Abstract: The present invention discloses an identifying method of hand-written Latin letter. The present invention considers many hand-written styles of Latin letter, extract many stable characteristics of Latin letter of different hand-written styles, and classify the Latin letter aggregation each time with one characteristic, so that the whole standard Latin letter aggregation is classified into many small Latin letter aggregations with intersection to be the coarse classification candidate letter aggregations to be identified. When identifying the inputted hand-written Latin letter, obtain the coarse classification candidate letter aggregation that matches with the characteristics of the inputted hand-written Latin letter. Many stable characteristics ensure the identifying rate. The multilayer coarse classification candidate letter aggregations regulate the searching path and increase the identifying speed.

Abstract: A form processing program which is capable of automatically extracting keywords. When the image of a scanned form is entered, a layout recognizer extracts a readout region of the form image, a character recognizer recognizes characters within the readout region. A form logical definition database stores form logical definitions defining strings as keywords according to logical structures which are common to forms of same type. A possible string extractor extracts as possible strings combinations of recognized characters each of which satisfies defined relationships of a string. A linking unit links the possible strings according to positional relationships, and determines a combination of possible strings as keywords.

Abstract: A text input device receives, in its information input circuit, a letter indicating a destination of transmission as information on the destination of transmission. The text input device stores, in its word-finder with learning function, an input text and an output text in a state correlated with the information on the destination of transmission or its attribute. The text input device in its text learning circuit controls a change in storage caused by correlating an input text matched to a text entered with the information on the destination of transmission or its attribute stored and coincident with the information on the destination of transmission or its attribute entered. When a text matched to the text entered is output, the text input device in its text converter takes out and outputs at least one output text stored.

Abstract: Described is searching directly based on digital ink input to provide a result set of one or more items. Digital ink input (e.g., a handwritten character, sketched shape, gesture, drawing picture) is provided to a search engine and interpreted thereby, with a search result (or results) returned. Different kinds of digital ink can be used as search input without changing modes. The search engine includes a unified digital ink recognizer that recognizes digital ink as a character or another type of digital ink. When the recognition result is a character, the character may be used in a keyword search to find one or more corresponding non-character items, e.g., from a data store. When the recognition result is a non-character item, the non-character item is provided as the result, without keyword searching. The search result may appear as one or more item representations, such as in a user interface result panel.

Abstract: An apparatus (100) for handwriting recognition has a touch-sensitive display screen (240) providing a hand writing input area (270) capable of detecting hand-made user input. The apparatus also has a processing device (300) coupled to the touch-sensitive display screen and providing a user interface to a user. The handwriting input area (270) includes a writing start area (280) capable of switching between a first two-dimensional scope (282) and a second two-dimensional scope (282?), larger than the first two-dimensional scope. The processing device (300) is configured to handle said handmade user input as either a logical mouse event, associated with a control operation for said user interface, or a logical pen event, associated with handwriting.

Abstract: Illustrative embodiments provide a computer implemented method, a data processing system and a computer program product for transforming character data input between a first writing system and a second writing system. The computer implemented method comprises receiving character data input of a first writing system and ensuring the character data input contains normalized characters. A predefined transform is selected based on the character data input of the first writing system and output to a second writing system to transform the normalized characters of the first writing system to character data output of the second writing system, and providing the character data output to a display process.

Abstract: A method and system for preprocessing an image for Optical Character Recognition (OCR), wherein the image includes a plurality of columns is disclosed. Each column includes one or more of Arabic text and non-text items. The method includes determining a plurality of components associated with one or more of the Arabic text and the non-text items, wherein a component includes a set of connected pixels. On determining the plurality of components, a line height and a column spacing is determined for the plurality of components. The plurality of components are then associated with a column of the plurality of columns based on the line height and the column spacing. Subsequently, a set of characteristic parameters are calculated for each column and the plurality of components of each column are merged based on the set of characteristic parameters to form sub-words and words.

Abstract: A character recognizing system includes a portable electronic device, a location sensing system and a server system. The portable electronic device captures image of a target to produce a captured image. The location sensing system locates position of the portable electronic device to produce a position information. The server system receives the captured image and the position information via internet for executing recognizing motion.

