Abstract: A method an apparatus for detecting lines in medical images is disclosed, wherein a direction image array and a line image array are formed by filtering a digital image with a single-peaked filter, convolving the resultant array with second order difference operators oriented along the horizontal, vertical, and diagonal axes, and computing the direction image arrays and line image arrays as direct scalar functions of the results of the second order difference operations. Advantageously, line detection based on the use of four line operator functions along the horizontal, vertical, and diagonal directions in accordance with the preferred embodiments actually results in fewer computations than line detection based on the use of three line operator functions. In particular, because of the special symmetries involved, 3×3 second order difference operators may be effectively used.

Abstract: A method of automatically evaluating settings for three dimensional rendering engines establishes a mathematical equivalency to image quality such as may be RMS error. With this mathematical index, a gold standard setting may be established and used to compare clinical images with other settings. The image qualities for different settings may be fit to a functional model so as to provide rapid assessment of the affect on image quality of parameter adjustments. The time-cost for producing the image may also be fit to a surface to allow more complex time/image quality trade offs to be realized.

Abstract: An image composite processing method for composing a plurality of images, each having a duplicate portion, into a single composite image based on a retrieved duplicate portion, wherein the duplicate areas of every two adjacent composite subject images are compared and one of the duplicate areas is adopted based on the comparison result.

Abstract: This invention solves problems due to employing error degraded data in digital processing. It particularly solves multi-generation problems 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 devices for reduced-error processing of transform-coded data. After inverse transformation of transform data, high-precision numbers are converted to integers and clipped to an allowed range forming converted data. High-precision differences are obtained by subtracting the high-precision output of the inverse transform from the converted data. The converted data can be manipulated and sent to output devices which expect integer data. Processed high-precision numbers are formed by adding the high-precision differences to the processed converted data. Thus, the rounding and clipping errors are greatly reduced in the processed high-precision numbers.

Abstract: Detected data for a plurality of predetermined positions on a particular bill to be identified are each converted into relative-value data to a minimum value of the detected data for the individual positions, to extract a characteristic pattern of designs printed on the bill. The relative-value data are normalized by use of a relative value standard average to provide adjusted sample data. Then, for each of the positions, a distance value indicating how many times a difference of the adjusted sample data from the standard average is greater than a standard deviation (namely, standardization variable) is evaluated, to effect further normalization. The bill is then identified by judging whether or not the thus-normalized adjusted sample data, i.e., distance value for each of the positions, falls within a predetermined determination-standard-value range.

Abstract: A system for automatically generating a mosaic from a plurality of input images. The system sequentially executes an image alignment process, an editing process, and a combining process such that, from a sequence of images, the system automatically produces a seamless mosaic for various applications. Illustrative applications for the mosaic include: (1) a mosaic based display system including an image printing system, (2) a surveillance system and (3) a mosaic based compression system. The mosaic based display system permits a system user to display, manipulate and alter a mosaic. The mosaic based compression system exploits the temporal and spatial redundancy in image sequences and efficiently compresses the image information. The compression system can be used for compressing image information for storage in a storage device or can be used for compressing image information for transmission through a band-limited transmission channel.

Abstract: The present invention provides a method and apparatus for classifying image data. A video peak/valley counter may count peaks and valleys within a window of the input image data. The apparatus and method checks whether any peaks or valleys are located within a neighborhood of a pixel under consideration.

Abstract: An on-line handwritten Chinese character recognition apparatus based on character shapes uses a reference file for look-up of compressed codes-sequence codes in a conventional input method founded on dismantling by character shapes as reference when comparing character to reduce the number of matching templates required in the on-line handwritten Chinese character recognition system and to lower the time for matching. Input handwritten Chinese characters are dismantled into constituting radicals. Thereafter, with reference to code retrieving rules of the input method, and based on the character composing method of handwritten characters, radicals for comparison are retrieved. Then, the compression formula of the input method is used to compress the codes of the base radicals, and the compressed codes are compared with contents of the input method reference file to obtain the sequence codes of candidate characters.

Abstract: In a method for interpolating color values with a view to converting an initial color space into an output color space, the initial color space is divided into a multidimensional network. For grid points, converted components of the output color space are prescribed as interpolation nodes. The converted components, located between the grid points are obtained by interpolating said nodes. For the purpose of interpolation, the components of an initial color to be interpolated are broken down in a multiple of the grid point spacing and a remainder, and the remaining values are classified by order of magnitude. At the grid points of a sub-space comprising the initial color to be interpolated, the interpolation nodes are selected in a series determined by the classification of the remaining values, and the initial color is interpolated based on the selection of interpolation nodes.

Abstract: A method for encoding digital information to be transmitted from a first device to a second device, wherein the digital information is mapped according to a map of n rows and m columns which define a plurality of cells (n×m) and wherein each of said cells has a pixel value. The method is useful for compressing bitmapped images to be rendered by the simplest of graphics engines. Thus the method is particularly suited to the transmission of images to a relatively slow client on which the images are to be rendered in real time and on which any complex decompression would be too costly for such rendering. Raw compression ratios are comparable to those obtained with standard runlength encoding, but the number of drawing primitive calls required to render the compressed images directly is substantially less. If the size of the largest area exceeds predetermined limits it is subdivided into smaller areas automatically without having to specify the size and position.

