. FIG. 24 shows the occurrence probability for determining the mutation source / mutation destination for that by random numbers. Increasing the occurrence probability of the mutation source for the density block having a high use frequency (see FIG. 24A) and increasing the occurrence probability of the mutation destination for the concentration block having a low use frequency (FIG. 24B). reference). FIG. 25 is a flowchart showing the flow of such a mutation process. As shown in FIG. 25, an occurrence frequency table for genes LD0 to LD3 is obtained (step S21), the mutation source occurrence probability is changed (step S22), a random number is generated, and the change source is selected (step S23). Further, the mutation destination occurrence probability is changed (step S24), a random number is generated to select the change destination (step S25), and the mutation is executed (step S26). By processing the mutation in such a procedure, there is an effect that the use frequency of the LD (appearance frequency of the beam) is effectively equalized.

Next, the use of LD usage frequency (beam appearance frequency) information for mating will be described. FIG. 26 is an explanatory diagram showing an example of a mating method in the genetic algorithm. In addition, although the value of either 0 or 1 is contained in the chromosome, it is omitted in FIG. In the example shown in FIG. 26, it is assumed that there are two individuals A0 and A1. When mating the two individuals A0 and A1, the mating position is first determined. The method for determining the mating position is to randomly generate a random number and set the position. In the example shown in FIG. 26, a case is shown in which one mating position (position corresponding to the above-described density level) is determined and mated. In determining the mating position, selection is made with a random number, but the above-described dot occurrence frequency information is taken into consideration when the random number is generated. Once the mating position is determined, the individual A0's chromosomes P01 and P02 are mated to the individual A1's chromosomes P11 and P12.
A0 '= P01 + P12
A1 '= P11 + P02
Is generated. Even if the evaluation of the individuals A0 and A1 is not so good, if the information of the P01 portion of the individual A0 and the information of the P12 portion of the individual A1 are both excellent, the generated individual A0 ′ is An excellent evaluation value can be obtained. By performing the mating in this way, an effect is achieved that the evolution by the genetic algorithm can be advanced at high speed.

  As described above, according to the present embodiment, the image quality is evaluated with respect to the dither image output obtained based on the image data generated according to the dither matrix and the output engine characteristic data, and obtained based on the evaluation result. Genetic algorithm processing is executed based on the fitness of the dither matrix, and a dither matrix is output based on the generated gene information. This makes it possible to perform dither matrix optimization processing using a genetic algorithm regardless of the pattern size.

  In addition, according to the present embodiment, even when there is a difference in the amount of light among the multi-beams, the influence on the image quality can be reduced, so that the deterioration of the image quality can be prevented.

  In the present embodiment, it has been described that the use frequency of the LD (beam appearance frequency) is made uniform by using the bias of the LD use frequency (beam appearance frequency) as the evaluation function. Similarly, it is possible to make the frequency of use uniform on the reflecting surface of the polygon mirror 63 by the same method. Thereby, it is possible to make the appearance frequency of dots scanned and formed by the reflection surface of the polygon mirror 63 uniform, and to prevent unevenness due to the difference in reflectance of the reflection surface of the polygon mirror 63.

  FIG. 27 is an explanatory diagram showing scanning of a laser beam with respect to a cell (or a subcell). As shown in FIG. 27, in the case of a 6-sided polygon, F0, F1,. . . , F5 constitute the reflecting surface of the polygon mirror 63.

  Here, the evaluation value calculation will be described in detail with reference to FIG. First, as shown in FIG. 28A, the genes are sorted and divided for each reflection surface of the polygon mirror 63. This is only sorted for convenience of explanation, and if the following usage frequency can be calculated, it is not necessary to sort in particular.

  Next, as shown in FIG. 28 (b), after dividing the concentration direction into a plurality of areas, as shown in FIG. 28 (c), the number of genes 1 contributing to the gradation change is determined. Count and make a table.

