Abstract: A system for electronically distilling information from a business document uses a network scanner to electronically scan a platen area, having a business document thereon, to create a bitmap. A network server carries out a segmentation process to segment the scan generated bitmap into a bitmap object, the bitmap object corresponding to the scanned business document; a bitmap to text conversion process to convert the bitmap object into a block of text; a semantic recognition process to generate a structured representation of semantic entities corresponding to the scanned business document; and a document generation process to convert the structured representation into a structure text file.

Abstract: A method for predicting a processing result includes a process of performing a principal component analysis on a plurality of detected data obtained during a first standard processing, to construct a principal component analysis model; a process of obtaining residuals of the principal component analysis model as first residuals; a process of performing a second standard processing; and a process of obtaining a plurality of detected data from a plurality of detectors during the second standard processing. And, the detected data obtained during the second standard processing are applied to the principal component analysis model, to obtain second residuals. The method further includes a process of weighting the second residuals based on weighting references, and constructing a new multivariate analysis model with use of the weighted second residuals; and a process of predicting a processing result of the second standard processing with use of the multivariate analysis model.

Abstract: Techniques for efficiently and accurately organizing freeform handwriting into lines. A global cost function is employed to find the simplest partitioning of electronic ink strokes into line groups that also maximize the “goodness” of the resulting lines and the consistency of their configuration. The “goodness” of a line may be based upon its linear regression error and the horizontal and vertical compactness of the strokes making up the line. The line consistency configuration for a grouping of strokes is measured by the angle difference between neighboring groups. The global cost function also takes into account the complexity of the stroke partitioning, measured by the number of lines into which the strokes are grouped. An initial grouping of strokes is made, and the cost for this initial grouping is determined. Alternate groupings of the initial stroke grouping are then generated.

Abstract: System and method for analyzing an image. A received image, comprising an object or objects, is optionally preprocessed. Invariant shape features of the object(s) are extracted using a generalized invariant feature descriptor. The generalized invariant feature descriptor may comprise a generalized invariant feature vector comprising components corresponding to attributes of each object, e.g., related to circularity, elongation, perimeter-ratio-based convexity, area-ratio-based convexity, hole-perimeter-ratio, hole-area-ratio, and/or functions of Hu Moment 1 and/or Hu Moment 2. Non-invariant features, e.g., scale and reflection, may be extracted to form corresponding feature vectors.

Abstract: The subject invention leverages a scalable character glyph hash table to provide an efficient means to identify print characters where the character glyphs are identical over independent presentation. The hash table allows for quick determinations of glyph meta data as, for example, a pre-filter to traditional OCR techniques. The hash table can be trained for a particular environment, user, language, character set (e.g., alphabet), document type, and/or specific document and the like. This permits substantial flexibility and increases in speed in identifying unknown glyphs. The hash table itself can be composed of single or multiple tables that have a specific optimization purpose. In one instance of the subject invention, traditional OCR techniques can be utilized to update the hash tables as needed based on glyph frequency. This keeps the hash tables from growing by limiting updates that reduce its performance, while adding frequently determined glyphs to increase the pre-filter performance.

Abstract: Column misalignment between a character string recognized by a MICR and a character string recognized by optical character recognition is corrected to improve character recognition reliability. A magnetic data reading unit detects the magnetism of a character string printed in magnetic ink and outputs the detected magnetic data. An image reading unit optically reads the same character string printed in magnetic ink and outputs the image data. A magnetic data recognition unit applies a character recognition process to the magnetic data output by the magnetic data reading unit and outputs magnetic data recognition result. An image data recognition unit applies character recognition to the image data output by the image reading unit and outputs image data recognition result.

Abstract: A method and system for generating 3D images of faces from 2D images, for generating 2D images of the faces at different image conditions from the 3D images, and for recognizing a 2D image of a target face based on the generated 2D images is provided. The recognition system provides a 3D model of a face that includes a 3D image of a standard face under a standard image condition and parameters indicating variations of an individual face from the standard face. To generate the 3D image of a face, the recognition system inputs a 2D image of the face under a standard image condition. The recognition system then calculates parameters that map the points of the 2D image to the corresponding points of a 2D image of the standard face. The recognition system uses these parameters with the 3D model to generate 3D images of the face at different image conditions.

Abstract: Systems and methods are described that facilitate performing feature extraction across multiple received input features to reduce computational overhead associated with feature processing related to, for instance, optical character recognition. Input feature information can be unfolded and concatenated to generate an aggregated input matrix, which can be convolved with a kernel matrix to produce output feature information for multiple output features concurrently.

Abstract: A system and method for improved validation of images is provided. Validation of any target image may be performed by creating a model of graphical components of a source image or a reference model and comparing the model of graphical components of the source image or reference model with a model of graphical components of the target image in order to detect differences between the graphical components. In one embodiment, a framework may be provided with an analysis engine for building a model of graphical components of a source image to be compared with a model of graphical components of a target image. In another embodiment, a framework may be provided with a synthesis engine for building a model of a synthesized graphical component to be compared with a model of graphical components of a target image.

Abstract: A method for the timely processing of tactical images which allows a military pilot adequate time to track and engage a target. The method uses a correlation process and back propagated neural network to home in on a correct position parameters for the tactical image including the heading and the range for the tactical image.

