Abstract: An image processing apparatus includes an image data acquiring portion, an image determining portion, and an image data converting portion. The image data acquiring portion is configured to acquire first image data representing a color image. The image determining portion is configured to determine whether or not the image represented by the first image data is an image mainly composed of black characters. The image data converting portion is configured to convert the first image data to second image data by converting the first image data to binary image data when the image determining portion determines that the image represented by the first image data is an image mainly composed of black characters.

Abstract: There are provided methods and systems for use in performing feature sensitive captioning of media content. In one implementation, such a method includes detecting an aesthetically determinative feature of a media content unit selected by a user, and determining a captioning aesthetic for a caption of the media content unit based at least in part on the aesthetically determinative feature. The captioning aesthetic may include a background aesthetic and a text aesthetic. The captioning aesthetic may be utilized by a feature sensitive captioning application to produce a feature sensitive caption for the media content unit.

Abstract: Detecting the start of a credit roll within video program may allow for the automatic extension of video recordings among other functions. The start of the credit roll may be detected by determining the number of text blocks within a sequence of frames and identifying a point in the sequence of frames where a difference between the number of text blocks in frames occurring before the point and the number of text blocks in frames occurring after the point is greatest and exceeds a specified threshold. Text blocks may be identified within each frame by partitioning the frame into one or more segments and recording the segments having a pixel of a sufficiently high contrast. Contiguous segments may be merged or combined into single blocks, which may then be filtered to remove noise and false positives. Additional content may be inserted into the credit roll frames.

Abstract: In a control apparatus, a controller operates as: identifying a reading condition instructed for reading an image from a document; and determining a method of an analysis processing, the identifying including identifying a reading section instructed to read an image from the document. If an identified reading condition satisfies a first condition including that an identified reading section is a first reading section configured to read an image from a document while maintaining the document to be stationary, a first analysis processing configured to extract a first type region from a read out image is determined. If the identified reading condition satisfies a second condition including that the identified reading section is a second reading section configured to read an image from the document while conveying the document, a second analysis processing configured to extract a second type region from the read out image is determined.

Abstract: A method and system for preprocessing text containing region of a video The invention provides a method and system for preprocessing the text containing region of video for improving the optical character recognition input.

Abstract: Detecting text using stroke width based text detection. As a part of the text detection, a representation of an image is generated that includes pixels that are associated with the stroke widths of components of the image. Connected components of the image are identified by filtering out portions of the pixels using metrics related to stroke width. Text is detected in the image based on the identified connected components.

Abstract: A text line segmentation method for a document image containing printed text and handwriting, or document image containing skewed lines or printed text. Connected component (CC) are obtained for the document, and their bounding boxes and centroids are calculated. The CCs are categorized into three categories based on bounding box sizes: small objects, regular text objects, and large objects involving handwriting. The centroids of regular text objects are used in a cluster analysis to find the vertical centers of the N text lines. Then, each CC is classified into one of the N lines based on the vertical distance between its centroid and the vertical centers of text lines, and copied into to a corresponding object board. Extra spaces are removed from the object boards to obtain the line segments. The large object involving handwriting will be classified into one of the lines but absent from other lines.

Abstract: A method including determining a position of each glyph in an image of a text document, identifying word boundaries in the document thereby implying the existence of a first plurality of words, preparing a first array of word lengths based on the first plurality of words, preparing a second array of word lengths based on a second plurality of words of a text file including a certain text, comparing at least part of the first array to at least part of the second array to find a best alignment between the first and second array, deriving a layout of at least part of the certain text as arranged in the image of the text document at least based on the best alignment and the position of at least some of the glyphs in the image. Related apparatus and methods are also described.

