Abstract: A method and apparatus for inspecting the surface of articles, such as chips and wafers, for defects, includes a first phase of optically examining the complete surface of the article inspected at a relatively high speed and with a relatively low spatial resolution, and a second phase of optically examining with a relatively high spatial resolution only the suspected locations for the presence or absence of a defect therein.

Abstract: An image scanning apparatus efficiently compresses and encodes image data obtained by scanning both sides of an original sheet by using a line memory storing three lines. Image data output from a first scanner and image data output from a second scanner are alternately output from a switching unit on an individual line basis. The line memory stores three consecutive lines of the image data output from the switching unit. The three lines include a current line, an immediately preceding line and a second immediately preceding line immediately preceding the immediately preceding line, the current line including a current pixel being encoded. Reference pixels are extracted from among pixels included in the current line and the second immediately preceding line. A pixel value of the current pixel is encoded by using a predicted pixel value of the current pixel as a parameter so as to output encoded image data corresponding to both sides of the original sheet.

Abstract: An image processing method and device computes the image quality as an evaluation value, determine the image quality based on this evaluation value, and employ a restoration process in accordance with image quality when this image quality is poor.

Abstract: An image processing apparatus and image processing method for performing a smoothing process to a decoded image signal. Edge information of a pixel is detected in response to a spatial change in the pixel value of the image signal and the smoothing process is performed to the image signal based on the edge information. The present invention reduces noise generated in the plain area in a picture without degrading an edge in the picture.

Abstract: A method and apparatus are provided to allow threshold values needed for the desired number of gradations to be set through a simple process when converting image data with numerous values into a low number of gradations. The desired number of gradations is achieved without shading being noticeable even for images having shading. The cumulative frequency distribution of the pixels for each brightness is determined from image data having a brightness range with designated gradations. The average brightness and standard deviation of the portion that is primarily background is determined from this cumulative frequency distribution. The first threshold value is determined on the basis of this average brightness and standard deviation, the second through n-1st threshold values corresponding to the meaningful information are determined on the basis of this first threshold value. An n value conversion process is accomplished using these threshold values.

Abstract: A dithering device converts a multi-gradation image into a binary image by comparing, in sequence, values of pixels forming the multi-gradation image with threshold values of corresponding elements forming a dither matrix. For determining the element corresponding to the pixel, the dithering device uses an array representing a relationship between a first group of positions of the elements to be used currently and a second group of positions of the elements to be used next. By referring to the array, the dithering device determines the position of the element from the second group based on the position of the element in the first group so as to compare the next pixel with the threshold value of the determined element.

Abstract: A free-lying cell classifier. An automated microscope system comprising a computer and high speed processing field of view processors identifies free-lying cells. An image of a biological specimen is obtained and the image is segmented to create a set of binary masks. The binary masks are used by a feature calculator to compute the features that characterize objects of interest including free-lying cells, artifacts and other biological objects. The objects are classified to identify their type, their normality or abnormality or their identification as an artifact. The results are summarized and reported. A stain evaluation of the slide is performed as well as a typicality evaluation. The robustness of the measurement is also quantified as a classification confidence value. The free-lying cell evaluation is used by an automated cytology system to classify a biological specimen slide.

Abstract: An image processing method comprises the steps of setting a trash image candidate by defining the set of locations of the quadratic differential value of the lightness component of pixel signals, which changes from the value exceeding a positive first threshold value to the value less than a negative second threshold value for the adjacent pixels in a digital image, as the contours of such trash image candidate, extracting the trash area by defining it from among the trash area candidates having the changes of lightness components and saturation components between inner adjacent pixels within a specific range, and interpolating the pixel signals in the trash area in order of larger pixel numbers having effective pixels by use of the signals of effective circumferential pixels. Here, the pixels whose interpolation has been executed are used as new effective pixels, thus making it possible to interpolate the pixels in the trash area from the outer side to the inner side one after another.

Abstract: A method and apparatus for creating threshold arrays having a minimized gray scale variance without geometric artifacts. The invention comprises the steps of processing each candidate pixel in the threshold array using a variance minimization technique and applying a sparse image calculation on the same pixel. The preferred embodiment of the invention comprises the application of a variance minimization technique on a threshold array. The threshold array is analyzed on a pixel-by-pixel basis. Each candidate pixel that is to be turned on is examined in relation to the previous pixels that already exist. The candidate pixel is examined to determine its effect on the variance of the smoothed gray level determined by the convolution of the pixel pattern with a smoothing kernel. The candidate pixel is also examined in relation to the distribution of the angles it forms with the previously selected pixels. The preferred embodiment uses a variant of the combinatorial Hough transform.