Abstract: The disclosed architecture is a new feature extraction approach to handwriting recognition. Given an handwriting sample (e.g., from an online source), a sequence of time-ordered dominant points are extracted, which include stroke-endings, points corresponding to local extrema of curvature, and points with a large distance to the chords formed by pairs of previously identified neighboring dominant points. At each dominant point, a multi-dimensional feature vector is extracted, which includes a combination of coordinate features, delta features, and double-delta features.

Abstract: A method for determining the orientation of Chinese words is provided. The amount of dark pixels in each column of a Chinese word image is calculated. Then, a first point, a second point, and a third point are determined. The first point and the second point correspond to the columns with the largest and the second largest amount of dark pixels, respectively. The third point is located between the first point and the second point. The Chinese word is right-side up if the third point is located on the left side of the Chinese word. The Chinese word is upside down if the third point is located on the right side of the Chinese word.

Abstract: A scanning apparatus and a method thereof include a scanning unit scanning a document and outputting a scanned result, at least one external storage unit detachably attached to the apparatus, at least one internal storage unit, and a controller detecting an attachment state of the external storage unit and storing the scanned result in one of the external storage unit and the internal storage unit according to the attachment state of the external storage unit. The scanning unit of the scanning apparatus is combined with a user scanning unit and a user printing unit into a combination apparatus, and the scanned result is printed in a printing apparatus spaced-apart from the scanning apparatus by a distance, thereby removing cables between the scanning or printing apparatus and a personal computer.

Abstract: A input method selects a character from a plurality of characters of a logographic script, and identifies characters proximate the selected character. One or more candidate characters are then selected based on a composition input and the proximate characters.

Abstract: A method and apparatus for determining an orientation of a document including Korean text are presented. A binarized pixel image is created from the document image. Contiguous pixels are grouped and labeled using a bounding box. A spanning stroke may be detected from a group of the contiguous pixels. The orientation of the document is determined by comparing counts associated with spanning strokes in the left, right, top, and bottom halves of the bounding boxes.

Abstract: An exemplary method for online character recognition of East Asian characters includes acquiring time sequential, online ink data for a handwritten East Asian character, conditioning the ink data to produce conditioned ink data where the conditioned ink data includes information as to writing sequence of the handwritten East Asian character and extracting features from the conditioned ink data where the features include a tangent feature, a curvature feature, a local length feature, a connection point feature and an imaginary stroke feature. Such a method may determine neighborhoods for ink data and extract features for each neighborhood. An exemplary Hidden Markov Model based character recognition system may use various exemplary methods for training and character recognition.

Abstract: A method for classifying a handwritten input character is disclosed. Character models are used. Each character model is associated with an output character and defines a model specific segmentation scheme for that output character and an associated segment model. The model specific segmentation scheme defines a minimum length corresponding to a number of points in a stroke of the output character and a minimum length threshold. Using each of the character models, the input character is decomposed into segments and the segments are evaluated against the segment model of the respective character model to produce a score indicative of the conformity of the segments with the segment model. The character model that produced the highest score is selected and the input character is classified as the output character associated with the character model that produces the highest score.

Abstract: In embodiments consistent with the subject matter of this disclosure, a user may input strokes as digital ink to a processing device. The processing device may partition the input strokes into multiple regions of strokes. A first recognizer and a second recognizer may score grammar objects included in regions and represented by chart entries. The scores may be converted to a converted score, which may have at least a near standard normal distribution. The processing device may present a recognition result based on highest converted scores according to a recurrence formula. The processing device may receive a correction hint with respect to misrecognized strokes and may add a penalty score with respect to chart entries representing grammar objects breaking the correction hint. Incremental recognition may be performed when a pause is detected during inputting of strokes.

Abstract: A method of classifying a character string formed from a known number of hand-written characters is disclosed. The method starts by determining character probabilities for each hand-written character in the character string. Each character probability represents a likelihood of the respective hand-written character being a respective one of a plurality of predetermined characters. Each predetermined character has a respective character type. Character templates having the known number of characters are next identified. Each character template has a respective predetermined probability and represents a respective combination of character types. Character sequence probabilities corresponding to each of the character templates having the known number of characters are next determined. The character sequence probabilities are a function of the predetermined probability of the respective character template and the character probabilities of the hand-written character in the character string.