Abstract: Disclosed techniques include receiving an electronic image containing depictions of characters, segmenting at least some of the depictions of characters using a first segmentation technique to produce a first segmented portion, and performing a first character recognition on the first segmented portion to determine a first sequence of characters. The techniques also include determining, based on the performing the first character recognition, that the first sequence of characters does not match the depictions of characters. The techniques further include segmenting at least some of the depictions of characters using a second segmentation technique, based on the determining, to produce a second segmented portion, and performing a second character recognition on at least a portion of the second segmented portion to produce a second sequence of characters. The techniques also include outputting a third sequence of characters based on at least part of the second sequence of characters.

Abstract: Character code data and vector drawing data are both listed and provided in a re-editable manner. Electronic data is generated in which information obtained by vectorizing character areas in an image and information obtained by recognizing characters in the image are stored in respective storage locations. As for the electronic data generated in this manner, because character code data and vector drawing data generated from the input image are both presented by a display and edit program, a user can immediately utilize the both data.

Abstract: A word segmentation method for processing a document image applies clustering analysis to the spacing segments of a line. The spacing segments are generated by thresholding a one-dimensional vertical projection profile of the line. Taking advantage of the bimodal distribution of spacing length distribution of text lines, a k-means clustering algorithm is used, with the number of clusters pre-set to two, to classify the spacing segments as either character spacing or word spacing. Moreover, k-means++ initialization is used to enhance performance of cluster analysis. The clustering result such as cluster centers and compactness is used to prune single-word text line, single table item, etc. The locations of the word spacing segments are then used to segment the line of text into words.

Abstract: Systems and methods for sequence transcription with neural networks are provided. More particularly, a neural network can be implemented to map a plurality of training images received by the neural network into a probabilistic model of sequences comprising P(S|X) by maximizing log P(S|X) on the plurality of training images. X represents an input image and S represents an output sequence of characters for the input image. The trained neural network can process a received image containing characters associated with building numbers. The trained neural network can generate a predicted sequence of characters by processing the received image.

Abstract: A method for binarizing a scanned document images containing gray or light colored text printed with halftone patterns. The document image is initially binarized and connected image components are extracted from the initial binary image as text characters. Each text character is classified as either a halftone text character or a non-halftone text character based on an analysis of its topology features. The topology features may be the Euler number of the text character; a text character with a Euler number below ?2 is classified as halftone text. The gray-scale document image is then divided into halftone text regions containing only halftone text characters and non-halftone text regions. Each region is binarized using its own pixel value statistics. This eliminates the influence of black text on the threshold values for binarizing halftone text. The binary maps of the regions are combined to generate the final binary map.

Abstract: A imaging scanner identifies first and second locations in a first and second captured image of a document, analyzes each character in the identified locations, and produces a first and second string, each including a character and a confidence value. The device determines that a first measurement of the confidence values in each of the first and second string is beyond a range of a first threshold. The device compares the confidence value for each character in the first string with a corresponding confidence value in the second string, selects a character from one of the first or second string with a higher confidence value; and produces a combined string including the selected characters and the confidence value associated with each selected character.

Abstract: An automated document processing system is configured to normalize zones obtained from a document, and to extract articles from the normalized zones. In one configuration, the system receives at least one zone from the document, and applies at least one zone-breaking factor, thereby creating normalized sub-zones within which text lines are consistent with the at least one zone-breaking factor. The normalized sub-zones may be evaluated to obtain a reading order. Adjacent sub-zones are joined if text similarity exceeds a threshold value. Weakly joined sub-zones are separated where indicated by a topic vectors analysis of the weakly joined sub-zones.

Abstract: There are provided methods and systems for use in performing feature sensitive captioning of media content. In one implementation, such a method includes detecting an aesthetically determinative feature of a media content unit selected by a user, and determining a captioning aesthetic for a caption of the media content unit based at least in part on the aesthetically determinative feature. The captioning aesthetic may include a background aesthetic and a text aesthetic. The captioning aesthetic may be utilized by a feature sensitive captioning application to produce a feature sensitive caption for the media content unit.