Abstract: Systems and methods for performing intelligent image processing. Image processing systems and methods in accordance with the present invention may select algorithms for processing collections of images by comparing algorithmic region of interest (aROI) data to stored human visual region of interest (hROI) data to select from a database of available transformation algorithms an optimal algorithm or group of algorithms to be used in transforming data comprising the collection or collections of images. The selected algorithm(s) may then be used, for example, in data compression, image enhancement or database query functions.

Abstract: A method and apparatus for selecting colors for blocks for use in a Block truncation Coding scheme is disclosed. The method includes finding the largest cluster of pixels and selecting that color for the first color in the block. The average color of the remaining pixels is assigned to the second color. The invention may be applied in iterative fashion if more than two colors are to be assigned to the block.

Abstract: A histogram-based segmentation of images in a video signal via color moments is initialized by a user defining regions in objects of interest from one or more images, key frames or pictures of the video signal. For each rectangle a normalized average color moment and associated co-variance matrix are determined which define a color class for that rectangle. From the normalized average color moment and associated co-variance garbage parameters are generated. Segmentation is then performed on a block basis on each image of the video sequence, a normalized color moment being generated for each block. Using a log likelihood test the closest color class for the block is determined. Based upon the closest color class and the garbage parameters for that color class a final determination is made in a two stage test as to whether the block belongs to the closest class or to a “garbage” class.

Abstract: An image data processing apparatus for extracting a basic texture pattern as an identifiable feature of an image so that accurate image search can be performed by comparing images using the extracted basic texture patterns. The image data processing apparatus inputs image, detects textural periodicity in the image, extracts the basic texture pattern based on the calculated periodicity, normalizes the extracted basic pattern and shifts the normalized basic pattern. The apparatus stores the normalized and shifted basic pattern to an image database with the image from which the basic pattern was extracted.

Abstract: A method and apparatus for manipulating digitized images stored as variable length encoded bitstreams such as JPEG format in a manner that reduces memory and processor resource requirements. A prescan means sequentially decompresses the bitstream to identify the location of encoded pixel image areas. Designated ones of these locations are recorded or stored in a prescan table. After the prescan operation has been performed on the image, image manipulations such as rotating, cropping, and zooming can be performed on a selected portion of the image by directly accessing only the encoded pixel image areas to be manipulated, without the need to sequentially decode and store all the encoded image areas in order to locate the ones of interest.

Abstract: A three-dimensional model of a scene is reconstructed from a plurality of images of the scene taken from different viewpoints. A set of voxels that include the scene is defined, and each of the voxels in the set of voxels is processed. Each voxel is projected onto a set of pixels in each of the images. A first color-standard-deviation of not-occluded pixels is determined across all images. A second color-standard-deviation of not-occluded pixels is determined for each individual image. A mean of the second color-standard-deviations is determined across all images. A coloring threshold that is a function of the mean is established. If the first color-standard-deviation is less than the coloring threshold, the voxel is colored. Otherwise, the voxel is left transparent. The coloring threshold is established adaptively and depends on the color variation across the projection of the voxel in each individual image.

Abstract: An improved technique for compressing a color or gray scale pixel map representing a document using an MRC format includes a method of segmenting an original pixel map into two planes, and then compressing the data or each plane in an efficient manner is disclosed. The image is segmented by separating the image into two portions at the edges. One plane contains image data for the dark sides of the edges, while image data for the bright sides of the edges and the smooth portions of the image are placed on the other plane. A thresholding technique and apparatus for generating a bitmap is also disclosed that may be used to accomplish these tasks is also disclosed.

Abstract: The method of the present invention discovers patterns in a protein sequences in two phases. In a sampling phase, preferably proper templates corresponding to a group of protein sequences are generated. Patterns corresponding to the templates are then generated and stored in memory. In a convolution phase, the patterns stored in memory are combined to identify a set of maximal patterns.

Abstract: Apparatus and method for improving recognition of patterns such as alphanumeric characters. A known recognition system is expanded to further include a complementary recognition system which is linked with the primary recognition system. An image that can not be positively recognized by the primary recognition system is passed on to the complementary recognition system and any characters not positively recognized by the complementary recognition are again passed on to a correction system. At the correction system, an operator classifies unrecognized characters which are then used to teach the complementary recognition system. Thus, the classified data of the correction system provide the training data for a continuous training process which is coupled with the correction system by a pattern adaptation system.

Abstract: The architecture of the inventive correlator is, in a preferred embodiment, an array of correlation cells each containing a delay pipe, a math unit and an accumulator. An array of these correlation cells are tiled together to allow simultaneous processing by all cells. The array is disposed so that each cell accumulates an output value in a result surface. There is no electrical limit to the number of correlation cells that may be tiled together. A preferred embodiment uses nine cells tiled together into a 3×3 correlation result surface. Other embodiments have been tested in accordance with the present invention having twenty-five cells tiled together into a 5×5 correlation result surface. A stream of compare pixel values is presented to the array wherein each compare pixel value is presented to each cell concurrently. A reference memory supplies the appropriate reference pixel values to the cells to enable all calculations for that compare pixel value to be done concurrently.