Abstract: A similarity calculation unit (213) treats each of a plurality of classifications as a target classification, and calculates a similarity between a reference feature value, which is a feature value extracted from image data of the target classification, and a recognition feature value, which is a feature value extracted from recognition target data, which is image data to be recognized. An influence calculation unit (214) calculates an influence on the similarity with regard to each partial image of the recognition target data by taking as input the similarity with regard to each of the classifications and the recognition feature value. A feature identification unit (215) changes the recognition feature value in accordance with the influence.

Abstract: The present application discloses a high-precision mapping method and device, which relates to the field of autonomous driving. A specific implementation includes: acquiring global initial poses of multiple point clouds, where the point clouds are point clouds of a location for which a map is to be built and are collected by a lidar using a multi-circle collection mode; dividing the multiple point clouds into multiple spatial submap graphs according to a spatial distribution relationship of the multiple point clouds; optimizing, for each spatial submap graph, global initial poses of point clouds belonging to the spatial submap graph to acquire global poses of the point clouds in each spatial submap graph; and stitching the multiple spatial submap graphs together according to global poses of the point clouds in the multiple spatial submap graphs to acquire a base graph of the map to be built.

Abstract: Embodiments relate to automatic perspective and horizon correction. Generally, a camera captures an image as an image file. Capture-time orientation data from one or more sensors is used to determine the camera's attitude with respect to a defined reference frame. The orientation data and/or attitude can be registered into metadata of the image file and used to generate axis lines representative of the camera's reference frame. A reference line such as a horizon can be automatically identified from detected line segments in the image that align with one of the axis lines within a predetermined angular threshold. The reference line can be used to generate a camera transformation from a starting orientation reflected by the camera attitude to a transformed orientation that aligns the reference line with the reference frame. The transformation can be applied to the image to automatically correct perspective distortion and/or horizon tilt in the image.

Abstract: A line-detection system computes, using a local memory, a result of a partial conversion of image-space pixel data from image space to Hough space. The result is analyzed for edges corresponding to a line present in the partial conversion. The line is compared against other lines detected in previously computed partial results to identify a longest line in the image.

Abstract: Applicants have discovered a multi-layer quality control/quality assurance system that provides higher quality data, without human intervention, by rejecting images that do not achieve a pre-determined quality threshold. In some embodiments of the present invention, there is provided a new method of processing an image from a data set through a pipeline wherein quality control/assurance allows to determine a quality of the image processing, the method comprising; receiving a test image; pre-processing the test image; registering the test image to a reference image; calculating a test image quality using a correlation of image intensity values between corresponding locations of the test image and the reference image; providing a plurality of training images and calculating training image quality distribution statistics for the training images with respect to the reference image; and relating the test image quality to the training image quality distribution.

Abstract: An embodiment is a computer-implemented method for detecting a straight line in a digital image comprising a plurality of pixels comprising the steps: detecting an edge in the digital image, generating a first straight line which passes through a first pixel of the detected edge, generating a second straight line which passes through a second pixel of the detected edge, which is different from the first pixel, determining at least two intersections with a boundary of the digital image for each generated straight line, determining a set of two parameter values for each generated straight line based on the respective determined at least two intersections, wherein the set of two parameter values uniquely determines the respective generated straight line, and detecting the straight line in the digital image based on the determined sets of two parameter values.

Abstract: In a particular embodiment, a method includes receiving line segment data at a processing core. The line segment data is associated with multiple candidate line segments associated with a first group of pixels of an image. The line segment data includes an angle value and/or a distance value for each of the multiple candidate line segments. The method further includes identifying, at the processing core, a set of line segments of the multiple candidate line segments by comparing angle values and/or distance values associated with the multiple candidate line segments. The method also includes determining, at the processing core, a representative line segment based on the set of line segments of the multiple candidate line segments. The method further includes storing, by the processing core, line segment information based on the representative line segment.

Abstract: Methods for drawing a rotated rectangle on an electronic device are provided. In one aspect, a method includes receiving content configured for display and analyzing the content to identify a skew angle associated with the content. The method also includes receiving an input request to draw a rectangle overlaid upon the content, and providing, for display, the rectangle overlaid upon the content based upon the input request, the rectangle rotated according to the skew angle. Systems and machine-readable media are also provided.

