Abstract: In a facsimile machine, when a received image data having resolution of standard mode is stored in a reception data memory, the image data is decoded and then converted to image data of two fine or four super fine lines and stored in a first recording data memory. The image data stored in the first recording data memory is subjected to a thin-out processing at a set reduction ratio, then stored in a second recording data memory and finally recorded on a recording sheet.

Abstract: Before printing received image data, a facsimile apparatus performs a forcible white data converting operation on a predetermined range of the end regions of lines in the received image data, detects white lines having white data only from the image data subjected to the forcible white data converting operation, and performs the white-line skipping function and the ink donor film saving function based on the result of the white line detection.

Abstract: A method and associated apparatus for enhancing the display or print of greys in a two-tone digitized image consisting of storing a portion of a binary image map including a central bit under examination to be displayed and in testing if the said central bit is representative of an isolated dot or whether in a region around the said central bit in one of a plurality of predetermined directions there is an isolated dot at a predetermined distance from the said central bit, to display the central bit as a dot of reduced dimensions in such a way as to convert the display of an isolated dot in the display into a plurality of smaller dots with aggregate areas equivalent to the isolated dot and distributed around the location of the isolated dot identified from the image map.

Abstract: An image data processing apparatus binary-codes an image signal obtained from an image sensor of the CCD type or the like so as to output a binary-coded image signal. The image signal obtained from the image sensor A/D-converted into digital image data. Based upon the digital image data, average values of luminance values of the digital image data are calculated for an array of pixels in a preselected area of the CCD. This average value is used as a threshold level for binary coding the image data for these pixels at the center portion of this preselected area. Furthermore, to detect coutours this data processing apparatus calculates a gradient in the luminance values of the pixels in a portion of the preselected area of the CCD in an X direction and a Y direction, and the gradient value is used to obtain binary-coded data in accordance with the above-described method.

Abstract: A method and system implements a high addressability characteristic into an error diffusion process. A grey level value representing a pixel is received. The grey level value is interpolated to generate subpixel grey level values which correspond to a second resolution. A threshold circuit thresholds the interpolated grey level value. In parallel to the interpolation circuit and threshold circuit is an error circuit which generates a plurality of possible error values. One of the plurality of possible error values is selected based on the number of subpixels exceeding a threshold value. A portion of the selected error value is then diffused to adjacent pixels on a next scanline.

Abstract: An image reader sets the level of illumination provided by a light source and the gamma correction provided by a density correction circuit in dependence upon a user selected density setting button on a control panel.

Abstract: Digital image data containing binary pixel values are reproduced, without any loss of detail, by a reproduction unit, such as a laser printer, having a lower base resolution than the resolution of the image data, by subjecting the image data to a resolution conversion, the resolution being reduced in one direction, to a value corresponding to the resolution of the reproduction unit, and enlarged in the remaining direction.

Abstract: There is provided a system and a method thereof for producing enhanced signals from an irreversible compression first processor in which there were performed irreversible compression encoding of signals and subsequently a compression decoding thereof. The method includes defining at least two sets of signals, wherein each set contains signals having a selected property of a signal, feeding the decoded signals of the first processor to a second processor, and processing the decoded signals by the method of projection onto convex sets, thereby imposing conformity to the sets of signals.

Abstract: The present invention is made to reduce the effect of variation of pulses for driving a printer when recording image data and to improve the gray level representation of picture images to be printed. Data incoming through an image reading device are quantized and processed two pixels at a time by a gray level image processing circuit. The two-pixels' data are first discriminated by an edge discriminator to determine whether they represent an edge portion within the image or not. An object image, on which an operation for inter-pixel spacing shall be conducted, is selected according to the discrimination result. The arithmetic and logic portion performs the above-mentioned operation by two pixels at a time if the data correspond to a non-edge portion and performs correction for enhancement if the data correspond to an edge. This may eliminate the effect of dot displacement due to variation of drive pulses of a printer and improve the quality, in particular, of a gray level image portion.

Abstract: Outline vectors representing an arrangement of pixels of a binary image obtained by a binary image obtaining unit are extracted by an outline extractor. From the extracted outline vector, isolated points are extracted. These isolated points are stored in an isolated point data generator 7 as information indicative of the position of isolated point. The outline vectors other than the isolated points are stored in the vector generator 3 as vector data. When a binary image is reproduced, the stored position information of the isolated points is developed to produce vector data by the isolated point data developer 8. Accordingly, since the isolated points are stored not in the form of outline vectors, but position information, the storage capacity for storing the outline vectors can be small. This construction is effective in halftone image in particular.

