Abstract: In some aspects, a debris detector includes one or more processors and a memory storage device to store instructions executable by the one or more processors to: receive a frame captured by a camera, receive a subsequent frame captured by the camera, compare the frame to the subsequent frame to determine a difference, and based on determining that the difference is greater than a threshold, send a notification indicating that debris has been detected. The notification may include location data identifying a location where the debris is located. Receiving the notification may cause a robot vacuum to initiate cleaning the location identified in the notification.

Abstract: An image processing method, an image processing device, an electronic device, and a storage medium are provided. The image processing method includes: obtaining an input image, wherein the input image includes M character rows; performing global correction processing on the input image to obtain an intermediate corrected image; determining the M character row lower boundaries; determining the relative offset of all pixels in the intermediate corrected image according to the M character row lower boundaries, the first image boundary and the second image boundary of the intermediate corrected image; determining the local adjustment offset of all pixels in the intermediate corrected image according to the relative offsets of all pixels in the intermediate corrected image; and performing local adjustment on the intermediate corrected image according to the local adjustment offsets of all pixels in the intermediate corrected image to obtain the target corrected image.

Abstract: A magnetic disk device includes a plurality of magnetic heads, a storage unit that stores a table storing, for each of the magnetic heads, a track skew value obtained by adding a head shared value and a head unique value, and a control unit that reads out the track skew value corresponding to one of the magnetic heads to be operated from the table and performs seek control.

Abstract: An information processing apparatus includes a processor configured to acquire a result of character recognition of a character string formed on a medium and read by scanning that is subject to character recognition and replace a character or a symbol in a subject with a reference character string that is referred to by the character or the symbol.

Abstract: Disclosed herein are various techniques for more precisely and reliably (a) positioning top and bottom border edges relative to textual content, (b) positioning left and right border edges relative to textual content, (c) positioning mixed edge borders relative to textual content, (d) positioning boundaries of a region of background shading that fall within borders of textual content, (e) positioning borders relative to textual content that spans columns, (f) positioning respective borders relative to discrete portions of textual content, (g) positioning collective borders relative to discrete, abutting portions of textual content, (h) applying stylized corner boundaries to a region of background shading, and (i) applying stylized corners to borders.

Abstract: Aspects of the present disclosure provide techniques for document classification through signal processing. Embodiments include receiving a document for classification. Embodiments include generating an image of the document. Embodiments include producing a signal representation of the document based on numbers of non-white pixels in each horizontal scan line or vertical scan line of the image of the document. Embodiments include comparing the signal representation of the document to signal representations of previously-classified documents. Embodiments include determining, based on the comparing, a classification for the document. Embodiments include performing additional processing with respect to the document based on the classification for the document.

Abstract: An image processing apparatus includes a determination unit configured to determine a region of the image on which to perform character recognition processing, a decision unit configured to decide, based on a number of black pixels in contact with the region determined by the determination unit, whether to perform the character recognition processing on an expanded region obtained by expanding the region determined by the determination unit rather than on the region determined by the determination unit, and a character recognition unit configured to perform the character recognition processing on that region of the image decided by the decision unit.

Abstract: Systems and methods for automatic enrollment and identity verification based upon processing a captured image of a document are disclosed herein. Various embodiments enable, for example, a user to enroll in a particular service by taking a photograph of a particular document (e.g., his driver license) with a mobile device. One or more algorithms can then extract relevant data from the captured image. The extracted data (e.g., the person's name, gender, date of birth, height, weight, etc.) can then be used to automatically populate various fields of an enrollment application, thereby reducing the amount of information that the user has to manually input into his mobile device in order to complete the enrollment process. In some embodiments, a set of internal and/or external checks can be run against the data to ensure that the data is valid, has been read correctly, and is consistent with other data.

Abstract: Disclosed are a training method and apparatus for a CNN model, which belong to the field of image recognition. The method comprises: performing a convolution operation, maximal pooling operation and horizontal pooling operation on training images, respectively, to obtain second feature images; determining feature vectors according to the second feature images; processing the feature vectors to obtain category probability vectors; according to the category probability vectors and an initial category, calculating a category error; based on the category error, adjusting model parameters; based on the adjusted model parameters, continuing the model parameters adjusting process, and using the model parameters when the number of iteration times reaches a pre-set number of times as the model parameters for the well-trained CNN model. After the convolution operation and maximal pooling operation on the training images on each level of convolution layer, a horizontal pooling operation is performed.