Abstract: Analog image data a SEM are converted into digital data, and are processed by a spatial filtering processing, histogram processing, threshold value setting, three-valued image data processing, noise reduction and the like. Area of a pattern in the three-valued image data is calculated by a labelling and calculation processing, and a pattern is sequentially detected by comparing the area of the pattern with a reference area value. The comparison and detection of the same or similar patterns repeated in the SEM image are performed by using the area of the pattern, and are not performed by a shape of the pattern, thereby resulting a precise detection at high speed by using a microprocessor. Since it is possible to perform a pattern recognition from the area value even though the pattern does not have a characteristic, it is possible to precisely detect and recognize a pattern image in high speed.

Abstract: A quantizer controller identifies a target bit value equal to a total number of bits available for encoding a frame. A total amount of distortion in the frame is modeled according to predicted quantization values for each one of the blocks. The predicted quantization values are characterized according to an amount of energy in each block and a number of bits available for encoding each block. Quantization values for each block are optimized by minimizing the modeled distortion in the frame subject to the constraint that the total number of bits for encoding the frame is equal to the target bit value. Each block is then encoded with the optimized quantization value.

Abstract: An input density for a subject pixel position (x, y) is retrieved in S230. A modified density value I'(x, y) is calculated in S240 based on the input density value I(x, y) and the error e(x, y) stored in the corresponding error buffer. Then, in S250, a discrimination value for the subject pixel position (x, y) is retrieved from the bilevel threshold value matrix Mat2( ) which is produced through the blue noise characteristic method. One of two threshold values is assigned to the retrieved discrimination value. Thus, a threshold value T is determined for the subject pixel (x, y). Then, an output density value O(x, y) is determined based on the modified density value I'(x, y) and the threshold value T in S260, S270, and S280. Then, an error value E(x,y) is calculated based on the output density value O(x, y) and the modified density value I'(x, y). The error value E(x,y) is distributed to the error buffer 16 in S300 dependently on the error distribution matrix Bat( ).

Abstract: There is disclosed an image processing apparatus such that after completion of one scan of a scanner unit, a controller obtains a background value in a range to be read by one scan from an image processing IC and compares the background value with a binarization threshold value used in the present scan, and when it is equal to or less than a predetermined value, the background value is used in the next scan, and when it is equal to or larger than the predetermined value, by correcting the threshold value, an image can be read at a high quality.

Abstract: An image processing and reproducing apparatus comprises an image signal output device for outputting an image signal representing an image, a plurality of image displaying devices, each of which reproduces a visible image from the received image signal and displays the visible image, and a transmission line for transmitting the image signal, which has been outputted from the image signal output device, to the plurality of the image displaying devices. The image signal output device is provided with a first image processing section for carrying out principal image processing on an original image signal representing an original image. Each of the image displaying devices is provided with a second image processing section for carrying out necessary image processing, which is other than the principal image processing, on an image signal, which has been obtained from the principal image processing. The image processing and image displaying are thereby carried out quickly, and the cost of the apparatus is kept low.

Abstract: Dither thresholds are assigned one after the other to matrix locations in the process of generating a dither matrix used for printer half-toning. The matrix location to be assigned the next threshold is chosen by locating the tightest cluster or largest void in the dot pattern that will result from the gray level with which the threshold being assigned is associated. Measures of cluster tightness for low-range and high-range thresholds are based on the areas of Voronoi partitions associated with respective candidate locations. For mid-range thresholds, a Gaussian-filter output is used as the measure. In both cases, ties between candidate locations are resolved by applying a further criterion, which depends on the candidate locations' proximities to locations assigned thresholds the same as the one being assigned or differing from it by only one.

Abstract: According to an image quality enhancing circuit and a method therefor, an impulse for each of an input luminance and color signals is detected. A trimmed mean of a predetermined-sized window is output when the impulse is detected. Otherwise, the input signals are bypassed. Accordingly, noise-reduced luminance and color signals are output. An enhanced luminance signal is output by receiving the noise-reduced luminance signal in a picture unit and independently equalizing histograms for subimages divided on the basis of the mean value of the received luminance signal. A local contrast defined as a difference between the value of an input sample with respect to the enhanced luminance signal and each value obtained by low-pass filtering the samples in a predetermined-sized window including the input sample, and the input sample value is adaptively weighted according to the detected local contrast, so that a changed luminance signal is output.

