Abstract: A system and method for selective color-correction when applying three-dimensional texture to a printed substrate hinges either on a metric that is elicited from the image, or as determined by an operator. The metric may be associated with the human visual acuity curve for textural information. Other textural metrics may also be used.

Abstract: The present invention provides an image processing device which can reduce a storage capacity and perform color correction in accordance with the types of image. The image processing device of the present invention has an input correction section 6 including: input color correction section 24 for, using a matrix coefficient, correcting strength of signals for color components in image data of a supplied image; a matrix coefficient storage section 26 for storing a plurality of matrix coefficients in accordance with types of image; and control section 25 for reading a matrix coefficient corresponding to the type of a supplied image out of the matrix coefficients stored in the matrix coefficient storage section 26 and setting the matrix coefficient that has been read in the input color correction section 24.

Abstract: In image processing carried out by means of repeated execution of process set which includes N unit processes (where N is an integer equal to 3 or greater), prior to execution of the process groups, the N unit processes are assigned to a number M (where M is an integer equal to 2 or greater, but less than N) of processing modules for executing the unit processes. First, on the basis of execution results of unit processes included in the process group executed immediately prior, there is estimated an estimated load of unit processes of each class in the next process group. Then, on the basis of the estimated load, the N unit processes are assigned to the processing modules. During this process, a selected unit process which is one of the unit processes not yet assigned to a processing module, is assigned to one of the processing modules.

Abstract: A method for processing image data that employs a stamp field that contains data representing a distance metric for use with a variety of error diffusion operations. The distance metric possesses distance-order preservation permitting the distance from a previous dot to be determined by incrementing the distance metric, without knowing precisely where the previous dot was printed.

Abstract: A system and method for printing an image on a document by generating RGB data from a scanned image, converting the RGB data to CMYK data and identifying one of a first and a second type of screening. The type of screening is identified for a first one of the CMYK data based upon the density of the first one of the CMYK data and the density of at least one of the other CMYK data.

Abstract: An efficient method and system for eliminating halftone screens from scanned documents while preserving the quality and sharpness of text and line-art is disclosed. The method and system utilizes one or more independent channels with different sensitivities (e.g., Max, High, and Low) to provide high quality frequency and magnitude estimation. The most sensitive channel (Max) derives the frequency estimate, and the remaining channels (e.g., High and Low) are combined to create the screen magnitude. The Max channel is the most sensitive and will usually report the existence of frequencies even when the screen is very weak. Therefore, the screen frequency must be additionally qualified by the screen magnitude. The screen magnitude can be interpreted as the level of confidence that the local neighborhood represents half-toned data.

Abstract: A pattern which does not appear at a flat portion in normal binarization processing is set as a code pattern, and a code formed from this pattern is attached. At this time, code attachment with little degradation in image quality is implemented by selecting an unnoticeable pattern.

Abstract: A method for multi-level halftoning an input digital image to form an output digital image which includes a multi-level error diffusion halftoning process incorporating periodic dither signals whose amplitudes are adjusted as a function of the error-diffusion texture characteristics.

Abstract: A print engine/controller suited to use with a drop on demand print head. The print engine/controller works with compressed page data having both JPEG contone image layers and a bi-level image plane compressed using a Group 4 facsimile protocol. It receives compressed image plane's and effects expansion and printing in a pipeline fashion. It consists of a high speed serial interface (27) (such as a standard IEEE 1394 interface), a standard JPEG decoder (28), a standard Group 4 Fax decoder (39), a halftoner/compositor unit (29), a tag encoder (30) by which to place infrared tags into a printed page, a line loader/formatter unit 31 feeding an interface 32 to the print head (33). The decoders (28, 39) and encoder (30) are buffered to the halftoner/compositor (29).

Abstract: An adjusting dot position method for overlapped dots in a printer is disclosed. The pixels of an image are first input. Then, it determines whether the pixel is an overlapped dot or not. If yes and a pixel on a left side is a blank dot, the magenta, cyan or yellow in the pixel is moved to the left side pixel. Then, it determines whether the pixel is an overlapped dot or not. If yes and a pixel on an upper side is a blank dot, the magenta, cyan or yellow dot in the pixel is moved to the upper side pixel.