Abstract: The present disclosure relates to method and system for automatically identifying one or more issues in one or more tickets of an organization. An issue identification system retrieves a sequence pattern from ticket data received from one or more data sources. The issue identification system generates one or more first sub-sequence patterns of the n-grams from the sequence pattern. Further, frequency of occurrence and Part-of-Speech (POS) weightage of each of the one or more first sub-sequence patterns of the n-grams are determined by the issue identification system. A first score is determined for each of the one or more first sub-sequence patterns of the n-grams based on both the frequency and the POS weightage. Upon determining the first score, the issue identification system identifies one or more issues in the one or more tickets automatically based on the first sub-sequence pattern of the n-grams associated with a highest first score.

Abstract: Systems and methods for mobile image data capture and processing are disclosed. The techniques encompass receipt or capture of digital image data, detecting an object such as a document depicted in a digital image corresponding to the digital image data, processing the digital image to improve image quality, classifying the object from the processed image data, and extracting useful information from the object. Processing may improve image quality by correcting artifacts such as distortion, skew, blur, shadows, etc. common to digital images captured using mobile devices. Classification is based on identifying unique features (and/or combinations thereof) within the image data and determining whether the identified features indicate the object belongs to a class of known objects having similar characteristics, or is unique to all known classes. Extraction is based in whole or in part on object classification. All operations may be performed using mobile technology exclusively.

Abstract: An image processing method of processing a gray-scale image of character strings arranged in a plurality of lines of text, for character recognition, includes setting at least a part of a range, in which the character strings in the gray-scale image area are distributed, as a search area, and setting, for each of a plurality of points arranged along one of two axes, representing coordinates in the image, corresponding to an arrangement of the line of text, a plurality of lines extending across the search area, with the point as a starting point, and accumulating, for each of the lines, pixel data in the line within the search area.

Abstract: A image manipulation technique allows a user to correct an image using samples obtained from other images. These samples may be obtained from one or more other images in a library of images. Matching techniques may identify an image that best matches the image to be corrected, or may aggregate or average multiple images that are identified as containing an area corresponding to the area to be corrected. Identification of the image or images to use as the source of the samples may be automatic or manual. The images may be from a library of images under the control of the user or from a library of images maintained by another person or service provider. Application of the samples to correct the image may be manually or automatically directed.

Abstract: A method of adjusting sampling precision of a navigation device is disclosed in the present invention. The sampling precision represents counts per inch (CPI) or dots per inch (DPI) of the navigation device. The method includes determining a predetermined mode of the navigation device, obtaining resolution of a display, and adjusting the sampling precision according to the resolution and the predetermined mode, so that the sampling precision of the navigation device can be accordingly increased and decreased due to variation of the resolution.

Abstract: Systems and methods for enhanced object detection are provided in this disclosure. One embodiment of a method includes providing front side illumination of an object, capturing a first image of the object, and providing backside illumination of the object. Some embodiments include capturing a second image of the object, and determining an edge of the platform area assembly. Similarly, some embodiments include aligning the object from the second image with the edge of the platform area assembly to determine a location of the object, determining based on the location of the object, a pixel in the first image that does not belong to the object, and modify the pixel from the first image that does not belong to the object to create an altered image.

Abstract: Auto-correcting perspective distortion in document images is described. Line segments are identified in a distorted document image. Line segments are combined to form candidate lines. Candidate lines are grouped into a substantially horizontal group of lines and a substantially vertical group of lines. Linear equations are calculated to fit the substantially horizontal group of lines. Linear equations are calculated to fit the substantially vertical group of lines. The distorted document image is transformed based on two linear equations associated with two substantially horizontal lines and two linear equations associated with two substantially vertical lines.