Abstract: A visible seal or label containing a serial number is placed in plain view on the product packaging. The visible label contains the serial number as well as a first public key encrypted version of the serial number. A second or hidden label inside of the package has thereon a second a second encrypted version of the serial number made using a second public key. The hidden label may be secured inside of the package out of sight or may be placed on the back of the visible label and therefore viewable through a transparent case when opened or visible when peeled off. The private keys are known only to the manufacturer. Using a corresponding public key provided by the manufacturer, the consumer, law enforcement agent, or customs inspector can verify that the encrypted version matches the serial number. An advantage to this method is that only the manufacturer can produce matching pairs.

Abstract: Pixels of an image are classified (e.g., into foreground and background pixels). In a set of possible values for the pixels, each value is able to serve as a threshold for classifying the pixels. From among the set, a value is selected for the threshold that causes the classification of the pixels to emphasize a characteristic of the image.

Abstract: A multi-tone-level image is converted into an M-tone-level image using M-1 threshold values, where M-1.gtoreq.2. Middle-tone-level pixels in the M-tone-level image are treated as unknown pixels, tone levels other than the middle tone levels are given to the unknown pixels based on the states of pixels in the neighborhood of the unknown pixels, and thus produces an N-tone-level image, where M>N.gtoreq.2. Two threshold values of the M-1 threshold values are used for determine the middle-tone-level pixels, the two threshold values corresponding to the lightest tone level of a first image region and the darkest tone level of a second image region respectively. The first image region has tone levels which are darker than the middle tone levels and the second image region has tone levels which are lighter than the middle tone levels.

Abstract: A method and apparatus implementing dynamic thresholding techniques in processes that use image subtraction. The subtracted image is separated into a positive difference image using a positive image map and a negative difference image using a negative image map. A positive threshold map and a negative threshold map process a Gaussian filtered first image to generate respective positive and negative dynamic threshold images. The positive and negative threshold map mask out irrelevant areas and implement adaptive gain in pixel thresholds proportional to a difference or contrast between foreground or feature of interest and local background. Dynamic thresholding is a subtraction providing a subtracted positive image and a subtracted negative image. The subtracted positive and subtracted negative images are combined to provide a merged image which is binarized. Edge or Interface detection involves edge detection and true peak detection to find the edge peaks in a pre-image and post image.

Abstract: A volumetric data set having a high intensity, dominant structure, near smaller, less intense structures is smoothed and segmented to separate the dominant structure. The dominant structure is masked out to result in the smaller, less intense structures which are displayed. For example, dominant structures such as the blood pool in atria, ventricles, and the aorta of a 3D angiography data set is segmented and removed from the data set using mathematical morphology operations to result in an enhanced data set. The enhanced data set is displayed by conventional means such as a maximum intensity projection (MIP) at different views to yield images of coronary arteries.

Abstract: Document image thresholding using foreground and background is disclosed. A document (16) is scanned and a region (30) and a sub-region (32) of the document is selected. The intensity value of a pixel in the region is compared to an average intensity value of pixels in a foreground cluster and to an average intensity value of pixels in a background cluster (50). The pixel is assigned to the foreground cluster (53) if the intensity value of the pixel is closer to the average intensity of the foreground cluster, or to the background cluster (54) if the pixel intensity is closer to the average intensity value of the background cluster. A new average intensity value is calculated for the foreground cluster (55) and the background cluster (56). A new pixel is compared to the new average intensity values for the foreground and background clusters and the process is repeated until all pixels in the region have been compared (57).

Abstract: A dither method converts multi-bit grey level pixel values into an image array of binary level pixel values, wherein the image array is arranged as a raster of scan lines.

Abstract: A technique which combines the advantages of blue noise and clustering obtains more pleasant "blueish" effects in the rendering of the grey levels for which the original multicell mask generates unwanted periodic representations. Such original multicell clustered dither array patterns which are undesirable are replaced by less periodic ones while preserving all benefits of the classical methods, and in particular, leaving the nicest patterns unchanged. An automated process redistributes the locations of the threshold values corresponding to undesirable grey levels to obtain more pleasant "blueish" effects in the rendering of the grey levels.