Abstract: There is provided an image processing apparatus comprising an image signal generator having a first processing unit for processing the image signal in a first mode and generating a first processed signal, a discriminator for discriminating characteristics of the image signal, the discriminator having a selector for selecting the first processed signal according to a discrimination result; and a pulse width modulated signal generator for processing the first processed signal from the selector by using a pattern signal and for generating a pulse width modulated signal.

Abstract: This invention provides an image processing method and apparatus. Two types of edge emphasis processing operations having different degrees of edge emphasis are performed for image data of a read original. When paper quality of the original instructed by an operation unit is recycle paper, the edge emphasis processing having a smaller degree of the two types of edge emphasis processing operations is selected. The selected (multi-value) image data is binary-converted, and the binary-converted image data is output, thereby obtaining good image quality with respect to the read original regardless of the paper quality of the read original.

Abstract: A method and apparatus for converting a binary image into a gray scale image includes a pipeline having a number of nearly identical stages. At each stage, modifications occur to provide the gray scale image by following a sequence of steps including: (1) determining a sample mean value representing an averaging of pixel values within a neighborhood containing a selected pixel; (2) determining a weighted variation value representing pixel value variations within the neighborhood; and (3) adjusting the pixel value of the selected pixel toward the sample mean value if the difference between the pixel value and the sample mean value exceeds the weighted variation value. The adjustment to the pixel value is in the direction of the sample mean value. Preferably, the weighting of the variation value is different at each stage. At each stage along the pipeline, low pass filtering is provided to remove high frequency noise generated by the process.

Abstract: This disclosure relates to an image processing method and apparatus for applying binarizing processing to multivalue image data so as to produce binary image data, reproducing an image having shades of light and dark based upon the binary image data, and outputting the image. On the basis of the value of the inputted multivalue image data, a search is performed to determine whether there is a pixel whose value is "1" at the periphery of each pixel of the binarized image data with regard to the image where the value of the image data is small, i.e., a highlighted portion of the image. A pixel defining an isolated point is detected through this search. Next, with regard to a pixel defining an isolated point of the binarized image, if a pixel whose value is "1" exists comparatively near this pixel, the value of this pixel of interest is diffused near the pixel. If a pixel whose value is "1" exists comparatively far from this pixel, the value of this pixel of interest is diffused far from the pixel.

Abstract: In a facsimile apparatus which can be connected to a data processing apparatus for transmitting/receiving a digital signal, when a specific pattern which is to overlap two or more lines of print without a space between the lines is to be output, a pattern is developed so that the patterns in the upper and lower lines are made continuous with each other. When characters of various fonts are to be mixed, arbitrary patterns from another apparatus can be registered in the facsimile apparatus with arbitrary codes in addition to an output operation using patterns corresponding to codes registered in advance in the facsimile apparatus. For this reason, any codes can be output during an output operation of one page, and various patterns can be mixed.

Abstract: A block-based error diffusion technique allows for the printing of colored images having arbitrary screen orientations. The technique improves reproducibility by only compensating for errors arising from quantization in selected "critical blocks".

Abstract: An image processing apparatus can deal with a plurality of kinds of video signals. The apparatus includes a circuit for generating a signal indicating the kind of an input video signal, and magnification varying circuit for varying the aspect ratio of an image of the input video signal in accordance with a result of discrimination by the signal generation circuit. The aspect ratio of a pixel (picture element) is automatically corrected in accordance with the input video signal by the above-described circuits. Accordingly, it becomes unnecessary to manually set the aspect ratio of a pixel in accordance with the kind of the input video signal. Hence, operation is greatly simplified.

Abstract: An image processing apparatus for converting color images into monochromatic pattern images. The image processing apparatus includes: means for detecting the area of a color image; means for converting a color image into a monochromatic pattern image; and control means for varying such image processing in accordance with the result of detection made by the means for detecting the area. When a thin line is detected in a color area, the method of processing is determined based on the width of the detected thin line. For example, when the width of the thin line is smaller than a predetermined value, the area of such thin line may be outputted as a plain image instead of a pattern image so that it may easily be seen or recognized. The contour of a color area may be made conspicuous by combining a plain image and a pattern image.