Abstract: An image processing apparatus includes a small area divider that divides, on the basis of edge information of an image, the image into multiple small areas each including multiple pixels; an attribute probability estimator that estimates attribute probability for each of the small areas, which is probability that the small area is attributed to a specific area to be detected; an adjacent-small-area connection strength calculator that calculates connection strength that quantitatively indicates a degree to which small areas adjacent to each other among the multiple small areas are attributed to the same area that is the specific area or a non-specific area; and a specific area detector that detects the specific area on the basis of the attribute probability and the connection strength.

Abstract: Techniques for performing foreground analysis are provided. The techniques include identifying a region of interest in a video scene, detecting a static foreground object in the region of interest, and determining whether the static foreground object is abandoned or removed, wherein said determining comprises performing a foreground analysis based on tracking information and pruning one or more false alarms using one or more track statistics.

Abstract: A server 2 includes an extraction unit 21, an analysis unit 22, a first estimating unit 24, an information acquisition unit 25 and a second estimating unit 26. The extraction unit 21 extracts an image area for each article. The analysis unit 22 analyzes the image area to acquire analysis information. The first estimating unit 24 narrows down candidates estimated to correspond to the article in the image area based on the analysis information. When the candidates were able to be narrowed down, the information acquisition unit 25 acquires additional information additional information of a reference article. The second estimating unit 26 attempts a narrowing process based on the additional information of the reference article in addition to the analysis information, for the image area including a spine, which is an image area in which candidates were unable to be narrowed down.

Abstract: A system and a method are disclosed that determine images with co-occurrence groups of individuals from an image collection. A value of a similarity metric is computed for each pair of images of the image collection, the value of the similarity metric being computed based on a comparison of the number of individuals in common between the images of the pair and the total number of individuals identified in both images of the pair. The collection of images is clustered based on the computed values of the similarity metric. At least one co-occurrence group is determined based on the results of the clustering, where a co-occurrence group is determined as a cluster of images that have a similar combination of individuals.

Abstract: An image processing apparatus includes a first path information calculating unit, a second path information calculating unit, and a path selecting unit. The first path information calculating unit calculates first path information which is information representing a first path for separating areas from an image. The second path information calculating unit calculates second path information representing a second path for separating the areas from the image, the second path being the reverse of the first path. The path selecting unit selects one of the first path information calculated by the first path information calculating unit and the second path information calculated by the second path information calculating unit.

Abstract: An expression, for which complementary information can be outputted, is extracted from a document obtained by character recognition for an image. Complementary information related to the extracted expression is outputted when a character or a symbol adjacent to the beginning or the end of the extracted expression is not a predetermined character or symbol. Output of complementary information related to the extracted expression is skipped when the character or symbol adjacent to the beginning or the end of the extracted expression is the predetermined character or symbol. A problem that complementary information unrelated to an original text is outputted is prevented even when a false character recognition occurs.

Abstract: Detecting text using stroke width based text detection. As a part of the text detection, a representation of an image is generated that includes pixels that are associated with the stroke widths of components of the image. Connected components of the image are identified by filtering out portions of the pixels using metrics related to stroke width. Text is detected in the image based on the identified connected components.

Abstract: Embodiments of the present application relate to a method for recognizing information, a system for recognizing information, and a computer program product for recognizing information. A method for recognizing information is provided. The method includes locating a card zone for each frame within a card image frame sequence comprising a plurality of frames, locating an information zone within each card zone, dividing each information zone into at least one character zone, de-blurring a character zone corresponding to a same region across all the frames in the card image frame sequence, and recognizing character string information based on the de-blurred character zone.

Abstract: A system for segregating handwritten information from typographic information on a document may include a memory, an interface, and a processor. The memory stores an electronic document image of a document where the electronic document image includes pixels and each pixel has a characteristic. The processor may receive, via the interface, the electronic document image and may identify first, second and third most frequently occurring characteristics of the pixels of the electronic document image. The pixels having the first most frequently occurring characteristic represent a background of the document. The processor may determine the typographic information of the document as represented by pixels having the second most frequently occurring characteristic. The processor may determine the handwritten information of the document as represented by pixels having the third most frequently occurring characteristic.