Abstract: Methods and systems for estimating single or multi-colored toner coverage on a printed page. One method includes generating color plane bitmaps corresponding to each color in a contone bitmap and calculating pixel coverage values for each of the color plane contone bitmaps. Another method includes receiving image data to be printed, generating color separations for the image data corresponding to available toner colors, printing each color separation on a printed page, scanning each printed page comprising the color separation and calculating toner coverage on each of the scanned printed pages.

Abstract: A method for image processing data representing a page to be printed includes the step of respectively receiving, in a data processing apparatus, external to a printer, data representing a bi-level black layer and a contone layer of a page to be printed. In the data processing apparatus, a halftoning process is applied to data representing the contone layer to form a bi-level cyan, yellow, magenta, and black (CYMK) layer. In the data processing apparatus, a compositing process is applied to data representing the bi-level black layer to composite the bi-level black layer over the bi-level CYMK layer to form a bi-level CYMK image of the page to be printed. The data representing the halftoned and composited layers is compressed to generate compressed image data. The compressed image data is written to the printer. An expansion process is applied to the compressed image data in the printer to generate data representing the page to be printed. The page is printed with the printer.

Abstract: In a black character detection processing, RGB data read from a line buffer is converted into YIQ data. For a subject pixel, a subject area centered on the subject pixel is set, and a judgment pixel that has the lowest brightness in the subject area is set. Then, a judgment area is set as being centered on the judgment pixel. If the absolute values of the averages of chroma components I and Q in the judgment area are both less than the corresponding thresholds T1 and T2, the subject pixel is determined as being an achromatic pixel. Black codes are added to fine-line pixels and edge pixels if the fine-line pixels and edge pixels are determined as being achromatic. Thus, the fine-line pixels and edge pixels can be specified as to be printed in black monochromatic ink. Therefore, black character or similar fine-line or edge part images can be outputted with high quality.

Abstract: An image processing apparatus includes a reduced image generating part generating a reduced image having a lower resolution than that of an original digital image. A first extracting part distinctly extracts a first foreground image and a first background image from the reduced image. A first continuous tone converting part converts, into a continuous tone, a halftone dot pattern of the first background image extracted from the reduced image. A combining part combines the first background image of which halftone dot pattern is converted into the continuous tone and an image based on the first foreground image extracted by the first extracting part.

Abstract: Historically, the creation of digital line screens was considered to be a subset of the creation of the digital cluster dot screens. The geometric constraints necessary for digital cluster dot screens were imported into the creation of digital line screens. Accordingly, the number of available angles or digital line screens was also significantly limited. In various exemplary embodiments of the systems and methods according to this invention, a digital line screen cell is defined such that the vectors defining the digital line screen cell are not necessarily isometric and are not necessarily at right angles to each other. By using a high-addressability grid that has different resolutions along x and y axis of the grid, the systems and methods according to this invention allow the components of the vectors along the high-addressability direction to be noninteger multiples of the components of the vectors along the other direction of the grid.

Abstract: In a system, such as in a digital copier, for classifying image data derived from an original image, the image data is classified by type. An algorithm is applied to the gray levels of pixels surrounding each pixel of interest. The algorithm determines whether a local maximum or minimum is part of an image of closely-spaced lines.

Abstract: A method for halftoning an image, including receiving image data including a plurality of color separations and comparing image data for first, second, third, and fourth separations to first, second, third, and fourth halftone screens. The screens are characterized by first, second, third, and fourth pairs of frequency vectors, respectively. The fourth pair of frequency vectors is identical to one of the first, second, and third pairs of frequency vectors. Further, the first, second, third, and fourth screens have an associated fill-in sequence, where the fill-in sequence of the fourth screen is such that overlap between the separation corresponding to the fourth screen and the separation corresponding to the one of the first, second, and third screens to which the fourth screen has an identical pair of frequency vectors does not occur until a combined dot area coverage of the separations is greater than 100%.

Abstract: A method and a system for de-screening an image signal. The system comprises a filter bank, a control module and a blend module. The filter bank filters the image signal and produces a set of filter output signals. The control module generates at least one control signal based on the image signal and some of the filter output signals. The blend module dynamically blends the filter output signals in accordance with the control signal to produce a de-screened output signal.