Abstract: An image scanning device has a light source for emitting a beam, a CCD circuit board having a CCD sensor, an optical unit for guiding the emitted beam from the light source to the CCD sensor, a CCD sensor driving circuit connected with the CCD circuit board with a signal cable, a clock generating circuit for generating a clock to be transferred to the CCD sensor, a reset pulse generating circuit for generating a reset pulse for resetting an output buffer of the CCD sensor, a clamp circuit for clamping an output of the CCD sensor, and an analog-to-digital converter for converting an analog signal into a digital signal. The clock generating circuit is disposed on the CCD circuit board.

Abstract: An image compensation system which provides dilation or erosion of image features using halfbitting or fullbitting in the rendition of bitmap images, especially on a write-white printer. A region of pixels of an image is isolated which includes two or more correctable pixel locations. A set of state determination rules, based on the formation of pixels in the isolated region, is used to determine a corrected binary pixel state for each of the correctable pixels. Corrections for one correctable pixel may be considered in the state determination rules for adjacent correctable pixels. A single enhanced output pixel is provided for each image input pixel, thereby preserving the original image resolution. Performing enhancements on multiple input pixels simultaneously increases the system throughput.

Abstract: An improved method for providing a more faithful and pleasing rendition of a pixel pattern is described herein. This method first places the original pixel pattern in a bit mapped memory. A small portion of this bit mapped pattern, called an input tile, acts as an index for a look-up table containing a corrected pixel pattern portion (an output tile) corresponding to the target pixels to be corrected in the input tile. The corresponding output tile is then printed or otherwise displayed at a higher dots-per-inch (dpi), in both the horizontal and vertical directions, than the original pixel pattern to provide a corrected and higher resolution pattern. This process may also be performed using an algorithm which performs a correction on the target pixels in the input tile to generate an output tile instead of indexing a corresponding output tile in a look-up table. If the original pixel pattern is detected to be part of an image, as opposed to text, no correction of the pixel data is made.

Abstract: A method of producing a unit pattern on a printing cylinder for printing an endless pattern including the steps of: storing pixel signals expressing pixels of an original picture in memory means; distinguishing the pixel signals into first pixel signals which approximate to respective their corresponding pixel signals distant therefrom by a repetition length in a primary scanning direction and second pixel signals which are much different from the same; producing a cutting line extending across the original picture in a secondary scanning direction, said cutting line passing through the pixels of the original picture expressed by the first pixel signals as much as possible; extracting pattern pixel signals which exist between the cutting line and an imaginary line distant therefrom by the repetition length from the pixel signals; smoothing the pattern pixel signals close to the cutting line and the imaginary line; and recording the unit pattern on the printing cylinder on the basis of the smoothed pattern pix

Abstract: An image processing system with iterative error diffusion for halftoning a gray image made up of gray image signals to a binary image made up of binary image signals. The system uses two operational modes A and B to perform the error diffusion. During operational mode A, gray image signals stored in a scanline buffer are quantized to form binary image signals. Quantization errors resulting during operational mode A are distributed symmetrically to a selected set of error signals that have corresponding gray image signals in the scanline buffer and outside the scanline buffer. During a first iterative operation in mode B, error signals having corresponding gray image signals in the scanline buffer are dispersed symmetrically to the selected set of error signals.

Abstract: A facsimile apparatus of this invention discriminates a reception-possible resolution of a reception-side facsimile apparatus. When the discriminated resolution does not coincide with the resolution of read document information, the original reading resolution is converted to one acceptable at reception side, and the converted resolution is displayed as a transmission resolution.

Abstract: A video signal transient edge enhancing device and method which can clearly display an image by properly enhancing transient edges of video signals having different frequency bands according to the respective frequency bands. The device of the present invention comprises a plurality of transient edge enhancers for enhancing transient edges by respective proper bandwidths with respect to a plurality of input video signals having different frequency bands, and an enhancer selector for selecting one enhancer among the transient edge enhancers according to a selection control signal determined according to bands of the input video signals and outputting a video signal enhanced by the selected enhancer.

Abstract: A facsimile system for scanning a graphic image to produce a stream of output bits representative of an m.times.n array of picture elements of the image, each output bit being corrected to compensate for the average tone of a subarray of surrounding picture elements. The average tone is measured by a count of those gray levels, representative of the picture elements of the subarray, that exceed a nominal threshold. Thus the tone of the subarray dynamically adjusts the threshold of the gray level to output bit conversion of each picture element.