Abstract: In a document analysis system that receives and processes jobs from a plurality of users, in which each job may contain multiple electronic documents, to extract data from the electronic documents, a method of automatically pre-processing each received electronic document using a plurality of image transformation algorithms to improve subsequent data extraction from said document is provided. The method includes: electronically partitioning each received electronic document page into pieces; automatically processing each piece of the received electronic document page using each of a plurality of image pre-processing algorithms to produce a plurality of image variations of each piece; and analyzing the outputs of subsequent processing and data extraction, on each of the image variations of the pieces to determine which output is best, from the plurality of outputs for each piece.

Abstract: The present technology proposes techniques for extracting forms and other types of documents from images taken with a mobile client device. By calculating and making adjustments along a document's detected borders, an input image can be transformed such that the document within the image may be properly aligned and background clutter completely removed. The resulting text fields of the extracted document are thus upright, aligned and locatable at predictable points.

Abstract: An information processing apparatus for performing image processing for document data created by a document creation application to generate print data of each page and sending the generated print data to an image forming apparatus, the information processing apparatus including: a control section for specifying, based on document data, a region where drawing object data included in the document data exists as a drawing object region, and detecting blank space in each print page based on the specified drawing object region.

Abstract: A system and method are provided for processing scanned documents by digitizing the scanned documents, converting the digitized documents to a JPEG2000 file, encoding content information corresponding to the digitized documents using spatial capabilities of JPEG2000's Region of Interest feature, and creating an image file having the digitized documents and the region of interest information for forwarding to a computing device, such as a handheld computing device, for display. The system and method are especially useful in Digital Mail applications which entail digitizing and delivering mail documents to recipients.

Abstract: An image overlaying device includes an image inputting unit, a memory and a controller. The controller is configured to obtain template data which define a plurality of layout areas and to extract each of the plurality of layout areas from the obtained template data. The controller determines positions of the extracted plurality of layout areas. The controller stores, in the memory, layout order information corresponding to an order of the layout area and further stores, in the memory image order information corresponding to an order of the image data of the documents The controller determines corresponding image data of documents corresponding to each of the plurality of layout areas and generates overlaid image data by laying out, in each of the plurality of layout areas, the determined corresponding image data of the documents.

Abstract: Extracting card data comprises receiving, by one or more computing devices, a digital image of a card; perform an image recognition process on the digital representation of the card; identifying an image in the digital representation of the card; comparing the identified image to an image database comprising a plurality of images and determining that the identified image matches a stored image in the image database; determining a card type associated with the stored image and associating the card type with the card based on the determination that the identified image matches the stored image; and performing a particular optical character recognition algorithm on the digital representation of the card, the particular optical character recognition algorithm being based on the determined card type. Another example uses an issuer identification number to improve data extraction. Another example compares extracted data with user data to improve accuracy.

Abstract: Techniques are presented for analyzing audio-video segments, usually from multiple sources. A combined similarity measure is determined from text similarities and video similarities. The text and video similarities measure similarity between audio-video scenes for text and video, respectively. The combined similarity measure is then used to determine similar scenes in the audio-video segments. When the audio-video segments are from multiple audio-video sources, the similar scenes are common scenes in the audio-video segments. Similarities may be converted to or measured by distance. Distance matrices may be determined by using the similarity matrices. The text and video distance matrices are normalized before the combined similarity matrix is determined. Clustering is performed using distance values determined from the combined similarity matrix.

Abstract: A method and a system for providing a text-based representation of a portion of a working area to a user are provided. The method includes acquiring an image of the entire working area and performing a fast OCR process on at least a region of interest of the image corresponding to the portion of the working area, thereby rapidly obtaining an initial machine-encoded representation of the portion of the working area and immediately presenting it to the user as the text-based representation. Parallelly to the fast OCR process, a high-precision OCR process is performed on at least the region of interest of the image, thereby obtaining a high-precision machine-encoded representation of the portion of the working area. Upon completing the high-precision OCR process, the high-precision machine-encoded representation of the portion of the working area is presented to the user as the text-based representation, in replacement of the initial machine-encoded representation.