Abstract: Conventional design tools were not developed for designing square zero-shift supercells. Conventionally, solutions that enable square zero-shift supercells were found by trial and error or by exhaustive analysis. According to a first design criterion of this invention, a non-square supercell in a first frame of reference has a diagonal that is equal in length to the diagonal of a square supercell in a second frame of reference rotated at a desired screen angle to the first frame of reference. The screen angle is a function of the lengths of the sides of the non-square supercell in the first frame of reference. According to a second design criterion, if the area of the corresponding square supercell in the second frame of reference is an integer, a square zero-shift supercell can be designed based on the lengths of the sides of the non-square supercell in the first frame of reference.

Abstract: There is disclosed a method of dynamic generation of a linearized halftone matrix from a high bit resolution halftone matrix, for use in a printer device, for example a domestic or office printer device or a commercial high resolution printer device. A high bit resolution halftone threshold matrix is converted into a vector format (700). A tone correction function is applied by selecting a variable number of index values, each index value representing a threshold level. A tone corrected two-dimensional 8 bit threshold level matrix, is then applied to a corresponding print image plane comprising a plurality of pixels to obtain a two dimensional print data for each of a plurality of colors of an image. The print data is printed as a plurality of dots by a print head to form a printed image using a halftoning method.

Abstract: A printing adjustment method includes providing a plurality of solid and screened density values produced by a proofing device that represent intended density values. The method also includes providing a plurality of solid and screened density values produced by a press output device. The method also provides calculating, in response to selected ones of the plurality of density values produced by the press output device and selected ones of the plurality of density values produced by the proofing device, required percent dot values to be used to print on the press output device a plurality of adjusted density values that approximately correspond to the intended density values. In a particular embodiment, the plurality of solid density values produced by the press output device are varied approximately linearly in density along a first axis, the first axis approximately perpendicular to direction in which output of the press output device is produced.

Abstract: An image data of M-gradations in a pixel is converted into multivalued image data of N-gradations (M>N>2) in a pixel. A remarked pixel converted into the multivalued data is determined whether to be a predetermined gradation TJ (N>TJ>2) in which a tone jump occurs. When the remarked pixel converted into the multivalued data is determined to be the predetermined gradation TJ, it is determined whether a pixel of the predetermined gradation TJ exists in pixels being adjacent to the remarked pixel. When the pixel of the predetermined gradation exists, the gradation of the remarked pixel is changed to the gradation other than the predetermined gradation so as to prevent the tone jump, and an average density after multivalued image data conversion is preserved.

Abstract: A method of automatically selecting screen characteristics value to be used for printing a color image, comprising the steps of: forming a digital representation of the color image, the digital representation comprising a plurality of separations; dividing at least one of the plurality of separations into a plurality of regions; and assigning at least one screen characteristic value to each of the plurality of regions.

Abstract: A halftone screening method creates screens of a plurality of colors, where each of the screens is divided into a plurality of cells, and intentionally changes a dot position of certain colors within cells of the screens of the certain colors relative to a dot position of a reference color within a cell of the screen of the reference color, so as to suppress a color tone change caused by positional errors of the screens.

Abstract: A halftone dot image HI composed of halftone dots is produced by halftone dot meshing of an object image OI. A smoothed object image SOI and a smoothed halftone dot image SHI are then produced by smoothing the object image and the halftone dot image. An interference moire image IM is produced by obtaining a difference between the smoothed object image SOI and the smoothed halftone dot image SHI.

Abstract: Disclosed is a method, system, and program for managing calibration files in a printing system. Patches are printed using a screening algorithm and incorporating at least one output appearance factor. A calibration file is generated from measured color values of the printed patches mapping a color space for the printed patches to a color space of a printer used to print the patches. Information is associated with the calibration file indicating the printer and at least one output appearance attribute for use in selecting one calibration file to use when printing a print job.

Abstract: An image processing apparatus, which performs color conversion for input tone data in an RGB color space to obtain tone data for a CMY color space, and which performs halftone processing to convert the tone data in the CMY color space to image reproduction data, is characterized in that a gamma characteristic A for an output density for a tone in the RGB color space is identical to a gamma characteristic B for an output density for a tone in the CMY color space in the halftone processing. In accordance with the invention, for the color conversion, even when the tone data for the RGB color space between the grid points of a color conversion table is obtained by interpolating the tone data for the CMY color space, the halftone process is performed for the RGB tone data in accordance with the same gamma characteristic B. Thus, image reproduction data can be obtained that provide the same output density as the output density allocated for the tone data of the RGB color space.

