Abstract: A method for assisting in the creation of a logical structure model, which stores, from an image in which character strings associated respectively with a plurality of logical elements constituting a logical structure are described, the logical elements, character strings associated with the logical elements, and the logical structure, wherein character strings in an input image and the logical structure among the character strings in the input image are extracted, a logical element is selected among the plurality of logical elements according to the degrees of similarity between the extracted character strings and the character string associated respectively with the plurality of logical elements stored in the logical structure model, a character string associated with the selected logical element and a character string in the input image associated with the logical element based on the logical structure among the extracted character strings in the input image are extracted.

Abstract: An image recognition method is conducted by recognizing logical elements based on a logical structure model set to correspond to the logical structure of an image of individual character strings, collecting information processed with the logical structure model of images of a logical structure, acquiring a recognition result when recognizing an image of a logical structure by processing information collected with a post-update logical structure model, and outputting warning information about the post-update logical structure model to an output unit when a result of the comparison is a non-match.

Abstract: There provides an apparatus for and a method of generating a classifier for detecting a specific object in an image. The apparatus for generating a classifier for detecting a specific object in an image includes: a region dividing section for dividing, from a sample image, at least one square region having a side length equal to or shorter than the length of shorter side of the sample image; a feature extracting section for extracting an image feature from at least a part of the square regions divided by the region dividing section; and a training section for performing training based on the extracted image feature to generate a classifier. By using the apparatus for and method of generating the classifier, it becomes possible to make full use of recognizable regions of objects to be recognized with variable aspect ratios and improve speed and accuracy for recognizing in complex backgrounds.

Abstract: Ruled lines are extracted and fitted into a real 2-D space. Correspondence between fitted cells and template cells of a ruled line template is determined. For each pair of cells corresponding to each other, the position of each pixel in the template cell is mapped into a real position in the real 2-D space based on an affine transformation between the cells. A pixel value based on pixel values of a plurality of pixels in the image with positions adjacent to the real position is generated as a pixel value of the pixel in the template cell corresponding to the real position. A synthesized image corresponding to the image is generated by merging the ruled lines of the ruled line template with the pixels in the template cells having the pixel values as generated. A form template is obtained based on the synthesized images corresponding to the plurality of images.

Abstract: A method for processing a document image includes: performing horizontal and vertical text line extraction on the document image; providing an overlapping matrix, a value of an element of the overlapping matrix indicating an overlapping relation between horizontal and vertical text lines; merging the overlapping matrix in the vertical and horizontal direction; determining one or more text overlapping regions in the document image, based on the values of the elements of the merged overlapping matrix; counting the total number of strokes or pixel points in the horizontal and vertical text lines, respectively, within one of the one or more text overlapping regions; and determining an orientation of the text overlapping region is horizontal if the total number of strokes or pixel points in the horizontal text lines is larger than that in the vertical text lines, otherwise, determining the orientation is vertical.

Abstract: A replay control method of controlling reply means for replaying video content executed by a computer, the method includes: accepting one or more keywords; retrieving, from pieces of correspondence information each containing fraction part information specifying a piece of video content and a fraction part in the piece of video content, and a word string expressed in the fraction part, each piece of correspondence information whose word string contains at least one of the accepted one or more keywords; and making the replay means replay the fraction part specified by each retrieved piece of correspondence information.

Abstract: A character area extracting device includes a reflective and non-reflective area separation unit separating image data into reflective and non-reflective areas, and binarizing the image data by changing a first threshold value when it is inappropriate; a reflective area binarizing unit separating the reflective area into character and background areas, and binarizing it by changing a second threshold value when it is inappropriate; a non-reflective area binarizing unit separating the non-reflective area into the character and background areas, and binarizing it by changing a third threshold value when it is inappropriate; a reflective and non-reflective area separation evaluation unit; and a line extracting unit connecting the character areas of the reflective and non-reflective areas and extracting positional information of the connected character areas in the image data.

Abstract: A document recognizing apparatus includes a display control unit which displays a document data including a character string related to a character string selected by a user, and an area that includes at least a character string of the document data.

Abstract: A logical structure analyzing apparatus includes an extracting unit that extracts word candidates from a form, a first generating unit that classifies each of the word candidates into a group of heading candidates or a group of data candidates to generate, based on positions of the word candidates on the form, first candidates sets each including one heading candidate and one data candidate identifiable by the heading candidate, and a second generating unit that combines the first candidate sets to generate second candidate sets that each include plural heading candidates that differ and one data candidate. The apparatus also includes a removing unit that, based on positions of the heading candidates and the data word candidate in each second candidate set, removes from among the second candidates sets, a determined set including a data item and headings identifying the data item, and an output unit that outputs the determined set.