  Subsequently, as shown in FIG. 28D, it is determined whether the frequency of use is high or low for each density area, and the degree of bias is calculated for each density area. For the calculation of the bias, a test method generally used in statistics may be used, or the deviation of the number of used reflective surfaces of the polygon mirror 63 may be simply accumulated from the average value for each density area. Then, the calculated deviation values calculated for each density area are integrated in all density directions to obtain an evaluation value of the deviation in frequency of appearance of dots scanned and formed by each reflecting surface of the polygon mirror 63. The less this bias, the better the gene is given.

  Then, by using the information on the deviation of the appearance frequency of the dots formed by the respective reflection surfaces of the polygon mirror 63 calculated as described above as an evaluation function, it is possible to significantly suppress the deterioration of the image quality. Is done. Further, by applying the information on the deviation of the appearance frequency of dots scanned and formed by the respective reflecting surfaces of the polygon mirror 63 to the probability of occurrence of mutation and mating, more efficient mutation and mating as in the case of multi-beam LD. Is possible.

  As a result, even when the reflection surface of the polygon mirror 63 has variations, the influence on the image quality can be reduced, so that the deterioration of the image quality can be prevented.

[Second Embodiment]
Next, a second embodiment of the present invention will be described with reference to FIGS. The same parts as those in the first embodiment described above are denoted by the same reference numerals, and description thereof is also omitted. In the first embodiment, a method of creating a dither mask based on a binary gene has been described. In the present embodiment, a method of creating a dither mask based on a multi-value gene will be described.

  In the description of the first embodiment, the chromosome constituting the gene is “1” or “0”. However, in this embodiment, for example, in the case of 2-bit image processing, the chromosome is “3” or “2”. There are four types, “1” and “0”.

  FIG. 29 is an explanatory diagram showing a dither mask creation method based on the multi-value gene of the present embodiment. As shown in FIG. 29, the position of each column (# 0, # 1...) Is read sequentially from the left column (# 0) to the right (# 255), and “1” appears first. The value corresponding to the selected column is set as the threshold value for that position in the first dither matrix. Further reading to the right side, the value corresponding to the column in which “2” appears first is set as the threshold value for that position in the second dither matrix. Further reading to the right side, the value corresponding to the column in which “3” appears first is set as the threshold value for that position in the third dither matrix. At this time, “2” existing before “1” present at the beginning is ignored and does not contribute to creation of the dither matrix. Similarly, “3” existing before “1” and “2” present at the beginning is ignored and does not contribute to the creation of the dither matrix. Similarly to the case of binarization, when reading from the left column (# 0) to the right side (# 255) in order, each “1”, “2”, “3” present first "1," "2," and "3" other than "" are ignored and do not contribute to the creation of the dither matrix.

  By creating the dither matrix in this way, the magnitude relation of the threshold values at the respective positions of the first dither matrix, the second dither matrix, and the third dither matrix is not reversed. Further, by using such a gene, crossover and mutation which are operations of the genetic algorithm are effectively performed.

  Here, a mutation method in the genetic algorithm of the present embodiment will be described. FIG. 30 is an explanatory diagram showing an example of a mutation method in the genetic algorithm of the present embodiment. The example shown in FIG. 30 is a method of exchanging (0, 1) (1, 2) (2, 3) (3, 0) with each other. May be. In the example shown in FIG. 30, all “3” are entered at # 255.

  Here, another method of mutation will be described. FIG. 31 is an explanatory diagram showing another example of a mutation method in the genetic algorithm. The example shown in FIG. 31 is a method of selecting chromosomes of information “1” to “3” to be mutated using, for example, random numbers and moving the information in the horizontal direction.

  Note that the mating process is performed in the same manner as in the first embodiment, and a description thereof will be omitted.

  Further, as shown in FIG. 32, the chromosomes may be rearranged so that the generated gene does not become an invalid gene. Of course, it is obvious that there is a method of handling a lethal gene in the lethal gene judgment process.

  In this way, this method can also be applied to multi-value dithering of 2 bits or more.