Abstract: The present invention provides a method and apparatus of inspecting a surface of an object article capable of precisely detecting strains or miniature defects present in the object surface. Specifically, the invention provides a method of inspecting surface irregularity of an object article having a surface of uniform or regular brightness, which comprises the steps of: gaining brightness information for a plurality of two-dimensionally distributed pixels by taking a picture of the article surface; finding stains on the article surface in response to each information obtained in the brightness information gaining step to produce a first output; finding miniature defects smaller in size than a unit pixel in response to each information obtained in the brightness information gaining step to generate a second output; switching the first output and the second output into appropriate electrical signals in an controlled manner; and displaying the switched electrical signals on a display in a viewable condition.

Abstract: An advanced low cost adaptive frame grabber is described. Its main area of application is in fingerprint identification and verification systems, but it could successfully be used in other biometric identification systems, such as facial, iris or eye identification and/or verification systems. The invented framegrabber is also very suitable for all applications in other areas, like medical electronics, car electronics etc., where automatic image enhancement and/or transformation from grey scale to line art is required. The adaptive framegrabber automatically estimates the dynamic range of the acquired image and sets up a proper offset and measurement range for a video analog to digital converter (ADC). It could also be operated in two different modes, in one the image is acquired as a full grey scale image, and in the other--as a line art (black and white) image. Full controls over the image size and resolution are provided.

Abstract: A dispersed-dot stochastic dither array is provided for rendering halftone images having excellent visual quality and are created by a minimum density variance method. By minimizing the variance in the number of dots within each local region of the image, a smooth and dispersed distribution of dots may be obtained. For color printing, a separate threshold array is generated for each of the color planes, however, the stochastic screens are interlocked so that the threshold arrays are generated while considering the other color threshold arrays. In this manner, a blue noise distribution may be produced by the individual arrays as well as by any combination of the individual arrays. When generating a single threshold array for a color plane, a particular criterion is used to determine where the next threshold value should be located, and the selection of a threshold location in each array considers the criterion for all the threshold arrays being generated.

Abstract: An improved interpolation method in which a threshold value used for determining a pixel value of a pixel generated by interpolation according to a context (state value of adjacent pixels). In the interpolation method, the ambiguity between the interpolation value and the threshold value is removed by using the context (state value of the reference pixels around the interpolated pixel), thereby reducing the blocking and smoothing phenomena in the restored binary image.

Abstract: In case of subjecting input graduation image data of 8 bits per pixel to multilevel dither processing, to convert them into image data having gradients of 3 bits per pixel on the basis of dither threshold planes, in a combination of threshold arrays extending over the plural dither threshold planes, dither thresholds for raising spatial fre0 quency, as shown in FIG. 8A, are arranged in low density areas of the input graduation image data, and dither thresholds for making spatial frequency lower than that in the low density areas, as shown in FIG. 8C, are arranged in half tone areas and high density areas of the input graduation image data.

Abstract: An image data processing apparatus for adding a special effect to an input image data, which is provided with function data generating circuit for generating function data in accordance with positions of pixel data for each pixel data forming the image data, random number generating circuit for generating different random numbers for each pixel data, and output circuit for judging based on the function data and the random number data if the corresponding pixel data is valid data or invalid data, outputting the pixel data only when valid data.

Abstract: An interpolation method in which a threshold value used for determining a pixel value of a pixel generated by interpolation according to a context (state value of adjacent pixels). In the interpolation method, the ambiguity between the interpolation value and the threshold value is removed by using the context (state value of the reference pixels around the interpolated pixel), thereby reducing the blocking and smoothing phenomena in the restored binary image.

Abstract: A method or apparatus for processing an input frame of image data comprising converting the input frame into a plurality of output frames, the spatial frequency spectrum of each of said plurality of output frames being that of a respective predetermined frequency band of the spatial frequency spectrum of the input frame and applying separately to at least one of said plurality of output frames a respective gain.

Abstract: In this proposal a method of multilevel halftoning is described that allows the definition of halftone patterns using more than two gray levels at a time. This feature can be employed to eliminate the texture contours produced when shades formed by a pattern of two levels blends into a shade formed by a single gray level. The method also gives control over the degree to which a given output gray level is used. This can be employed to limit ink coverage when certain levels require more ink to produce than do others. According to the method: initial pixel intensity ranges are obtained; the initial pixel intensity ranges are compared to at least two overlapping threshold arrays; and selection of enhanced pixel intensity levels are determined for halftone cells according to the comparison results of the initial pixel intensity ranges to at least two overlapping threshold arrays wherein the enhanced pixel intensity levels are a function of results from the comparison.