Abstract: A method for processing an electronic document is provided. The electronic document includes a plurality of text fields and a text label associated with each of the plurality of text fields. The method includes step of extracting the plurality of text fields from the electronic document. The method includes step of grouping the plurality of extracted text fields to generate a plurality of groups. The method includes step of labeling the plurality of groups based on a first pre-defined criteria to generate a plurality of labeled groups. The method includes step of distributing the plurality of labeled groups in a plurality of queues based on a second pre-defined criteria. The method includes step of transmitting the plurality of labeled groups from the plurality of queues to one or more crowdworkers based on a third pre-defined criteria.

Abstract: A system and method of detecting separator lines in a web page may include determining coordinates of visible web elements on a web page, generating an edge image of the web page based on the coordinates of the web elements, filtering edges belonging to non-separator line elements within the edge image, detecting horizontal lines within the edge image, detecting vertical lines within the edge image, and filtering short lines within the edge image. A system for detecting separator lines in a web page may include a memory device, and a processor communicatively coupled to the memory, in which the processor determines coordinates of visible web elements on a web page, generates an edge image of the web page based on the coordinates of the web elements, filters edges belonging to non-separator line elements within the edge image, detects horizontal lines within the edge image, detects vertical lines within the edge image, and filters short lines within the edge image.

Abstract: A system processes an image in an electronic device, by determining whether a text character is included in an image and extracting the determined text character from the image. The extracted text character is stored in association with the image.

Abstract: Methods for correcting distortions in an image including text, or an image of a page that includes text, are disclosed. The methods include identifying reliable and substantially straight lines from elements in the image. Vanishing points are determined from the lines. Parameters associated with a rectangle are determined. A coordinate conversion is performed.

Abstract: Disclosed are a device, a method, and a system for enhancing color. The device comprises a unit used to extract a foreground portion from a whiteboard color image to serve as whiteboard contents; a unit used to stretch R, G, and B channel values of each of a plurality of foreground pixels forming the whiteboard contents; a unit used to adjust color tone of each of the foreground pixels; a unit used to adjust a ratio of color saturation degree to color intensity of each of the foreground pixels so as to cause the ratio to approach a ratio expectation value; and a unit used to increase the color saturation degree and the color intensity of each of the foreground pixel so as to cause the two to approach a color saturation degree expectation value and a color intensity expectation value Id, respectively.

Abstract: An image processing apparatus extracts a line segment included in an image, and includes a density gradient direction determining section that determines a direction, in which density changes, of each processing unit composed of a predetermined number of pixels of an image, and an line segment extracting section that regards a couple of processing units whose density gradient direction are opposite each other as a processing unit pair and extracts a processing unit group including a plurality of processing unit pairs allocated in a row in a direction perpendicular to the density gradient directions as a line segment.

Abstract: This invention is a method for rectifying an input digital image including warped textual information. The method includes analyzing the input digital image to determine local orientations for a plurality of local image regions and determining an orientation vector field by interpolating between the determined local orientations for a lattice of positions. A set of streamlines are determined responsive to the orientation vector field. A global deformation function is formed by interpolating between the streamlines and is used to form a rectified image.

Abstract: A new approach is proposed that contemplates systems and methods to support block-based compression of a compound image by skipping “don't care” blocks in the layers of the image while neither introducing significant overhead nor requiring changes to the compression method used. The block-based compression approach first segments a compound image into multiple layers and then recomposes a new set of image layers, possibly with new dimensions, from only the non-“don't care” blocks in the layers of the original image. The approach may later decompress the compressed image layers and restore the image by copying the decompressed blocks to their respective positions in the original image.

Abstract: The disclosure relates to recognizing data such as items or entities in content. In some aspects, content may be received and feature information, such as face recognition data and voice recognition data may be generated. Scene segmentation may also be performed on the content, grouping the various shots of the video content into one or more shot collections, such as scenes. For example, a decision lattice representative of possible scene segmentations may be determined and the most probable path through the decision lattice may be selected as the scene segmentation. Upon generating the feature information and performing the scene segmentation, one or more items or entities that are present in the scene may be identified.