It is sectional drawing which shows schematic structure of the laser printer which concerns on the 1st Embodiment of this invention. It is a schematic diagram which shows the structure of a laser optical system unit. It is a functional block diagram concerning dither matrix optimization of a video control unit. It is a schematic diagram which shows a mode that an example of the cell appeared easily. It is a schematic diagram which shows the cell arranged in a line exemplarily. It is a schematic diagram which shows the information (gene information) which paid its attention to the difference between the gradations of FIGS. FIG. 5 is a schematic diagram illustrating information (gene information) focusing on the difference between gradations in FIG. It is explanatory drawing which shows the preparation method of the dither mask based on a gene. It is explanatory drawing which shows an example of the method of mating in a genetic algorithm. It is explanatory drawing which shows another example of the method of mating in a genetic algorithm. It is explanatory drawing which shows an example of the method of the variation | mutation in a genetic algorithm. It is explanatory drawing which shows another example of the method of the variation | mutation in a genetic algorithm. It is explanatory drawing which shows the process outline | summary of a simulation process part. It is explanatory drawing which shows an example of the process in an evaluation value calculation part. It is explanatory drawing which shows an example of the process in an evaluation value calculation part in detail. It is a graph showing a gamma characteristic. It is a graph showing the spatial frequency characteristics. It is explanatory drawing which shows another example of the process in an evaluation value calculation part. It is explanatory drawing which shows a digital perimeter. It is explanatory drawing which shows the halftone dot pattern which is a halftone dot expressed by the group of two types of 3 pixel dots. It is explanatory drawing which shows the scanning of the laser beam with respect to a cell (or subcell). It is explanatory drawing shown about evaluation value calculation. It is explanatory drawing which shows the case where the dot which consists of 5 pixels is formed by LD0-3. It is a flowchart which shows the flow of a dither matrix optimization process roughly. It is explanatory drawing which shows the generation | occurrence | production probability for determining a variation origin / mutation destination with a random number. It is a flowchart which shows the flow of a variation | mutation process. It is explanatory drawing which shows an example of the method of mating in a genetic algorithm. It is explanatory drawing which shows the scanning of the laser beam with respect to a cell (or subcell). It is explanatory drawing shown about evaluation value calculation. It is explanatory drawing which shows the preparation method of the dither mask based on the gene for multiple values based on the 2nd Embodiment of this invention. It is explanatory drawing which shows an example of the method of the variation | mutation in a genetic algorithm. It is explanatory drawing which shows another example of the method of the variation | mutation in a genetic algorithm. It is explanatory drawing which shows a mode that the chromosome was resorted.

Explanation of symbols

2 Genetic algorithm processing means 3 Simulation means 4 Evaluation means 5 Dither data output means 63 Rotating polygon mirror 100 Image forming apparatus 113 Dither matrix optimization apparatus, image processing means

Claims (10)