Abstract: An improved interpolation method in which a threshold value used for determining a pixel value of a pixel generated by interpolation according to a context which is a state value of adjacent pixels. In the interpolation method, the ambiguity between the interpolation value and the threshold value is removed by using the context, thereby reducing the blocking and smoothing phenomena in the restored binary image.

Abstract: An improved interpolation method in which a threshold value used for determining a pixel value of a pixel generated by interpolation according to a context which is a state value of adjacent pixels. In the interpolation method, the ambiguity between the interpolation value and the threshold value is removed by using the context, thereby reducing the blocking and smoothing phenomena in the restored binary image.

Abstract: An image reading apparatus suitable for use as a facsimile apparatus, a coping apparatus or the like, and a method for detecting open/close of an original feeding unit in the image reading apparatus.

Abstract: A method and system for detecting and displaying clustered microcalcifications in a digital mammogram, wherein a single digital mammogram is first automatically cropped to a breast area sub-image which is then processed by means of an optimized Difference of Gaussians filter to enhance the appearance of potential microcalcifications in the sub-image. The potential microcalcifications are thresholded, clusters are detected, features are computed for the detected clusters, and the clusters are classified as either suspicious or not suspicious by means of a neural network. Thresholding is preferably by sloping local thresholding, but may also be performed by global and dual-local thresholding. The locations in the original digital mammogram of the suspicious detected clustered microcalcifications are indicated. Parameters for use in the detection and thresholding portions of the system are computer-optimized by means of a genetic algorithm.

Abstract: Embodiments of the invention include a method and apparatus for determining values for data reproduced or retrieved from data storage media such as in holographic memory systems. The method includes initially approximating the value of data members retrieved initially from the data storage medium of interest (e.g., by their intensity levels based on an absolute scale), using those approximations collectively to establish one or more threshold levels for defining ranges for the data values to be determined, and determining values for the retrieved data members by comparing the initial data value approximations of the retrieved data members to the established threshold levels. Alternatively, the threshold level(s) are set initially and adjusted iteratively based on the initial approximations of the individual data members until a final established threshold level is established.

Abstract: A dither matrix is produced by creating an array of dots and dividing the array into a mutiplicity of regions, each of the regions having a multiplicity of dots. Borders of the regions are altered to have substantially continuous randomly irregular boundaries, and dither values are assigned to each of the regions.

Abstract: A method and apparatus for inspecting the surface of articles, such as chips and wafers, for defects, includes a first phase of optically examining the complete surface of the article inspected at a relatively high speed and with a relatively low spatial resolution, and a second phase of optically examining with a relatively high spatial resolution only the suspected locations for the presence or absence of a defect therein.

Abstract: A method for determining whether a pixel of a digitized image is a background or object pixel includes the steps of determining whether the pixel is in a static state or a transient state according to the variation in its gray-level compared with the gray-level of an adjacent pixel located in a first direction, setting a static threshold if the pixel is in the static state or a dynamic threshold if the pixel is in the transient state, comparing the gray-level of the pixel and the applicable threshold to determine whether the pixel is a background or object pixel, and repeating the determination with respect to static or dynamic thresholds, as applicable, in a second direction.

Abstract: The present invention describes a method for automatic cropping of images containing regions where intensity levels are uniform and other regions where intensity levels vary considerably. An image to be automatically cropped is scaled down to a grid and divided into non-overlapping blocks. The mean and variance of intensity levels are calculated for each block. Based on the distribution of variances in the blocks, a threshold is selected for the variance. All blocks with a variance higher than this threshold variance are selected as regions of interest. The regions of interest are then cropped to a bounding rectangle.

Abstract: An idealized clustered screen is characterized spot clustering which is optimized so that clusters are about the same size and shape as they grow and are approximately equally distanced though an image. Pixels are swapped within the screen to vary its appearance. An optimization process iteratively checks a quality metric based on the optimum conditions, as pixels are swapped until an optimum screen appearance is obtained.