Abstract: An image conversion method using an algorithm for scaling rasterized images in both the horizontal and vertical direction. Non-integer remainders of multiplication functions are incorporated into the conversion process to provide highly accurate conversion results. The process provides expansion or contraction of the image in a best fit mode in order to provide the ideal amount of scaling to coincide with the physical requirements of the output device. In addition, conversion with maintenance of the aspect ratio of the original image can be effected with highly accurate image reproduction results.

Abstract: A method of improving the contrast in a natural scene image, in which the image is converted from an original set of color coordinates to an expression where one term has a relationship to overall image intensity. A global histogram of the image is derived for that term, which plots the populations of pixels at each possible level of intensity in the image. The signal describing the histogram is operated on with a filter having the characteristic of weakening strong peaks and valleys in the function, but not effecting flat portions of the signal. The filtered histogram signal is used for controlling the TRC mapping in a device at which the image is to be printed. Contrast is improved in areas likely to have the most image information, using the method described above, and further, dividing the image into a number of segments, each describable by a local histogram signal for that image segment. Each local histogram signal is compared to the global histogram, to determine which signals are flatter.

Abstract: Apparatus for image noise reduction which employs an adaptive, acuity-preserving, multi-directional and multi dimensional smoothing method. The method and apparatus of the present invention are applicable, inter alia, for (a) adaptive spatial noise reduction in still images, (b) adaptive temporal noise reduction in time changing image sequences, and (c) adaptive spatio-temporal noise reduction by combining the first two approaches (a) and (b).

Abstract: An improved method for providing a more faithful and pleasing rendition of a pixel pattern is described herein. This method first places the original pixel pattern in a bit mapped memory. A small portion of this bit mapped pattern, called an input tile, acts as an address for a look-up table containing a corrected pixel pattern portion (an output tile) corresponding to the target pixels to be corrected in the input tile. To derive the output tiles stored in the look-up table, a high resolution bit map is created from an input tile by scaling the input tile. The resulting pattern is then smoothed using a filter. The middle portion of the high resolution bit map corresponding to the two original target pixels intended to be corrected is then divided up into a number of pixel areas, where each area represents a pixel to be printed. A threshold number is then compared to the pixels in each area, and a corresponding output pixel is then stored, printed, or otherwise displayed.

Abstract: Method and apparatus enabling color or black/white bi-level or multi-level halftone images to be displayed or printed on various raster output devices, in particular on display devices (cathode ray tubes, plasma displays, liquid crystal displays) or on printing devices (printers, telefaxes). The originality of the present invention is found in the dither matrix R.sup.c*n which is obtained by applying on a c-fold replication of a well-dispersed threshold matrix D.sup.n a one-to-one discrete rotation of Pythagorean angle .alpha.=arctan(b/a) or .beta.=arctan(a/b) around an arbitrary point, where {a,b,c} are Pythagorean triplets satisfying Diophantine equation a.sup.2 +b.sup.2 =c.sup.2 and n is an integer. The images produced using the disclosed method are visually pleasing due to the absence of visible artifacts and to the small typical structure size. In addition, this method offers very smooth intensity gradations throughout the dynamic range.

Abstract: A texture image processing system for image processing apparatus comprises, first calculator for calculating a maximum value and a minimum value of texture image data; second calculator for calculating an average value and a gain value with employment of said maximum value and minimum value; and synthesizing processor for synthesizing texture image data by image picture data, this synthesizing processor synthesizing the texture image data with the image picture data by multiplying a difference between the texture image data and the average value by the gain value. Also, synthesizing processor synthesizes the texture image data with chromaticity information of the image picture data, or chromaticity information of the image picture data with predetermined chromaticity information, chromaticity information of the texture image data.

Abstract: In order to prevent a line break phenomenon in reproduction of an image of an original in halftone dots, RGB signals of an original (2) are contour-enhanced by a USM circuit (10), and inputted in a halftone dot signal producing circuit (30) and an edge signal producing circuit (40). The halftone dot signal producing circuit (30) converts image signals (Y-IMA to Bk-IMA) to halftone dot signals (DOT1 to DOT10) at the timing of a clock signal (CL1), to output the halftone dot signals (DOT1 to DOT10) to a synthesizing circuit (11) in response to a first output start signal (ENST). However, an output start time is delayed from the time of receiving the signal (ENST) by a time required for the halftone dot processing.