In a dither matrix optimizing apparatus that performs dither matrix optimization processing according to the characteristics of an electrophotographic image forming means using a multi-beam,
Evaluation means for evaluating image data generated according to the dither matrix, based on the appearance frequency of dots generated in each beam of the multi-beam;
The fitness of the dither matrix is obtained based on the evaluation result by the evaluation means, the genetic information describing the dither matrix is entered based on the obtained fitness, and the remaining gene information is mated and mutated. A genetic algorithm processing means for generating new genetic information by performing
Dither data output means for describing and outputting a dither matrix by the genetic information generated by the genetic algorithm processing means;
A dither matrix optimization apparatus comprising: The genetic algorithm processing means determines that the fitness is better as the deviation in the frequency of appearance of dots generated in each beam of the multi-beam which is the evaluation result in the evaluation means is smaller.
2. The dither matrix optimizing device according to claim 1, wherein: The mating process and mutation process for the genetic information in the genetic algorithm processing means uses information on the appearance frequency of dots generated in each beam of the multi-beam,
The dither matrix optimizing apparatus according to claim 1 or 2, characterized in that: In a dither matrix optimizing apparatus that performs dither matrix optimization processing according to the characteristics of an electrophotographic image forming means using a scanning optical system with a rotating polygon mirror,
Evaluation means for evaluating the image data generated according to the dither matrix based on the appearance frequency of dots formed by scanning by each reflecting surface of the rotary polygon mirror;
The fitness of the dither matrix is obtained based on the evaluation result by the evaluation means, the genetic information describing the dither matrix is deduced based on the obtained fitness, and the mating process and the mutation process are performed on the remaining gene information. A genetic algorithm processing means for generating new genetic information by performing
Dither data output means for describing and outputting a dither matrix by the genetic information generated by the genetic algorithm processing means;
A dither matrix optimization apparatus comprising: The genetic algorithm processing means determines that the fitness is better as the deviation in the appearance frequency of dots formed by the respective reflecting surfaces of the rotary polygon mirror, which is the evaluation result in the evaluation means, is smaller.
The dither matrix optimization apparatus according to claim 4, wherein The mating process and mutation process for the genetic information in the genetic algorithm processing means uses information on the appearance frequency of dots formed by scanning by each reflecting surface of the rotary polygon mirror.
6. The dither matrix optimizing device according to claim 4, wherein the dither matrix optimizing device. In an image forming apparatus provided with an electrophotographic image forming means using a multi-beam,
Evaluation means for evaluating image data generated according to the dither matrix, based on the appearance frequency of dots generated in each beam of the multi-beam;
The fitness of the dither matrix is obtained based on the evaluation result by the evaluation means, the genetic information describing the dither matrix is entered based on the obtained fitness, and the remaining gene information is mated and mutated. A genetic algorithm processing means for generating new genetic information by performing
Dither data output means for describing and outputting a dither matrix by the genetic information generated by the genetic algorithm processing means;
Image processing means for performing gradation expression using the dither matrix output by the dither data output means;
With
The image forming unit forms an image according to the gradation expression in the image processing unit;
An image forming apparatus. In an image forming apparatus comprising an electrophotographic image forming means using a scanning optical system with a rotating polygon mirror,
Evaluation means for evaluating the image data generated according to the dither matrix based on the appearance frequency of dots formed by scanning by each reflecting surface of the rotary polygon mirror;
The fitness of the dither matrix is obtained based on the evaluation result by the evaluation means, the genetic information describing the dither matrix is entered based on the obtained fitness, and the remaining gene information is mated and mutated. A genetic algorithm processing means for generating new genetic information by performing
Dither data output means for describing and outputting a dither matrix by the genetic information generated by the genetic algorithm processing means;
Image processing means for performing gradation expression using the dither matrix output by the dither data output means;
With
The image forming unit forms an image according to the gradation expression in the image processing unit;
An image forming apparatus. In a dither matrix optimization method for performing a dither matrix optimization process according to the characteristics of an electrophotographic image forming means using a multi-beam,
An evaluation step for evaluating the image data generated according to the dither matrix based on the appearance frequency of dots generated in each beam of the multi-beam,
The fitness of the dither matrix is obtained based on the evaluation result of the evaluation process, the genetic information describing the dither matrix is deduced based on the obtained fitness, and the mating process and the mutation process are performed on the remaining gene information. A genetic algorithm processing step for generating new genetic information by performing
A dither data output step of describing and outputting a dither matrix by the genetic information generated by the genetic algorithm processing step;
A dither matrix optimization method comprising: In a dither matrix optimization method for performing a dither matrix optimization process according to the characteristics of an electrophotographic image forming means using a scanning optical system with a rotating polygon mirror,
An evaluation step for evaluating image data generated according to the dither matrix, based on the appearance frequency of dots formed by scanning by each reflecting surface of the rotary polygon mirror;
The fitness of the dither matrix is obtained based on the evaluation result of the evaluation process, the genetic information describing the dither matrix is deduced based on the obtained fitness, and the mating process and the mutation process are performed on the remaining gene information. A genetic algorithm processing step for generating new genetic information by performing
A dither data output step of describing and outputting a dither matrix by the genetic information generated by the genetic algorithm processing step;
A dither matrix optimization method comprising:

Images