Abstract: Features characterizing at least two significant portions or regions of the printed pattern on a model document are extracted from a scanned-in image of the model document. Statistical characteristics of each significant region are calculated from the extracted features and saved in a memory. In addition, geometrical relationships, e.g. distances and angles, between the regions are also saved in a memory. The geometrical relationships are represented by the coordinates of the regions in a predetermined coordinate system, e.g. the x-y coordinate system corresponding to the scan area on the platen of a scanner. When a new or trial document is scanned and examined through a search window, features are extracted from the search window blocks and compared to the statistical characteristics of the model document previously stored in memory. If there is a match, the coordinates of the search window having the matching features in the trial document are recorded in a candidate list.

Abstract: The present invention provides a method and apparatus for classifying image data. A video average may be determined from the image data. Peak detection may then be limited to an area within the image data where the video average is above a first threshold. Valley detection may be limited to an area where the video average is below a second threshold.

Abstract: A method of and apparatus for detecting objects in computed tomography (CT) data includes the ability to define the types of objects to be detected, and at least one algorithm related to the detection of each type of object. Multiple types of objects can be detected and distinguished from one another. Each type of object exhibits an object detection rate related to the probability of the system detecting the corresponding object type, and a false detection rate related to the false identification of objects, different from the target objects, as the target objects. An overall system detection rate is related to a combination of the object detection rates. Each type of object can also be associated with a unique object false alarm rate, with a overall false detection rate being related to the combination of object false alarm rates.

Abstract: A color image data compressing apparatus comprising a color space converting section for receiving RGB data and outputting data for lightness and chromaticity, a plurality of space converting sections each for converting the data from actual space to a coefficient in each specified transform space, a plurality of quantizing sections each for quantizing the coefficient, and a coding section for coding the quantized data, in which the degree of quantization of the coefficient data is changed according to data for lightness.

Abstract: A system for automatic detection and retrieval of embedded invisible digital watermarks retrieves digital watermarks from halftone images. Specifically, by supplying an image to the system, through a process of autocorrelation and shifting, the embedded invisible watermark becomes visible. The process includes scarning or supplying an image to the system, calculating the global autocorrelation of the image, selecting a moving window size, conducting a piecewise localized autocorrelation for each window-sized portion of the image, retrieving the embedded, initially invisible, watermarks, normalizing the resultant image for visualization and displaying the resultant image with the now visible retrieved embedded digital watermarks.

Abstract: A message in the form of a sequence of bits is encoded into relatively short, non-binary signals. The message (or vector) is divided into two halves. The encoder first determines an encoding of the message into a vector half the length of the eventual encoded vector. The half-length vector forms the first half of the encoded, and is negated to form the second half of the encoded vector. An identical synchronization signal is added to both halves of the encoded vector. The synchronization signal is used to compensate for the possibility that the received vector (or message) is cyclically rotated from the original vector. During decoding, the decoder first obtains the synchronization signal by adding the two halves of the vector together, canceling the encoded message. The synchronization signal is used to find the original cyclical shift of the vector. The decoder obtains the half-length encoded message vector by taking the difference between the two halves of the aligned full-length encoded vector.

Abstract: A method, apparatus, and article of manufacture employing lexicon reduction using key characters and a neural network, for recognizing a line of cursive text. Unambiguous parts of a cursive image, referred to as “key characters,” are identified. If the level of confidence that a segment of a line of cursive text is a particular character is higher than a threshold, and is also sufficiently higher than the level of confidence of neighboring segments, then the character is designated as a key character candidate. Key character candidates are then screened using geometric information. The key character candidates that pass the screening are designated key characters. Two-stages of lexicon reduction are employed. The first stage of lexicon reduction uses a neural network to estimate a lower bound and an upper bound of the number of characters in a line of cursive text. Lexicon entries having a total number of characters outside of the bounds are eliminated.

Abstract: A method of reconstructing an image captured as a stream of image data, for example as input received from a linear sensor in unconstrained scanning, comprises reconstructing the image in the form of a plurality of tiles. Each tile comprises a pixel grid of predetermined dimension representing a specific spatial region of the image. The tiles tessellate a rectilinear image space. Tiles can be created when required and compressed when no longer active, thus minimizing memory requirements. Devices utilizing this method are provided. The method is especially appropriate for use in an unconstrained hand scanner, but can also be applied to panoramic capture with a digital camera.

Abstract: This invention solves problems due to employing error degraded data in digital processing. It particularly solves the multi-generation problem wherein transform data degrade during each inverse transform and forward transform cycle even without any processing due to the rounding and clipping errors. It provides methods, systems and apparatus for transform-domain correction of real-domain errors in the processing of transform, transform-coded, and coded data. After inverse transformation of transform data the high-precision numbers are converted to integers and clipped to an allowed range forming converted data. High-precision differences are obtained by subtracting converted data from the high-precision output of the inverse transform. The differences are re-transformed back to the transform domain and saved. In an alternate embodiment the differences are quantized, thus, only significant errors are saved. In still another embodiment the quantized differences are entropy encoded.