Abstract: An apparatus includes: a unit to identify first pixels belonging to a first portion regarded as a character string that is inserted with an overlap on a background in an expanded image generated by expanding a specific frame image included in video image data; a unit to determine whether any one of the first pixels is out of a display area that is a part of the expanded image and to calculate a movement amount for moving the first portion so as to make all of the first pixels accommodated in the display area when it is determined that any one of the first pixels is out of the display area; and a unit to identify a movement destination pixel for each of the first pixels or the like, according to the movement amount, and to replace a color of the movement destination pixel with a predetermined color.

Abstract: Upon receiving, for example, document data including a character string from outside, a character recognition device detects a line from a line-touching character-string image in which at least one character (such as number, alphabet letter, kana character, and Chinese character) touches (or overlaps) a line in the document data, tentatively removes the line, and estimates a character region. The character recognition device extracts a line-touching character image from the line-touching character-string image (original image) based on the estimated character region. The character recognition device creates a line-added reference character image by adding a quasi-line to a reference character image stored in advance.

Abstract: Precise grayscale character segmentation apparatus and method. The precise grayscale character segmentation apparatus comprises an adjustment and segmentation unit for adjusting and segmenting an inputted low resolution text line image undergone coarse segmentation, so as to generate an adjusted character image; a character image binarization unit for generating a binary character image from the character image inputted therein; a noise removal unit for removing noise information in the binary character image generated by the binarization unit; and a final character image segmentation unit for generating a precisely segmented character image from the binary character image from which noise has been removed.

Abstract: This method includes: extracting a feature vector for an input character from a reading result of the input character; calculating distances between the feature vector for the input character and vectors including average vectors stored in a system dictionary storing, for each character, the average vector and distribution information, and feature vectors stored in a user dictionary; extracting the top N character codes in an ascending order of the calculated distances; obtaining second distribution information for the character codes, which are included the user dictionary and in the top N character codes; calculating, for each of the top N character codes, a second distance with the feature vector for the input character, by using, for the character codes, which are included in the user dictionary and in the top N character codes, the second distribution information; and identifying a character code whose second distance is shortest.

Abstract: A dictionary creating apparatus registers probability distributions each including an average vector and a covariance matrix, in a dictionary. The dictionary creating apparatus organizes plural distribution profiles of character categories having similar feature vectors into one typical distribution profile, and registers the typical distribution profile and the character categories to be organized, associated with each other, in the dictionary, without registering eigenvalues and eigenvectors of all character categories, associated with each other, in the dictionary.

Abstract: A method for assisting in the creation of a logical structure model, which stores, from an image in which character strings associated respectively with a plurality of logical elements constituting a logical structure are described, the logical elements, character strings associated with the logical elements, and the logical structure, wherein character strings in an input image and the logical structure among the character strings in the input image are extracted, a logical element is selected among the plurality of logical elements according to the degrees of similarity between the extracted character strings and the character string associated respectively with the plurality of logical elements stored in the logical structure model, a character string associated with the selected logical element and a character string in the input image associated with the logical element based on the logical structure among the extracted character strings in the input image are extracted.

Abstract: A document recognizing apparatus includes a display control unit which displays a document data including a character string related to a character string selected by a user, and an area that includes at least a character string of the document data.

Abstract: The form data extracting apparatus, even input form does not have a logical structure stored in the generic logical structure DB, by using logical elements in the existing logical structure and a registered form obtained on the basis of (a) the logical structure, (b) pieces of position information of the logical elements, and (c) a relation between the logical elements. A logical element and a logical structure are extracted from the input form, and the extracted logical structure can be defined as a new registered form or a new logical structure.

Abstract: A grayscale character dictionary generation apparatus, comprising a first synthetic grayscale degraded character image generation unit for generating first synthetic grayscale degraded character images using binary character images inputted therein; a clustering unit for dividing each category of the first synthetic grayscale degraded character images generated by the first synthetic grayscale degraded character image generation unit into a plurality of clusters; a template generation unit for generating template for each of the clusters; a transformation matrix generation unit for generating transformation matrix in relation to each of the templates; and a second synthetic grayscale degraded character dictionary generation unit for obtaining character feature of every grayscale degraded character of each of the clusters using the transformation matrix, and for constructing eigenspace of each category of the synthetic grayscale degraded character, which is the second synthetic grayscale character dictionary.

Abstract: A logical structure analyzing apparatus includes an extracting unit that extracts word candidates from a form, a first generating unit that classifies each of the word candidates into a group of heading candidates or a group of data candidates to generate, based on positions of the word candidates on the form, first candidate sets each including one heading candidate and one data candidate identifiable by the heading candidate, and a second generating unit that combines the first candidate sets to generate second candidate sets that each include plural heading candidates that differ and one data candidate. The apparatus also includes a removing unit that, based on positions of the heading candidates and the data word candidate in each second candidate set, removes from among the second candidate sets, a determined set including a data item and headings identifying the data item, and an output unit that outputs the determined set.

Abstract: An image recognition apparatus recognizes the correspondence between character strings and logical elements composing a logical structure in an image in which the character strings are described as the logical elements to recognize each logical element.