Abstract: A method and/or system for screenshot orientation detection may include performing an initial optical character recognition (OCR) and/or an initial face recognition technique on a screenshot of an application. A determination of whether the screenshot orientation is correct may be made based on, for example, the initial OCR and/or the initial face recognition technique. In an event when the screenshot orientation is not correct, a determination of a correct screenshot orientation may be made. In this regard, the screenshot may be rotated (e.g., by a predetermined number of degrees). A subsequent OCR and/or a subsequent face recognition technique may be performed on the rotated screenshot. A determination may be made whether the screenshot orientation of the rotated screenshot is correct based on, for example, the subsequent OCR and/or the subsequent face recognition technique.

Abstract: Aspects of the present invention are related to systems and methods for determining a skew angle associated with a document image. According to a first aspect of the present invention, a rotation vector may be estimated for at least one layer in a vertical-edge buffer and a horizontal-edge buffer. According to a second aspect of the present invention, a rotation vector may be estimated directly from the vertical-edge buffer and the horizontal-edge buffer using a fixed-sized, progressively constrained histogram.

Abstract: Implementations relate to relate to rectification of distortion in an image. In some implementations, a method includes extracting edgelets from an image, each edgelet defined by a location of a pixel having an edge depicted in the image and defined by a direction of the edge. The method finds at least one vanishing point in the image based on intersections of pairs of the edges of the edgelets. Each vanishing point is selected based on a plurality of the edgelets in the image having edges approximately aligned with the vanishing point. Found vanishing points are used to transform the image to reduce image distortion.

Abstract: A face detection processing unit performs a face detection process by rotating an image in increments of a predetermined angle and acquires a rotation angle at which a face is detected. An angle correction unit acquires an angle between the face and a shoulder by pattern matching and corrects the rotation angle of the image. A human-image orientation identification unit identifies a correct orientation of a human image based on the rotation angle. An image-characteristic analysis unit analyzes a frequency distribution and a brightness distribution of a non-human image. A non-human image orientation identification unit identifies the correct orientation of a non-human image based on distribution characteristics of a frequency of brightness with respect to an axis in a predetermined direction. An image-data updating unit incorporates information regarding the correct orientation in the image data.

Abstract: Embodiments of the present invention relates to systems, computer-implemented methods, and computer program products for capturing and storing elements of a negotiable instrument for use in image recreation. In some embodiments, a method is provided that includes: (a) receive an image of the negotiable instrument, wherein the image of the negotiable instrument comprises one or more elements that are used for processing the negotiable instrument and non-element portions that are not used in processing the negotiable instrument; (b) capture, using the image capture device, images of one or more elements of the negotiable instrument; (c) store the images of the one or more elements of the negotiable instrument in the database; and (d) store, as white space, the non-element portions of the image of the negotiable instrument in the database.

Abstract: A system and method using a text extraction application for identifying words with multiple orientations from an image are described. The text extraction application receives an input image, generates progressively blurred images, detects blobs in the blurred images, outputs ellipses over the blobs, detects a word in the input image, orients and normalizes a first version of the word, generates an inverted version of the word, performs OCR on the first version and the inverted version of the word, generates confidence scores for the first version and the inverted version of the word and outputs text associated with the word.

Abstract: An image combining device may include an image reading unit that reads an image of a first and second partial area which have a portion overlapping; a template area setting section that sets apart an area of a predetermined shape as a comparison area from the overlapping area; a calculation area setting section that selects areas of the shape corresponding to the shape of the comparison area; a correlation coefficient calculating section that calculates a correlation coefficient between each of the image data values of the multiple areas; an area identifying section that identifies, from the multiple areas, the area in which the calculation result satisfies a predetermined condition; and a combining section that combines the images of the first and second partial areas in the positional relationship in which the area identified by the area identifying section and the comparison area overlap with each other.

Abstract: A method for creating a binary mask image from an inputted digital image of a scanned document, including the steps of creating a binarized image by binarizing the inputted digital image, detecting first text regions representing light text on a dark background, and inverting the first text regions, such that the inverted first text regions are interpretable in the same way as dark text on a light background. A method for comparing in a binary image a first pixel blob with a second pixel blob to determine whether they represent matching symbols, including the steps of detecting a line in one blob not present in the other and/or determining if one of the blobs represents an italicized symbol where the other does not.