Abstract: A method detects if an image is compressed. The method determines a block grid within the image and establishes blocks from the determined grid. The method then computes differences between samples inside the established blocks and differences between samples across the established blocks. The method determines that the image is compressed based on characteristics derived from statistics of the computed differences.

Abstract: A facsimile network includes a plurality of nodes (10)-(14) which are disposed at different locations in the world. An originating fax machine (16) is associated with one of the nodes (10) for transmitting a fax over a local public telephone network (PSTN) (18). When the node (10) receives the fax transmission in the normal transmission mode, it then converts it to a digitized compressed and encrypted image which is transmitted over a primary network (24), such as the Internet, to a destination node (14) by a predetermined route. The node (14) then decompresses and de-encrypts the image and converts it to a facsimile image for transmission to a destination fax (20) in a normal mode over a local PSTN (22). A node (12) operates as a central node which can also receive the transmission for routing directly to the fax or for directing via another network link to the node (14).

Abstract: A reliable method of illuminating and imaging an eye through eyeglasses uses a carefully selected subset of multiple monochromatic light sources, a camera with an imager that exhibits minimal blooming, and a narrow-bandwidth optical bandpass filter to filter out most of the ambient illumination while passing most of the light from the system's own illuminator.

Abstract: An adaptive OCR technique for character classification and recognition without the input and use of ground truth derived from the image itself. A set of so-called stop words are employed for classifying symbols, e.g., characters, from any image. The stop words are identified independent of any particular image and are used for classification purposes across any set of images of the same language, e.g., English. Advantageously, an adaptive OCR method is realized without the requirement of the selection and inputting of ground truth from each individual image to be recognized.

Abstract: An image processing apparatus obtains a sum A of image data values of pixels in a template image, a sum B of squares of image data values of pixels in a template image, a sum C of image data values of pixels in a sub-image to be processed, of a search image, a sum D of squares of image data values of pixels in the sub-image of the template image, further obtains a threshold value F in advance by using the obtained values A, B, C and D, the number P of pixels in the template image, and the preset value E. Moreover, the apparatus obtains a square of each difference between an image data value of each pixel in the sub-image and that of a corresponding pixel in the template image, and performs cumulative addition for each obtained squares. If the result of cumulative addition exceeds the above-mentioned threshold value, the apparatus closes processing evaluation of a similarity between the sub-image and the template image.

Abstract: When edit processing is to be performed for a plurality of originals to form visible images, all the originals are edited at once, and the resultant images are output. A DH (130) sequentially conveys originals to a read surface. A CCD (105) reads original images. When nine original images are completely read, the nine read original images are displayed on a CRT (131) to allow the user to perform edit processing for a desired original. The edit contents are stored in a predetermined storage device. When edit processing for the displayed original images is complete, the next nine originals are processed in the same manner as described above. When a series of edit operations are completed in this manner, and a copy instruction is given, a read operation is performed again from the first original, and each original is edited in accordance with the stored edit contents. The resultant images are then printed out.

Abstract: A method for determining the color of an illuminant and an apparatus therefor are provided.

Abstract: A particle analyzer includes a detection section to detect a parameter representative of characteristics of particles in a sample, a processing section to process the detected parameter, and an output section to output a result obtained by the processing section, wherein the processing section includes a distribution diagram producing section to produce a distribution diagram based on the parameter detected by the detection section, an area storing section to store a predetermined reference pattern beforehand, and a control section to permit the output section to output the reference pattern in addition to the distribution diagram.

Abstract: A method of producing a transform decomposition of data having an odd length the method comprising the steps of dividing the data into a portion having an even length of one element; performing a discrete wavelet transform on the even length data to produce low frequency subband data and high frequency subband data; adding the difference of the one element and an adjacent element to high frequency subband data. Preferably the transform is a Discrete Wavelet Transform utilised in the compression of image data.

Abstract: According to a method for improving digital images, an input image is digitized and represented by pixels, each pixel including three original color components represented by gray level values. The average gray level value for a first color is calculated to obtain a first color adjustment factor. A second color adjustment factor and a third color adjustment factor are calculated based on the first color adjustment factor, and the gray level values for the second color and the third color of all pixels are adjusted accordingly to obtain an image with emphasized first color components.

Abstract: A template is analyzed to determine step sizes for searching within a search area. The template is analyzed by first padding the template with data points to increase its size. Cross-correlation between the padded template and the original template leads to identification of an effective step size along multiple axes. Step sizes for each of a horizontal, vertical and a third axis are derived. Third axis step sizes may correspond to rotation, scaling factor, subsampling factor, linear distance, time or frequency. Windows of the search area, selected based on the step sizes, then are tested in a fast search by correlating the template to selected windows to derive correlation coefficients. Any tested window which has a correlation coefficient exceeding a given value is a potential match for the template and is subject to a refined stage of comparison.