Abstract: An image processing apparatus includes a first pixel replacing unit for replacing the positions of pixels in image data in accordance with a replacement rule so as to increase redundancy of the image data, an encoder for encoding the image data supplied from the first pixel replacing unit, an optical disk device for storing the image data encoded by the encoder, a decoder for decoding the image data supplied from the optical disk device, a second pixel replacing unit for replacing the positions of pixels in the image data supplied from the decoder to restore the image data before replacement by the first pixel replacing unit, an output unit for outputting the image data, a third pixel replacing unit for replacing the positions of pixels in the image data to be outputted on the output unit, and a pixel density varying unit for varying the pixel density of the image data supplied from the third pixel replacing unit.

Abstract: A method of operating a display including providing a display; acquiring an image of a structure by measuring, for each of a plurality of elements, a magnitude representing an element of the structure; providing a plurality of contrast setting control buttons including a first contrast setting button that directly and independently modifies dark parts of the image of the structure and a second contract setting button that directly and independently modifies light parts of the image of the structure; digitizing the magnitudes according to a direct mode by assigning a digitized value d representing the measured magnitude of each of the elements; displaying a direct image by assigning a luminosity l corresponding to the digitized value d of each of the elements and transmitting to the display the luminosity l of each of the elements; transforming the direct image to a transformed image by assigning to each the of elements a digital value i which corresponds to the digitized value d of each of the elements; and d

Abstract: A method is provided for image processing to remove noise and improve smoothing of nonimpulsive noise while preserving major identifying edges of the image, all of which are accomplished corresponding to a prescribed scale of interest. The method achieves nonlinear cluster filtering relying upon a maximum entropy principle. A single scale parameter governs filtering by dictating the spatial extent of neighboring data used for clustering. This, together with the signal characteristic (e.g., variation in gray scale) determines the scale parameter in the output space. The described method is unsupervised and data driven.

Abstract: An image processing device for effecting resolution conversion includes an image signal memory (1) for storing the input pixel levels I.sub.xy, and an enlargement/reduction calculator circuit (2) for calculating a grey level G.sub.XY of each output pixel corresponding to the average of the areas of input pixel intersecting with the output pixel. The error occurring upon quantization by a quantizer (4) is calculated by an error calculator (5), and, after being delayed by one pixel by an error memory (6), is fed back to a image signal compensator circuit (3) to be added to the grey level S3 to obtain the compensated gray level S4, which is quantized by the quantizer (4) to obtain the quantized level O.sub.XY.

Abstract: An image processing system is capable of image compression including smoothing and edge detection for representing an image with a minimal amount of data and image reconstruction from the minimal data. The system comprises an array of pixel locations having intensity data representing the intensities of an image. The system also comprises a processor having a smoothing module for iteratively smoothing the intensity data to eliminate noise associated therewith. The processor also has an edge detection module for iteratively detecting pixel locations defining edges of regions of substantially constant intensity across the image. The processor further comprises a reconstruction module for reconstructing the image from the pixel locations defining edges and the intensity data associated with said pixel locations.

Abstract: The invention relates to an image processing apparatus having the function to encode or decode image data and discloses a technique in which an optimum encoding method is selected in accordance with the characteristics of an image and a high efficient encoding with little deterioration in image is executed.

Abstract: An image reading apparatus includes a projector for projecting light through a translucent film so as to produce a projected image from an original image on the film, a correction part for adjusting a quantity of the light, an image sensor for reading out an image from the projected image and outputting an analog signal indicating the read-out image, a signal conversion part for converting the analog signal into a digital signal indicating the read-out image, and a control part for measuring a peak value of digital signals supplied by the signal conversion part with respect to an effective range of the original image, and for allowing the correction part to adjust appropriately the quantity of the projected light in response to the measured peak value of the digital signals.

Abstract: In an image reading apparatus, the standard data for the normalization of the image read data are obtained with use of a line memory of one line from the standard data of a plurality of lines. The white level data and the black level data are used for the normalization. The weight average of the input data of the standard data received successively one line by one converges gradually. The converged average is used for the correction of read data of an image, and the effect of image noises is reduced and the image read data is normalized at a high precision. In a continuous reading, only the white level may be updated for shortening the time for reading.

Abstract: An image processing apparatus including an image data input function for inputting image data, and an image output function for outputting data on which image processing has been, which comprises characteristic quantities from said image data input by said image data input function, and a control function for performing suitable image processing control of the input image data by fuzzy assumptions in accordance with the characteristic quantities.