Abstract: Various embodiments of the present invention relate to a method, system and computer program product for detecting and recognizing text in the images captured by cameras and scanners. First, a series of image-processing techniques is applied to detect text regions in the image. Subsequently, the detected text regions pass through different processing stages that reduce blurring and the negative effects of variable lighting. This results in the creation of multiple images that are versions of the same text region. Some of these multiple versions are sent to a character-recognition system. The resulting texts from each of the versions of the image sent to the character-recognition system are then combined to a single result, wherein the single result is detected text.

Abstract: Method, system, device and computer program product for determining vanishing point candidates of a text portion in an image document distorted by perspective. The method includes the steps of image binarization, connected component analysis, estimating a number of text lines in a Cartesian coordinate system, transforming the text lines to data points in a homogenous coordinate system, assigning a confidence level to the data points, grouping a number of data points into a priority sample array, clustering the data points in the priority sample array into a number of sample groups and assigning a group confidence value to each sample group. A RANSAC algorithm is applied to determine among the data points a set of inliers, initiated with the sample group having the highest group confidence value. A vanishing point candidate is determined from the text lines corresponding to the set of inliers.

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: A reading machine that operates in various modes includes image correction processing is described. The reading device pre-processes an image for optical character recognition by receiving the image and determining whether text in the image is too large or small for optical character recognition processing by determining that text height falls outside of a range in which optical character recognition software will recognize text in a digitized image. If necessary the image is resized according to whether the text is too large or too small.

Abstract: An electronic device and method creates a part coordinate system of a three-dimensional (3D) product model based on dimensional information in a two-dimensional (2D) drawing of the product, determines transformation matrices of the part coordinate system relative to a machine coordinate system of the measuring machine, and automatically adjusts the orientation of the product model in the machine coordinate system, to align the part coordinate system with the machine coordinate system.

Abstract: A presentation document is repurposed to save paper and ink. An indication that a printout of a presentation document from a printer has been requested is received. The presentation document has content with a format that would result in a first quantity of pages being printed. The content has a plurality of slides and a plurality of shapes. Further, the presentation document is preprocessed to acquire data associated the presentation document. Group shaping is also performed to group the plurality of shapes into one or more groups based on shapes that have to be transformed together. Further, content transformation is performed to generate transformed content. In addition, a content re-layout is performed to determine placement of the transformed content.

Abstract: A “Text Rectifier” provides various techniques for processing selected regions of an image containing text or characters by treating those images as matrices of low-rank textures and using a rank minimization technique that recovers and removes image deformations (e.g., affine and projective transforms as well as general classes of nonlinear transforms) while rectifying the text or characters in the image region. Once distortions have been removed and the text or characters rectified, the resulting text is made available for a variety of uses or further processing such as optical character recognition (OCR). In various embodiments, binarization and/or inversion techniques are applied to the selected image regions during the rank minimization process to both improve text rectification and to present the resulting images of text to an OCR engine in a form that enhances the accuracy of the OCR results.

Abstract: An image reading device includes a conveying unit, reading unit, detecting unit, and control unit. The conveying unit conveys two sheets of the original one by one. The reading unit reads a sheet passing past a reading position. The detecting unit detects a sheet passing past a detection position upstream of the reading position in a conveying direction. The control unit controls the reading unit to start a reading process when a leading edge of sheet reaching a position upstream of the reading position by a first distance, controls the reading unit to stop the reading process when a trailing edge of sheet reaching a position downstream of the reading position by a second distance, determines an interspace between the trailing edge of sheet and a leading edge of a subsequently-conveyed sheet, and updates the first distance and the second distance based on the interspace.

Abstract: A device includes a character-data rotating section that rotates a regular-position character by a predetermined angle with respect to a reference point that is the center point of the background area of the regular-position character by using regular-position character data having a rotation angle of 0° and a center-point matching processing section that horizontally and/or vertically enlarges the background area of the rotated character data to cause the center point of the rotated character and the center point of BMP data to match each other even with respect to rotated character data. Thus, when multiple pieces of character data are arranged so that the center points thereof lie on a reference line, not only are the center points of the characters aligned along the reference line, but also bottom portions of the characters aligned with respect to the reference line.

Abstract: In one embodiment, a method of detecting document orientation includes capturing a document image, binarizing each subimage of the document image to retain textual content and eliminate graphic and noise content from the document image, detecting portrait or landscape orientation based on values computed from strip-based projection profiles, and detecting up or down text orientation based on a text-asymmetry ratio computed from strip-based projection profiles.