Abstract: A method for providing a correct e-dictionary search result for a document recognition result includes performing character recognition of a document in which Korean characters (Hangul) and Chinese characters are mixed and displaying a recognition result. If a character string to be searched is selected by a user from the recognition result, determining whether the selected character string corresponds to Hangul or Chinese characters, detecting a Hangul word or a Chinese word included in the selected character string, and outputting an e-dictionary search result corresponding to the detected Hangul or a Chinese word. Accordingly, the user can use an e-dictionary function without directly inputting a search word and obtain a correct e-dictionary search result for a document in which Hangul and Chinese characters are mixed.

Abstract: A practical and natural way of inputting syllables of scripts into a computer. In one example embodiment, This is achieved by selecting a base character from a set of characters using a digitizing tablet [1216]. The selected base character is then modified by drawing one or more natural handwritten modifying gestures to form a current desired syllable. An associated data of the formed current desired syllable is then inputted into a gesture-keypad-engine [1230] via the digitizing tablet [1216] upon completion of the drawing of the one or more natural handwritten modifying gestures. The gesture-keypad-engine [1230] then produces a current candidate syllable as a function of the inputted associated data of the formed current desired syllable. The produced current candidate syllable is then displayed on a display device [540].

Abstract: A memory footprint of an Modified Quadratic Discriminant Function (MQDF) pattern recognition classifier is reduced without resulting in unacceptable classification accuracy degradation. Covariance matrices for multiple classes are clustered into a smaller number of matrices where different classes share the same set of eigenvectors. According to another approach, different numbers of principal components are stored for different classes based on criteria such as class usage frequency, larger variation in writing, and the like, resulting in fewer principal components to be stored in memory.

Abstract: A live video stream captured by an on-device camera is displayed on a screen with an overlaid guideline. Video frames of the live video stream are analyzed for a video frame with acceptable quality. A text region is identified in the video frame approximate to the on-screen guideline and cropped from the video frame. The cropped image is transmitted to an optical character recognition (OCR) engine, which processes the cropped image and generates text in an editable symbolic form (the OCR'ed text). A confidence score is determined for the OCR'ed text and compared with a threshold value. If the confidence score exceeds the threshold value, the OCR'ed text is outputted.

Abstract: A character string recognition method for recognizing a character string may include a first step in which a first projection data of image data are calculated in a direction of the character string and a second step in which a position of the character string is detected on the basis of the first projection data. In the first step, the image data are divided into a plurality of segments in the direction of the character string and projection in the segment is calculated. The method may further include a third step in which a second projection data in the segment are calculated on the basis of the position of the character string and a fourth step in which a position where the second projection data exceeds a threshold value is detected as a boundary position of a character, and the threshold value may be changed according to pixel number between both ends of the character string.

Abstract: A handwriting processing apparatus and method effective for search of, e.g., a document file including handwriting is provided. When a handwriting characters are input to a coordinate input unit as a search key, a corresponding character in a dictionary is recognized for each of the handwritten characters, a search unit searches for a text code stored in a document file storage unit based on a text code of the corresponding character if the corresponding character is recognized and the search unit searches for handwriting trail data stored in a handwriting trail storage unit based on a handwriting trail of the handwriting character if the corresponding character is not recognized, thereby finding a desired document file.

Abstract: Exemplary methods, systems, and computer-readable media for developing, training and/or using models for online handwriting recognition of characters are described. An exemplary method for building a trainable radical-based HMM for use in character recognition includes defining radical nodes, where a radical node represents a structural element of an character, and defining connection nodes, where a connection node represents a spatial relationship between two or more radicals. Such a method may include determining a number of paths in the radical-based HMM using subsequence direction histogram vector (SDHV) clustering and determining a number of states in the radical-based HMM using curvature scale space-based (CSS) corner detection.

Abstract: The present invention provides a Chinese character search method. According to the method, the user first inputs the notation of the known character. When the input notation is received, some corresponding Chinese characters are shown in the display. Then, the user chooses the correct character for which he is searching. Next, symbols are added to this character to represent the unknown Chinese character in the word. Then, those words containing this character are searched and are shown in a display.