Abstract: Halftone pixels are distinguished from non-halftone pixels in pixels making up an image according to a predetermined algorithm based on a result of edge detection for determining whether the pixels are edge pixels. The pixels which have been determined to be non-halftone pixels according to the predetermined algorithm, are continuous to the pixels determined to be halftone pixels according to the predetermined algorithm including those which have been redetermined to be halftone pixels and are not lower than a predetermined threshold density in density are all redetermined to be halftone pixels.

Abstract: A maximum color identifying portion 41 identifies a maximum color from coloring material colors used to reproduce a color of a line segment after color conversion into coloring material colors by a character/line color converting portion 31 is applied to the rasterized image signal. A structure deciding portion 42 outputs a screen parameter flag that controls a screen parameter in compliance with the identified maximum color and a drawing direction of the line segment such that the screen process in the maximum color is executed at a screen angle, that is different from the drawing direction of the line segment, in a character/line screen processing portion 35. The character/line screen processing portion 35 applies the screen process to the image in accordance with the screen parameter flag by using the screen parameter at the screen angle that is different from the drawing direction of the line segment. Accordingly, disappearance of the line segment due to the screen process is avoided.

Abstract: Moiré free color halftoning is achieved through the use of two dot screens and one or two line screens. The line screens may be, for example hybrid line screens. Frequency vectors associated with the screens combine to produce moiré frequency vectors representing moiré frequencies above or below the visual range. In one embodiment lines screens are used to halftone yellow and black separations and dot screens are used to halftone cyan and magenta separations. The dot screens are oriented at 15 and 75 degrees. The line screens are oriented at 45 and 135 degrees. Selected screens and screen orientations minimize screen interaction with mechanical or optical artifacts of a rendering device, thereby minimizing mechanical and optical moiré.

Abstract: A printhead controller controls printing of a page by at least one printhead. The printhead controller, in use, receives data, in a compressed format relating to the page to be printed, the data including a page description comprising a bi-level black layer and a contone color layer. The printhead controller extracts from the compressed data relating to the bi-level black layer the contone color layer. Memory is provided for temporary storage of the expanded data. A dithering/compositing unit produces a ditherer page from the data temporally stored in the memory and feeds its out put to the printhead.

Abstract: A color image forming device and method prevent occurrence of pile height moires specific to electrophotographic images in low-frequency areas and conventional moires in a digital screen set of four colors. In the color image forming device, by at least one of a halftone generating part and scanning signal forming unit in an image processing unit, differences among screen angles of three dot patterns corresponding to cyan, magenta and black toner images are in the range about 25 to 40 degrees, preferably 30 degrees; the screen angle of a first dot pattern corresponding to a yellow toner image is made equal to that of a second dot pattern corresponding to the toner image of one of cyan, magenta and black; and the screen of the first dot pattern is brought about 150 to 210 degrees, preferably about 180 degrees out of phase with the screen of the second dot pattern.

Abstract: A program creates a unitary mask for use over a whole image, with one pattern. The mask need not be full image size; it may be tiled. In the field a printer forms a mask and prints with it, best in the same session, before another image needing another mask, whether due to different image activity or type, or operating conditions. Constraints are controlled to ensure solution. Printout can start before a mask (even the part needed to finish a top swath) is done, going columnwise in each swath. In other invention facets, a program forms a mask of size significant relative to the program and prints with it. Other facets check print conditions and form adapted masks; or find nozzles that cannot back up a known-failed nozzle and to back it up form masks with all able nozzles; or assess image activity and inject mask randomness accordingly.

Abstract: Provided is a method, system, program, and data structures for halftoning an input image comprised of at least two input color components. Each input color component provides input intensity values for the color component at pixel locations in the image. At least two halftoning screens are accessed. There is one screen for each color component and halftone output generated by at least one of the screens has a lines per inch (LPI) that is at least approximately twenty percent different than the LPI of halftone output generated by one other screen. The input image is separated into the separate color components. The accessed screen for each color component is applied to the input intensity values for the color component to produce output intensity values for the color component. The combined halftone outputs for all the color components form the output pixels.

Abstract: The system and method reduces aliasing which is formed by the patterns that are introduced by the halftone screens used in the printer. The system and method provides halftone-specific anti-alias filters for obtaining optimal effective resolution in printed images. A method of reducing aliasing in a digital image includes providing a digital image; selecting a halftone screen for halftoning the digital image, wherein the halftone screen has a directional component associated with it; filtering the digital image with an antialiasing filter, the antialiasing filter having been designed to have a directional frequency response that is optimized for the directional component of the selected halftone screen; and halftoning the filtered digital image.