Abstract: A first aspect of the invention relates to a method for creating a binary mask image from an a inputted digital image of a scanned document, comprising the steps of creating a binarized image by binarizing the inputted digital image, detecting first text regions representing light text on a dark background, and inverting the first text regions, such that the inverted first text regions are interpretable in the same way as dark text on a light background. A second aspect of the invention relates to a method for comparing in a binary image a first pixel blob with a second pixel blob to determine whether they represent matching symbols, comprising the steps of detecting a line in one blob not present in the other and/or determining if one of the blobs represents an italicized symbol where the other does not.

Abstract: Various embodiments of the invention provide systems and methods for extracting information from digital documents, including physical documents that have been converted to digital documents. For example, some embodiments are configured to extract information from a field in a digital document by identifying a block of tokens before (i.e., a prior block) and a block of tokens after (i.e., a post block) the field from which the information is to be extracted, where both the prior block and post block are known to be associated with the field type of the field (e.g., name, address, phone number, etc.).

Abstract: There is disclosed a method for seamlessly replacing areas in a digital image with corresponding data from temporally close digital images depicting substantially the same scene. The method uses localized image registration error minimization over a fixed preliminary boundary. A least cost closed path which constitutes a boundary for the area to be replaced is calculated using dynamic programming. The replacement area is blended such that image data information from one image is seamlessly replaced with image data information from another image.

Abstract: An image processing apparatus includes: a division section for dividing input image data into portions; an orientation determining section for calculating reliabilities of directions of image data of each portion when the directions are regarded as orientations, and setting an orientation with the highest reliability as an orientation of each portion; a display control section for generating display image data including an image of a target portion whose reliability of an orientation is less than a predetermined value and images of designation regions from which a user's input to designate the orientation of the target portion is entered; and a character recognition section for recognizing characters of each portion in such a manner that the orientation is designated from the designation regions or set by the orientation determining section. This allows prompt recognition of characters of a portion whose reliability of orientation is low, in accordance with a right orientation.

Abstract: Image alignment operations particular suited for High Dynamic Range (HDR) image generation are described. Image pyramids may be generated using reference and source images. Difference bitmaps, based on a number of pixel shift combinations in the x and y directions, can be divided into tiles and analyzed and, for each pyramid level, an optimal shift direction determined. The tiles can then be pruned using a threshold such that only those tiles contributing up to the threshold are projected to a subsequent pyramid level. The alignment vector for each level can be aggregated to determine a final alignment vector which can be used to shift the source image. This process may be repeated for another source image, and the two source images and reference image, once aligned, may be merged to generate an HDR image.

Abstract: An image processor includes a control unit, wherein the control unit includes an image obtaining unit that obtains an image including a document area corresponding to a document read by an image reader, an edge coordinate detecting unit that detects coordinate values of edge candidate pixels, a tilt calculating unit that calculates a tilt of the document area, or a tilt obtaining unit that obtains the tilt of the document area detected by a predetermined sensor, an edge end coordinate extracting unit that extracts a coordinate value of an edge end candidate pixel, a straight line calculating unit that calculates a straight line corresponding to the edge, and either one or both of a tilt correcting unit that corrects the tilt of the document area by rotating the document area, and an area clipping unit that clips the document area from the image.

Abstract: The present invention provides a method for an Optical Character Recognition (OCR) system providing recognition of characters that are partly hidden by crossing outs due to for example an imprint of a stamp, handwritten signatures, etc. The method establishes a set of template images of certainly recognized characters from the image of the text being processed by the OCR system, wherein the effect of the crossed out section is modelled into the template images before comparing these images with the image of a visually impaired crossed out character. The modelled template image having the highest similarity with the visually impaired crossed out character is the correct identification for the visually impaired character instance.

Abstract: Various embodiments create a source image from a web page and then process the source image effective to remove left and/or right border areas so that a reduced-size web page image created from the source image has a better chance of capturing relevant content for a viewer. In at least some embodiments, image processing techniques are utilized to identify repeating patterns of pixels along left and/or right border areas of a source image. The image processing techniques can process individual rows of pixels looking for patterns of pixel color values. Identified patterns of pixel color values are noted and then subsequently used to remove regions of the source image in which repeating patterns occur. Having removed these regions from a source image, the source image can be reduced in a manner directed to improving information density contained in a reduced-size web page image created from the source image.