Abstract: A gesture recognition system enabling control of a robotic device through gesture command by a user is provided, comprising a robotic unit, a video or infrared camera affixed to the robotic unit, computing means, and high and low level of control gesture recognition application code capable of enabling the system to locate points of left hand, right hand, upper torso and lower torso of the user in the video imagery and convert it to waveform data, correlate the waveform data to user command data, and form corresponding control voltage command(s) for production of electric current voltage(s) to drive one or more of the electric motors or actuators of the robotic device to thereby control same. In addition, a computer software program is provided for use in the gesture recognition system described above.

Abstract: A method for providing an intensity or brightness measurement using a digital image-capturing device comprising: selecting a target area within a field of view of the image-capturing device, the target area containing pixels; measuring the intensity or brightness of pixels in a target area; accumulating the intensity or brightness values of the pixels in the target area; and determining a pixel value representative of the intensity or brightness of the pixels in the target area. A device for making color measurements comprising an image-capture device, a processor or logic device, and a memory location for accumulating color data, and the processor or logic device is programmed to perform color measurements by accumulating the data for pixels located in the target area in memory, and determining a representative color value.

Abstract: An object to be imaged or detected is illuminated by a single broadband light source or multiple light sources emitting light at different wavelengths. The light is detected by a detector, which includes a light-detecting sensor covered by a hybrid filter.

Abstract: An image processing apparatus having an arithmetic circuit for generating pixel data of output pixels by arithmetically processing the pixel data of M horizontally adjoining pixels and N vertically adjoining pixels. A first temporary storage section readably stores the pixel data of the pixels ranging from the Mth pixel to the last pixel in each horizontal pixel line of an image. A second temporary storage section readably stores the pixel data of the pixels ranging from the head pixel to the (M?1)th pixel in each horizontal pixel line of the image. A third temporary storage section delays the pixel data stored in the first temporary storage section, receives the pixel data from the second temporary storage section, and simultaneously outputs pixel data of the M horizontally adjoining pixels and the N vertically adjoining pixels.

Abstract: The invention relates to a method for providing corrected data in order to generate a model of a monitoring area which is located in the respective visual range of at least two opto-electronic sensors for determining the position of detected objects, based on the amounts of raw data elements corresponding to object points in said monitoring area, said amounts of raw data elements being respectively detected by one of the sensors when the visual range is sensed and associated therewith, and the amounts of raw data elements from various sensors which are temporally linked to each other forming a group. The raw data elements corresponding to said amounts respectively comprise at least the coordinates of object points detected by the associated sensor in relation to the position of the detecting sensor.

Abstract: Methods to convert the resolution of digital images have been achieved. The methods invented support the decrease or the increase of the resolution of digital images, even enlargement in one direction and decimation in the other direction. The methods invented are applicable for any color space used. In case of decimation the method invented combines an interpolation of source pixels to calculate the color values of the destination pixels and omitting some rows of source pixels. A linear interpolation method is used to get a fast result of the interpolation. In case of an enlargement of a digital image in one or two directions the method invented combines the calculation of the color values of the destination pixels by an interpolation of the nearest source pixels with an extrapolation of the destination pixels being close to the edge of the image according to the scale of the conversion of resolution in one or two directions.

Abstract: The systems and methods described herein are directed at accelerating video encoding using a graphics processing unit. In one aspect, a video encoding system uses both a central processing unit (CPU) and a graphics processing unit (GPU) to perform video encoding. The system implements a technique that enables the GPU to perform motion estimation for video encoding. The technique allows the GPU to perform a motion estimation process in parallel with the video encoding process performed by the CPU. The performance of video encoding using such a system is greatly accelerated as compared to encoding using just the CPU. In another aspect, data related to motion estimation is arranged and provided to the GPU in a way that utilizes the capabilities of the GPU. Data about video frames may be collocated to enable multiple channels of the GPU to process tasks in parallel. The depth buffer of the GPU may be used to consolidate repeated calculations and searching tasks during the motion estimation process.

Abstract: An output mechanism of a character recognition device includes a category classifier for classifying image data of characters to be recognized for each category recognized in character recognition processing, a clustering processor for determining feature values related to shapes of characters included in the image data in each category classified by the category classifier, and for classifying the image data into one or more clusters based on the feature values, and a screen creator for creating a confirmation screen for displaying the image data for each cluster classified by the clustering processor.

Abstract: A Mixed Media Reality (MMR) system and associated techniques are disclosed. The MMR system provides mechanisms for forming a mixed media document that includes media of at least two types (e.g., printed paper as a first medium and digital content and/or web link as a second medium). In one particular embodiment, the mixed media document may also include hotspot images that uniquely identify a user associated with the mixed media document. Such individualized mixed media documents are used to identify users of the mixed media document, when orders for products or services are made using the mixed media documents through the MMR system.

Abstract: Methods to zoom a region of interest from a digital image have been achieved. The methods invented support according to the scale of zooming the decrease or the increase of the resolution of digital images. The region of interest to be zoomed is variable; the destination image has a fixed size. In case of a decrease of the resolution of the region of interest the method invented combines interpolating source pixels to calculate the color values of the destination pixels and omitting some rows of source pixels. In case of enlargement the method invented combines interpolation of the source pixels to calculate the color values of the destination pixels, extrapolation of the destination pixels being close to the edge of the image, and replication of some of the interpolated rows of the destination image to gain additional rows according to the scale of the increase of resolution (enlargement) of a digital image.