Abstract: An apparatus for Chinese language education includes a set of Chinese character pieces, a Chinese dictionary database, an input unit, and a processing unit. Each of the Chinese character pieces is associated with a distinct Chinese character, has the distinct Chinese character visibly indicated thereon, and is provided with a distinct machine-readable identification code. The Chinese dictionary database contains a plurality of dictionary entries, each of which corresponds to a respective one of the Chinese character pieces and includes information of the distinct Chinese character, such as pronunciation, evolution of a character form, stroke sequence and a radical of the distinct Chinese character, and meaningful Chinese phrases including the distinct Chinese character. The input unit reads the identification code of a selected one of the Chinese character pieces. The processing unit determines the dictionary entry corresponding to the selected one of the Chinese character pieces.

Abstract: Handwriting recognition techniques employing a personalized handwriting recognition engine. The recognition techniques use examples of an individual's previous writing style to help recognize new pen input from that individual. The techniques also employ a shape trainer to select samples of an individual's handwriting that accurately represent the individual's writing style, for use as prototypes to recognize subsequent handwriting from the individual. The techniques also alternately or additionally employ an intelligent combiner to combine the recognition results from the personalized recognition engine and the conventional recognition engine (or engines). The combiner may use a comparative neural network to combine the recognition results from multiple recognition engines. The combiner alternately may use a rule-based system based on prior knowledge of different recognition engines.

Abstract: An image display device and an operation method thereof are provided that include receiving signals corresponding to spatial coordinates of the pointing device, recognizing at least one character based on the received signals, and displaying a channel list including at least one of a channel number or characters based on the at least one recognized character.

Abstract: A method and system for implementing character recognition is described herein. An input character is received. The input character is composed of one or more logical structures in a particular layout. The layout of the one or more logical structures is identified. One or more of a plurality of classifiers are selected based on the layout of the one or more logical structures in the input character. The entire character is input into the selected classifiers. The selected classifiers classify the logical structures. The outputs from the selected classifiers are then combined to form an output character vector.

Abstract: To make searching for pictographic characters, such as Chinese characters, easier for novice learners of languages using pictographic characters, a subset of pictographic character parts of the pictographic character is generated. Then, the subset of the pictographic character parts is used to generate the pictographic character based on the subset of the pictographic character parts.

Abstract: An apparatus and method for processing a captured image and, more particularly, for processing a captured image comprising a document. In one embodiment, an apparatus comprising a camera to capture documents is described. In another embodiment, a method for processing a captured image that includes a document comprises the steps of distinguishing an imaged document from its background, adjusting the captured image to reduce distortions created from use of a camera and properly orienting the document is described.

Abstract: Various technologies and techniques are disclosed that improve handwriting recognition operations. Handwritten input is received in training mode and run through several base recognizers to generate several alternate lists. The alternate lists are unioned together into a combined alternate list. If the correct result is in the combined list, each correct/incorrect alternate pair is used to generate training patterns. The weights associated with the alternate pairs are stored. At runtime, the combined alternate list is generated just as training time. The trained comparator-net can be used to compare any two alternates in the combined list. A template matching base recognizer is used with one or more neural network base recognizers to improve recognition operations. The system provides comparator-net and reorder-net processes trained on print and cursive data, and ones that have been trained on cursive-only data. The respective comparator-net and reorder-net processes are used accordingly.

Abstract: Described is a technology by which handwriting recognition is performed using a precision constrained Gaussian model (PCGM) that requires far less memory than other models such as MQDF. Offline training, such as via maximum likelihood and/or minimum classification error techniques, provides classification data. The classification data includes basis matrices that are shared by classes, along with weighting coefficients and a mean vector corresponding to each class. The base matrices and weights are obtained by expanding a precision matrix for each class. In online recognition, received handwritten input (e.g., an East Asian character) is classified into a class, based upon the per-class mean vector and weighting coefficients, and the basis matrices, by a PCGM recognizer that outputs similarity scores for candidates and a decision rule that selects the most likely class.

Abstract: The automatic Arabic text image optical character recognition method includes training a text recognition system using Arabic printed text, using the produced models for classification of newly unseen Arabic scanned text, and generating the corresponding textual information. Scanned images of Arabic text and copies of minimal Arabic text are used in the training sessions. Each page is segmented into lines. Features of each line are extracted and input to Hidden Markov Model (HMM). All training data training features are used. HMM runs training algorithms to produce codebook and language models. In the classification stage new Arabic text is input in scanned form. Line segmentation where lines are extracted is passed through. In the feature stage, line features are extracted and input to the classification stage. In the classification stage the corresponding Arabic text is generated.