Abstract: An image processor performing super-resolution of converting an input image with first resolution to an output image with second resolution higher than the first resolution on consecutive input images includes a prediction unit predicting the output image with the second resolution of a current frame using the input image of the current frame and the output image obtained by the super-resolution on an input image of a previous frame; a generation unit generating a reduced image with the first resolution composed of pixels at different phases of the prediction image using a prediction image obtained by the prediction of the prediction unit; a difference calculation unit calculating a difference between the input image of the current frame and the reduced image; and an addition unit adding the difference up-sampled to the second resolution to the prediction image, thus generating the output image with the second resolution of the current frame.

Abstract: The invention relates to a method for securing a first image by means of graphical anti-counterfeiting means and to a method for securing an identification document with such graphical anti-counterfeiting means. The invention also relates to a secure identification document that allows detecting either a fraudulent modification of the existing personalization or a fraudulent falsified document. For that, graphical anti-counterfeiting image is inserted into an identification image, each image being defined by a plurality of pixels. The characteristic level (for example grey level) of each pixel i of the graphical anti-counterfeiting image is linked, by a function F, to a matrix ?i of pixels defined in the identification image, said pixels of the matrix ?i surrounding the location i of a pixel of the graphical anti-counterfeiting image, said function F taking into account the characteristic level (for example average grey level) G(?i) and the texture level T(?i) of said matrix ?i.

Abstract: A method of compensating for distortion in text recognition is provided, which includes extracting a text region from an image; estimating the form of an upper end of the extracted text region; estimating the form of a lower end of the extracted text region; estimating the form of left and right sides of the extracted text region; estimating a diagram constituted in the form of the estimated upper end, lower end, left and right sides, and including a minimum area of the text region; and transforming the text region constituting the estimated diagram into a rectangular diagram using an affine transform.

Abstract: Briefly, in accordance with one or more embodiments, an image-processing system is capable of receiving an image containing text, applying optical character recognition to the image, and then audibly reproducing the text via text-to-speech synthesis. Prior to optical character recognition, an orientation corrector is capable of detecting an amount of angular rotation of the text in the image with respect to horizontal, and then rotating the image by an appropriate amount to sufficiently align the text with respect to horizontal for optimal optical character recognition. The detection may be performed using steerable filters to provide an energy versus orientation curve of the image data. A maximum of the energy curve may indicate the amount of angular rotation that may be corrected by the orientation corrector.

Abstract: The present invention provides a method and a system for image identification and identification result output, which determines a location coordinate with respect to an image and a rotating angle based on at least one direction of the image according to features of the image. The image is compared to a plurality of sample images stored in a database according to the rotating angle so as to obtain at least one identification result. By means of the method and the system of the present invention, identification can be achieved with respect to various rotating angles and distances so as to improve the identification rate.

Abstract: An image generation method for generating image information of a color signal Y of an image A by using a color signal X of image A, and color signal X and color signal Y of an image B. The presence or absence of a point in color signal X of image B corresponding to each pixel position of color signal X of image A, and the position of the relevant corresponding point are estimated. To each estimated pixel position in color signal Y of image A, image information of the corresponding position in the second color signal Y of image B is assigned. Color signal Y at a pixel position in image A for which it is estimated that there is no corresponding point is generated by using the image information of color signal Y assigned to pixels having a corresponding point.

Abstract: An apparatus, method, and system for processing character data is provided, which selects a format of the character data to be used for generating print data. When a user instruction for printing character data according to character command data specifying the output of the character data is received, the format of the character data is selected based on the character command data.

Abstract: Various embodiments of the present invention relate to a method, system and computer program product for detecting and recognizing text in the images captured by cameras and scanners. First, a series of image-processing techniques is applied to detect text regions in the image. Subsequently, the detected text regions pass through different processing stages that reduce blurring and the negative effects of variable lighting. This results in the creation of multiple images that are versions of the same text region. Some of these multiple versions are sent to a character-recognition system. The resulting texts from each of the versions of the image sent to the character-recognition system are then combined to a single result, wherein the single result is detected text.