Abstract: A method embodiment herein begins by capturing a source image. The source image is segmented into first planes. The first planes can each comprise a mask plane and foreground plane combination. The binary images in the first planes are structurally analyzed to identify different regions of text, tables, handwriting, line art, equations, etc., using a document model that has information of size, shape, and spatial arrangement of possible regions. Then, the method extracts (crops out) these regions from the foreground plane to create second mask/foreground plane pairs. Thus, the method creates “second” planes from the first planes, so that a separate second plane is created for each of the regions. Next, tags are associated with each of the second planes (to create tagged mask/foreground plane pairs) and the second planes and associated tags are combined into a mixed raster content (MRC) document.

Abstract: An object area containing a text or a graphic is extracted from a document image containing the text or the graphic. Then, on the basis of the extracted object area and stored size information, a cropping area is determined that surrounds the object area with given margins. Further, setting of margins is received. Then, on the basis of the object area and the received setting of the margins, the cropping area is determined. Then, the cropping area determined on the basis of the object area and the size information or alternatively the cropping area determined on the basis of the object area and the setting of the margins is cropped from the document image.

Abstract: The use of optical character recognition (OCR) in mobile devices is becoming prevalent with the increasing use of mobile devices. One important application for OCR in mobile devices is recognizing and translating the text to a language understandable by the user. Techniques are provided for replacing symbols in an image, while reducing the artifacts as a result of re-rendering of the background image.

Abstract: A method and apparatus for encoding an image is provided. An image coding unit, including a region that deviates from a boundary of a current picture, is divided to obtain a coding unit having a smaller size than the size of the image coding unit, and encoding is performed only in a region that does not deviate from the boundary of the current picture. A method and apparatus for decoding an image encoded by the method and apparatus for encoding an image is also provided.

Abstract: As set forth herein, systems and methods facilitate providing an efficient edge-detection and closed-contour based approach for finding text in natural scenes such as photographic images, digital, and/or electronic images, and the like. Edge information (e.g., edges of structures or objects in the images) is obtained via an edge detection technique. Edges from text characters form closed contours even in the presence of reasonable levels of noise. Closed contour linking and candidate text line formation are two additional features of the described approach. A candidate text line classifier is applied to further screen out false-positive text identifications. Candidate text regions for placement of text in the natural scene of the electronic image are highlighted and presented to a user.

Abstract: Conversion of text-based images to vector graphics (VG) is disclosed. The text-based images may include images of equations, custom typefaces, or other types of text that may not be included in a font selection of an optical character recognition (OCR) device or an application stored on a viewing device. A textual image may be converted from a raster graphics (RG) image to a VG image, which may enable resizing and alignment of the VG image with body text. In some aspects, the server may determine a body size of a reference character in the VG image. The server may determine a baseline of the VG image that may be used to align the image with the body text.

Abstract: An image forming device performs functions including: dividing the image into a plurality of band images each including at least one sub-region; creating region data used to identify each sub-region included in the band image; updating the region data such that the region data identifies both first and second uniform sub-regions as a single uniform region of the image when the first uniform sub-region abuts the second uniform sub-region, the first and second uniform sub-regions being included in the first and second band image and classified as the uniform sub-region, respectively; and updating the region data such that the region data identifies both first and second nonuniform sub-regions as a single nonuniform region of the image when the first nonuniform sub-region abuts the second nonuniform sub-region, the first and second nonuniform sub-regions being included in the first and second band image and classified as the nonuniform sub-region, respectively.

Abstract: Some embodiments provide a for analyzing a document that includes a number of primitive elements. The method identifies boundaries between sets of primitive elements and identifies regions bounded by the boundaries. The method uses the identified regions to define structural elements for the document. The method defines a structured document based on the primitive elements and the structural elements.