Abstract: Recognizing handwritten words at an electronic device. A plurality of strokes is received at a common input region of an electronic device. The plurality of strokes in combination defines a word comprising a plurality of symbols, a relative geometry of a first subset of the plurality of strokes defines a first symbol and a relative geometry of a second subset of the plurality of strokes defines a second symbol such that the relative geometry of the first subset of the plurality of strokes is not related to the relative geometry of the second subset of the plurality of strokes, and at least one stroke of the first subset of the plurality of strokes is spatially superimposed over at least one stroke of the second subset of the plurality of strokes.

Abstract: Exemplary techniques are described for selecting radical sets for use in probabilistic East Asian character recognition algorithms. An exemplary technique includes applying a decomposition rule to each East Asian character of the set to generate a progressive splitting graph where the progressive splitting graph comprises radicals as nodes, formulating an optimization problem to find an optimal set of radicals to represent the set of East Asian characters using maximum likelihood and minimum description length and solving the optimization problem for the optimal set of radicals. Another exemplary technique includes selecting an optimal set of radicals by using a general function that characterizes a radical with respect to other East Asian characters and a complex function that characterizes complexity of a radical.

Abstract: Described is a technology by which handwriting recognition is performed using a semi-tied covariance modeling (STC) that requires far less memory than other models such as MQDF. Offline training, such as via maximum likelihood and/or minimum classification error techniques, provides classification data. The classification data includes semi-tied transforms that are shared by classes, along with a class-dependent diagonal matrix and a mean vector corresponding to each class. The semi-tied transforms and class-dependent diagonal matrices are obtained by processing a precision matrix for each class. In online recognition, received handwritten input (e.g., an East Asian character) is classified into a class, based upon the class-dependent diagonal matrices and the semi-tied transforms, by a STC recognizer that outputs similarity scores for candidates and a decision rule that selects the most likely class.

Abstract: A form processing apparatus extracts layout information and character information from a form document. A candidate extracting unit extracts word candidates from the character information. A frequency digitizing unit calculates emission probability of a word candidate from each element. A relation digitizing unit calculates transition probability that relationship between word candidates is established. An evaluating unit calculates an evaluation value indicative of a probability of appearance of word candidates in respective logical elements. A determining unit determines the element and a word candidate thereof as the element and a character string thereof in the form document, based on the evaluation value.

Abstract: A method of creating font format data from source font data includes analyzing the source font data to obtain glyph data for a plurality of glyphs, dissecting the glyph data, extracting midline data from the dissected glyph data, classifying the midline data as unique element data and common element data, associating unique element data and common element data to each glyph of the plurality of glyphs.

Abstract: An apparatus and method for searching a handwritten memo, which is input by a user using a digital pen interface, for a word corresponding to the user's query. The apparatus includes a preprocessing unit which removes unnecessary portions from digital ink data of an input query phrase and an input memo to reduce an information amount, a feature extraction unit which extracts a feature vector from the digital ink data having the reduced information amount, and a query searching unit which searches the memo for a portion matched with the query phrase in units of segments. Therefore, an accurate result can be obtained quickly when an existing memo or document is searched for desired content by inputting a query phrase using a digital pen.

Abstract: A hand held device for generating commands and transferring data between the hand-held device and a base device (including consumer electronic equipment). The hand-held device detects the motion of the device itself, interpreting the motion as a command, and executing or transferring the command. The motion of the device can include gestures made by the user while holding the device, such as the motion of throwing the hand-held device toward a base device. The commands generated by the user range from basic on/off commands to complex processes, such as the transfer of data. In one embodiment, the user can train the device to learn new motions associated with existing or new commands. The hand-held device analyzes the basic components of the motion to create a motion model such that the motion can be uniquely identified in the future.