Abstract: An image processing method for determining a characteristic of image data in each area in whole image area and performing binarization processing appropriate for the characteristic, the method comprises the steps of: calculating gradation levels around a pixel in the image data; selecting a threshold value matrix appropriate for the pixel, in response to the calculating step, from the group consisting of a threshold value matrix consisting of single threshold value elements for simple binarization, a threshold value matrix comprising different threshold value elements for halftone processing, and at least one threshold value matrix comprising interpolation of both the threshold value matrixes; and binarizing the pixel by a systematic dither method using the threshold value matrix which is selected in the selecting step.

Abstract: This system and method error-diffuse image data to generate printing decisions; scan and print stages are preferably included. Multiple image-data rows are held in longterm memory, usually external to an integrated circuit that does the processing; preferably data from plural rows at a time are received from that memory. Error from earlier-processed data is also held in longterm memory and fed to the circuit. Longterm-memory access is shared with many other functions; the invention keeps error diffusion from dominating that access. Processing preferably proceeds generally transverse (preferably skewed, ideally at 45.degree.) to data rows, with final error distribution into each pixel when it is the "left pixel" (or when it is the "next pixel" if it is at the top of a row group).

Abstract: An improved display of a color image is achieved by an optimum placement of color dots using an efficient stochastic screening process in which display elements are formed according to a priority ranking of color and according to a spatial-priority matrix. Various combinations of basic colors are ranked in priority according to visibility such that elements in a stochastic screen are first formed for the most visible colors. The number of elements in the screen that are used to represent more than one basic color is minimized. In preferred embodiments, a spatial-priority matrix is used to control the placement of colors in a stochastic screen such that an optimally uniform distribution of colors is achieved. A process for generating a spatial priority matrix for display devices that do not have a one-to-one display aspect ratio is described.

Abstract: Image data scanned by a scanner in a digital scanner, a digital copying machine or a facsimile machine is processed to detect an edge despite low contrast. A first MTF compensation portion enhances an edge area of the image data. An edge detector detects edge data from the first PTF co pensation portion. A second MTF compensation portion enhances the edge data from the equalization filter. An equalization filter calculates a binary cut-off level for input data. Another edge detector functions as a device for detecting edge data from the equalization filter by a Laplacian filter. Clamp circuits clamp a highest limit and a lowest limit of the binary cut-off level of signals from the equalization filter. A first comparator compares an output signal from the MTF compensation portion to an output signal from one of the clamp circuits. A second comparator compares an output signal from the first equalization filter to an output signal from the other clamp circuit.

Abstract: A plurality of uniform density pixel matrices Di are prepared so that each matrix Di will have all pixels of a uniform density i. First, in S110, the pixels of one uniform density pixel matrix Di are converted into binary values (0 or 1) while performing an error diffusion operation. Thus obtained binary value pixel matrix Fi is stored in the working memory 14. Then, while the uniform density value i is repeatedly incremented, the above-described converting-and-storing processes are repeated. Afterward, binary values throughout all the binary value pixel matrices are summed at each pixel position. Thus, in S150, an accumulated value matrix M1 is produced to have a corresponding element constructed from the total value. Next, threshold values of a dither matrix DM are determined based on the elements of the matrix M1. That is, the threshold values are determined one by one using the elements of the matrix M1 from its element having the lowest value.

Abstract: A method and apparatus for separating picture image areas in facsimile machines, scanners, and other picture processing equipment, and more particularly to a method and apparatus for separating picture regions by extracting features of graphic image and characters includes a vertical contouring circuit for producing vertical contour data by combining vertical components of an original picture and binary picture data; a horizontal contouring circuit for producing horizontal contour data from horizontal components of the binary picture data; a logic combiner for combining outputs from the vertical and horizontal contouring circuits and a halftone picture; a mode discriminator for selecting between a binary picture and contouring mode; an image area separator for separating image areas with an output from the logic combiner and a halftone picture, and a counter for executing pixel counting from a halftone picture to determine the probability of the presence of a character.

Abstract: A binarizing section generates binarized data from an image signal of a dot code on an information recording medium read by a code reading section. The binarizing section has a reference dot detection section, a dot area measuring section, a threshold value modifying section and a threshold value determining section. The reference dot detection section binarizes the image signal with a predetermined threshold value prior to generating binarized data to detect a reference dot from a binarized code image. The dot area measuring section measures the area of the reference dot detected by the reference dot detection section. The threshold value modifying section modifies the threshold value for binarization in such a manner that the area measured by the dot area measuring section approaches a predetermined target value. The threshold value determining section binarizes the image signal with the threshold value modified by the threshold value modifying section.