Abstract: An image processing apparatus in which a moire occurring in a half tone area can be eliminated by a suitable filter. A plurality of data blocks comprising N*N pixel data are transformed by means of a two-dimensional orthogonal transform so as to obtain an N*N matrix transformation factor block. An evaluation block is prepared which comprises N*N transformation factors each of which is the mean value of the absolute values of corresponding factors from a data block being considered and data blocks surrounding the data block to be determined. Mean values A[i] and B[i] (i=0 to L-1) of predetermined transformation factors are calculated, A[i] being mean values of factors included in a number L of first areas consecutively positioned along a diagonal line of the evaluation block, B[i] being mean values of factors included in a number L of second areas positioned adjacent to and lower in frequency to the corresponding first areas.

Abstract: An image processing unit comprises an image scanning unit for scanning an original at a plurality of predetermined line densities, a smoothing unit for smoothing the intensity of each pixel, the intensity of each pixel being corrected to be the average intensity of a predetermined number of pixels surrounding each pixel in a predetermined pixel area, a binarizing unit for binarizing the corrected pixel by a half-tone processing, and a pixel area changing unit for changing the pixel area based on the line density employed by the image scanning unit. The intensity of each pixel may be corrected, by using pixels included in the predetermined pixel area, based on the existence of a pixel corresponding to a contour or a fine line in the original in the pixel area. A clear image of a mesh picture is obtained without generation of moire, blurring of contours, or breaking of fine lines.

Abstract: There is disclosed an image processing apparatus capable of smoothing the image in obtaining a binary image, by eliminating notches through comparison of the input image data with certain notch image patterns stored in the apparatus in advance. In order to maintain an effect of smoothing even when the resolution of the input image data is increased, the image patterns employed for the above-mentioned comparison are varied according to the resolution of the input image data.

Abstract: A method for forming images composed of a matrix of dots in a printer, such as a laser printer or an LED printer, and a bit map display unit such as a CRT display, and more particularly a method for smoothing images by smoothing stairsteps included in the images to obtain a high resolution image. The method includes dividing a pixel into a plurality of subpixels along a vertical scanning line or a horizontal scanning line, and reversing a signal for a subpixel following a reversion signal to smooth the image as desired. The reversion signal may be provided by subjecting a specific pixel to a boolean operation, conjunction or disjunction, with the adjacent pixels next to the specific pixel.

Abstract: The present invention provides an image reader comprising an image reading line sensor for reading out data from an original document line by line to be converted into digital image data, a storage device capable of storing at least one line of read-out data, a sub-scanning device for controlling the image reading line sensor such that the line sensor accelerates up to a scanning speed which is equal to or higher than a record-scanning speed of an image data output device, a first controlling device for determining a first line and a last line to be read by the image reading line sensor in a sub-scanning direction, in relation to the sub-scanning speed of the image reading line sensor, and a second controlling device for controlling the storage device to sequentially store image data items which fall in an effective range on one line of image data read by the image reading line sensor, the storing of data being carried out during a fixed time period equal to or shorter than a minimum duration of a cycle in wh

Abstract: An image communication apparatus adapted to transmit image information with additional information appended thereto, includes a sender data position selection button for designating a position to which the additional information is to be appended, and a transmitter for transmitting the image information with the additional information appended to a leading portion or side of the image information in accordance with a designation by said sender data position selection button.

Abstract: An image processing apparatus, which stochastically binary-converts lower-bit data of a pixel of interest to improve image quality in a highlight portion, as well as in a low-density portion, is provided. In an image processing apparatus for quantizing multi-value image data to binary data, multi-value image data of a pixel of interest is divided into upper- and lower-bit data by a calculator (24). The lower-bit data is binary-converted by a comparator (25) using a normalized uniform random number generated by a random number generator (26) as a threshold value. The binary-converted data is added to the upper-bit data of the image signal of the pixel of interest by an adder (22), and the sum data is binary-converted by a comparator (18) using one of a plurality of weighted average values of a plurality of pixels near the pixel of interest, which values are calculated by a calculator (600), on the basis of binary data of pixels, which have already been binary-converted before the pixel of interest.

Abstract: An image processing apparatus performs an edge correction process for image signals of respective pixels sequentially inputted, each image signal having a density level corresponding to a pixel. An edge amount detector of the image processing apparatus calculates a difference between a density level of an image signal of a specified pixel from among the image signals and a product of a predetermined value and the sum of the density levels of peripheral pixels located at a predetermined distance from the specified pixel, and outputs a detection signal having a level of an edge amount of the calculated difference, and an edge amount convertor converts the edge amount of the detection signal into a converted edge amount in a predetermined manner, and outputs a conversion signal having level of the converted edge amount.