Abstract: A main CPU reads out CMY values as ? correction patch data from an NVRAM, and outputs a ? correction patch from a color printer section via a ? correction patch output section. The output ? correction patch (sheet) is read by a color scanner section 1 as RGB image data. The RGB image data is converted to CMY color signals by a color conversion section 131 and input to the main CPU via a buffer memory. The main CPU calculates patch read data from the CMY color signals and compares it with the ? correction patch data stored in the NVRAM. The main CPU then calculates an inverse function of a curve of the patch read data and forms ? correction tables of four colors, C, M, Y and K. The ? correction tables are used for full-color ? correction. On the other hand, the main CPU performs an arithmetic operation (error correction) of an error correction table h(x), with an output of black (K) at screen angle ?63° being f(x) and an output of mono-color black (K) at screen angle 90° being g(x).

Abstract: An image processing apparatus has interpolation calculation means which outputs, based on an input image signal P7, a binary signal P8=0 or 1 by calculating a formula, P8=S[i](xl,yl) EXOR (DF(xl,yl) AND IP(xl,yl)), using a screen pattern S[i](xl,yl), a difference pattern DF(xla,yla) based on screen patterns S[i] and S[i+1], and an interpolation pattern IP(xla,yla).

Abstract: The printer driver receives print data from an application software program. When the received print data is font data, the printer driver transmits the font data as it is to the printer. When the print data is other than font data, but is vector data, for example, the printer driver converts the received data into bitmap data of the host-based-type format. The printer driver then transfers the converted print data to the printer. In the printer, when the received print data is bitmap data, information on the bitmap data is registered in a packet structure, and is then set in a development buffer. When the received print data is font data, on the other hand, the font data is registered in the packet structure. The font data is then developed into bitmap data, and then set over the already-set bitmap data in the development buffer. Thus, composite bitmap data is created and printed out.

Abstract: An image processor includes conversion table storage for storing a plurality of conversion tables to determine a correspondence relationship between image reproduction data, including development area information, and tone data; halftone processing for receiving the tone data of respective colors and then finding, for each color, the image reproduction data at each dot by referring to one of the conversion tables; and an image reproducing engine for reproducing images by receiving the image reproduction data at each dot for each color and by determining a development area at the dot on the basis of the development area information included in the image reproduction data.

Abstract: A method for performing a digital multistage screening of a continuous-tone image with a screen having any desired screen resolution and any desired screen angle, in which the screening is carried out by reading out threshold values and comparing the threshold values read out with the color values of the continuous-tone image. Threshold value matrices are used in which the threshold values represent one or more screen dots of a given screen, with which the desired screen resolution and the desired angle are approximated. While the threshold values are being read out, an error between the coordinates of the desired screen and of the given screen is determined, and the error is compensated for by correcting the read address for the threshold value matrix, as soon as the error exceeds a predefined limiting value. Instead of the threshold value matrices, printing-value matrices can also be used.

Abstract: A method is provided for rendering a color image with a plurality of separations with a halftone process using a single screen for a plurality of separations. The screen is comprised of a plurality of pixel locations with associated threshold values and the image is comprised of a plurality of separation values. The method includes the steps of rendering a first one of the plurality of separations in accordance with the screen, wherein the plurality of pixel locations are turned on or off at a given pixel location based on a comparison of the image separation value at that pixel with the screen threshold value. The rendering of the next color separation is made in accordance with the rendering of the first separation and the screen, wherein for constant image separation values, pixel locations are turned on for the separation at pixel locations disposed in a highest available luminance region having a lowest available threshold value.

Abstract: The present disclosure provides a computer-implemented digital prepress method of interlacing images, including: providing a set of at least two graphic images to a computer; raster image processing the digital representations for each image to provide a set of screened color-separated digital files that correspond with each graphic image; interlacing the screened color-separated digital files of like color for the graphic images to produce one interlaced file for each color; optionally saving the interlaced files as digital files for each color; and outputting the interlaced digital files to a digital output device. The present disclosure also provides a lenticular product having lenticulated images prepared by the above method.