Abstract: A method, computer program product, and a data processing system for performing handwriting recognition of a language having character stroke order rules. A stroke parameter set describing attributes of a handwritten stroke is calculated, and a user input indicates a stroke order knowledge. A reference character dictionary includes a record having a plurality of reference parameter sets each defining attributes of reference character strokes. A stroke sequence number of the stroke parameter set is identified and at least one of the reference parameter sets are excluded from a comparison with the stroke parameter set based on the stroke sequence number.

Abstract: Systems and methods that exploit unique properties of a language script (e.g., condition joining rules for Arabic language) to enable a two tier text recognition. In such two tier system, one tier can recognize predetermined groups of linked letters that are connected based on joining rules of a language associated with the text, and another tier dissects (and recognizes) such linked letters to respective constituent letters that form the predetermined group of linked letters. Various classifiers and artificial intelligence components can further facilitate text recognition at each level.

Abstract: The present invention discloses a kind of integrated system to recognize comprehensive semantic information, comprising: an information receiver module, to receive information source expressed by any kind of natural languages or texts; a conversion module, to convert the said information source into the semantic information database according to its semantic meaning; a semantic database, composed of Chinese words, in which the Chinese characters are encoded as digits which can be used in computer system according to the coding scheme of radical attribute; and an output module, to convert and output the said digits. The present invention can comprehensively recognize any kind of information source which are expressed by texts or languages, capture all kinds of information or digital information through the electronic system, and comprehensively understand and recognize all these information according to the Chinese words semantic meanings, and then respond with integrated data in simulation way.

Abstract: Aspects of the present invention relate to the creation of an ink font. Based on characteristics of handwritten characters, the collection of characters may be scaled so as to adjust the size of the font to match predefined size values or relationships.

Abstract: An image processing apparatus includes: a recognition unit that recognizes a layout of a line including a character string in an image read from an original; a determination unit that determines a size of a region in which additional information is embedded so as to include at least a part of a line including a character string in the region, based on the layout recognized by the recognition unit; a dividing unit that divides the image read from the original based on the size of the region determined by the determination unit; and an embedding unit that embeds the additional information in the image divided by the dividing unit.

Abstract: An image processing apparatus includes an image receiving unit, a writing detection unit, a writing deletion unit, a character recognition unit, a character string generation unit, a translation unit and a translation image generation unit. The image receiving unit receives an image including a writing. The writing detection unit detects a position of the writing. The writing deletion unit deletes the writing from the received image based on the position of the writing. The character recognition unit recognizes characters in the image from which the writing has been deleted. The character string generation unit generates a character string by inserting a code representative of the writing into the recognition result based on the position of the writing. The translation unit translates the character string. The translation image generation unit generates, based on the translation result, an image of the translation result including an image corresponding to the writing.

Abstract: This computer Chinese character input method mainly includes: Select 10 elements corresponding to the 10 simplified Chinese character strokes, which are and Select 46 elements corresponding to the 46 stroke combination sets, whose representative visual representations are: Assign the above 10 elements and 46 elements to keys on a computer keyboard; Determine desired characters based on the elements input by a user using the keyboard mentioned above or other apparatus.

Abstract: A method for recognizing an image photographed by a camera and translating characters in connection with an electronic dictionary is provided. The method includes directly selecting an area to be recognized from the photographed character image and performing character recognition, translating and recognizing characters of a user's selected word in connection with dictionary data, and displaying translation result information of user's selected character or word in connection with dictionary data on a screen device. The recognition includes providing information on location of the selected character image area and location of the recognized character string words to the user, and then translating a character string or word in a location area selected by the user. The electronic dictionary-connected search and translation is for searching the character or word selected in connection with the electronic dictionary database, and providing translation result to the user.

Abstract: A “gliding” interface operates in a space of visual perceptions and tries to predict an intended pattern sequence to enable a high-speed recognition of Asian writing with a simple interface. When a search begins, the user is presented with a collection of visual patterns within a box. The user can “zoom in” on a visual pattern by moving the cursor toward it. As the user zooms closer, a new layer of patterns appears within the box. The new layer includes more complex visual patterns than those in the previous layer. The system is mistake-tolerant, with no single cumulation of visual patterns required to achieve a specific visual pattern; rather, a statistical algorithm selects the visual patterns most likely to match the unknown structure. The algorithm is updated as the user searches, tracking every visual pattern that the user traverses while using the interface. A database contains groups of visual patterns that describe each Kanji character.