Abstract: A method and an apparatus for proofing an image on a proofing device having a set of ColB colorants at a proofing resolution ResB, wherein the image is to be output on an output device having a set of ColA colorants at a resolution ResA different from ResB, the method including (a) providing the image in a first representation having the set of ColA colorants and the resolution ResA; (b) converting the image to a second, intermediate representation having the resolution ResA and an intermediate color set that is different from the set of ColA colorants and different from the set of ColB colorants; (c) converting the image from the second, intermediate representation to a third, intermediate representation having the intermediate color set and the resolution ResB; (d) converting the image from the third, intermediate representation to a fourth representation having the set of ColB colorants and the resolution ResB.

Abstract: Historically, the creation of digital line screens was considered to be a subset of the creation of the digital cluster dot screens. The geometric constraints necessary for digital cluster dot screens were imported into the creation of digital line screens. Accordingly, the number of available angles or digital line screens was also significantly limited. In various exemplary embodiments of the systems and methods according to this invention, a digital line screen cell is defined such that the vectors defining the digital line screen cell are not necessarily isometric and are not necessarily at right angles to each other. By using a high-addressability grid that has different resolutions along x and y axis of the grid, the systems and methods according to this invention allow the components of the vectors along the high-addressability direction to be noninteger multiples of the components of the vectors along the other direction of the grid.

Abstract: A method for halftoning an image, including receiving image data including a plurality of color separations and comparing image data for first, second, third, and fourth separations to first, second, third, and fourth halftone screens. The screens are characterized by first, second, third, and fourth pairs of frequency vectors, respectively. The fourth pair of frequency vectors is identical to one of the first, second, and third pairs of frequency vectors. Further, the first, second, third, and fourth screens have an associated fill-in sequence, where the fill-in sequence of the fourth screen is such that overlap between the separation corresponding to the fourth screen and the separation corresponding to the one of the first, second, and third screens to which the fourth screen has an identical pair of frequency vectors does not occur until a combined dot area coverage of the separations is greater than 100%.

Abstract: An image processing system includes an image input unit having a plurality of image input channels; a parallel image data processing unit that receives image data from the image input channels of the image input unit, divides the image data received into a plurality of divided data, each divided data including data in a number that is less than a predetermined number, and processes in parallel the divided image data; and a sequential image data processing unit that sequentially processes the divided image data output from the image parallel data processing unit while switching the divided image data for the image input channels.

Abstract: A method/algorithm enables the rapid development and tuning of multi-level and multi-frequency halftone screens to improve printer performance. The screens are primarily directed to laser printers. The method/algorithm generally involves generating a lattice of points by selecting a plurality of points along an imaginary line whose path extends along a surface of a geometric shape; assigning a dot shape to each of the lattice points; selecting a growth model for each dot shape; quantizing the lattice points and assigned dots to a specified grid of pixels; and generating multi-frequency screens from the lattice of points and associated dots.

Abstract: An image reproduction engine which causes toner to adhere to a development region of certain area located at a certain position within dots according to image reproduction data is utilized for image processing, wherein a halftone is expressed by means of halftone spots formed from a plurality of dots. The centroid of the halftone spot formed from a single dot or a plurality of adjacent dots is shifted from the center of the dot to an arbitrary position, thus achieving desired screen angles or desired pitches of halftone spots. As a result, screen angles related to an irrational tangent can be realized, and the pitches of halftone spots of a plurality of color screens can also be made uniform. 31.

Abstract: Methods and apparatus are provided for dynamically halftoning image data, substantially in real-time. Halftone threshold values are calculated and stored in memory, and halftone output values are calculated for the portions of the image data that will be displayed on a display device. Halftone screen parameters may be modified, and new halftone screen threshold values may be calculated for any affected halftone screens. For unaffected halftone screens, the previously stored threshold values may be retrieved from memory. In addition, the portion of the image to be displayed may be modified, and the halftone output values of the new portion may be calculated and then displayed.

Abstract: There is provided an image forming apparatus that enables the formation of high-quality multi-level images by ensuring that the size of blocks into which the image is to be divided is specified in accordance with the resolution of the image to be recorded. The apparatus divides input image gradation level data into a plurality of blocks and performs graduated recording by converting the divided image gradation level data into multi-level recording data in accordance with gradation conversion characteristics determined as keyed to relative positions in each of the blocks; the blocks are typically set to a size satisfying the following relation: (55/n)<X/p<(65/n), provided p is an integral multiple of 10; where X is the resolution of the image to be recorded in a given direction and expressed in dpi, p is the length of each block in said given direction and expressed in dots, and n is a natural number.