Abstract: An image processing apparatus in the present disclosure includes a color conversion processing unit, a modulation candidate creating unit, a color value modulating unit, and an output unit. The color conversion processing unit performs color conversion from an input color value of a target image to a corresponding output color value to obtain a tentative output color value. The modulation candidate creating unit creates, as modulation candidates, combinations of integer values obtained by rounding up or down the decimal fraction of each color component of a tentative output color value. The color value modulating unit selects a modulation candidate at random and performs modulation by replacing the tentative output color value with the selected modulation candidate. The output unit outputs the target image according to the color value obtained after the modulation.

Abstract: An image retrieval method includes reading from a first database a first image group mapped to an object as a retrieval target, and extracting an image feature quantity of a plurality of images of the first image group. A second image group mapped to an object as a search key is read from a second database and an image feature quantity of a plurality of images of the second image group is extracted. Similarity is determined between the first image group and the second image group, based on the image feature quantity of the plurality of images of the first image group and the image feature quantity of the plurality of images of the second image group. On an output device, as retrieval results, the plurality of images of the first image group are displayed in order of similarity of the object as the retrieval target based on determined similarity.

Abstract: There is provided an information processing apparatus including a selecting unit that selects figures from a candidate figure group based on recognition values obtained through character recognition with respect to an input image, and a display control unit that performs control to display the figures selected by the selecting unit.

Abstract: A computerized method for identifying a document. A signature may be determined for a first document and compared with a signature for each of one or more additional documents. A document similarity score may be determined and one or more similar documents may be identified based on the document similarity score.

Abstract: A predetermined feature point obtained from an input image is extracted. An image that indicates a locus specifying a predetermined graphic included in the input image and corresponds to a feature point is acquired using a Hough transform. A recognition target object is detected from an input image, based on a plurality of feature quantities, using an identifier generated by statistical learning using the plurality of feature quantities obtained from a locus image obtained based on a learning image including the recognition target object and a locus image obtained based on a learning image including no recognition target object.

Abstract: Device (10) for vehicle driving evaluation comprising: A means (11) to obtain at least one physical parameter whereby it possible at any time to determine the value of the speed and instantaneous acceleration of a traveling vehicle; A calculation and comparison unit (12) whereby it is possible, from said physical parameter, to calculate an effective parameter that depends on said instantaneous acceleration and to compare said effective parameter with a reference parameter; A driving evaluation unit (13), whereby it is possible to generate a vehicle driving energy score by measuring the variance between said effective parameter and said reference parameter. Corresponding vehicle driving evaluation process.

Abstract: Systems and methods for reconstructing a floorplan of a building for generating a three-dimensional model are provided. One aspect of the present disclosure is directed to a computer-implemented method for generating a three-dimensional model of a building. The method includes estimating a floor height and a ceiling height of the building. The method also includes identifying a core region of a two-dimensional graph, the core region corresponding to an interior of the building. The method includes determining a solution path that circumnavigates the core region and minimizes a cost formula, the cost formula providing an edge cost for each of a plurality of edges. The method further includes generating a three-dimensional model of the interior of the building based on the floor height, the ceiling height, and the solution path.

Abstract: The present invention provides a method and device for acquiring real-time video images of a terminal, sad method comprising: acquiring light intensity of an existing environment when the terminal acquires the real-time video images; comparing the light intensity of the existing environment with a light intensity threshold value set by the terminal system; acquiring images preset by the terminal and sending the preset images to an existing receiver which is communicating with the terminal if the light intensity of the existing environment is lower than the light intensity threshold value set by the terminal system. The present invention can ensure the quality of video conversation when the light intensity is weak, save power energy, improve the user experience, and is beneficial to promote the video communication terminal.

Abstract: A method for analyzing seismic data by generating a post-migration common image gather in a dip angle domain from measured seismic data; detecting concave features related to reflection events in the common image gather and apexes; filtering out part of the concave features in the common image gather in a vicinity of the detected apexes; applying a hybrid Radon transform to the filtered common image gather to separate residues of the concave features from other image features related to diffraction events; and applying an inverse hybrid Radon transform to an image containing the separated features related to diffraction events to obtain a transformed common image gather in the dip angle domain.

Abstract: There are provided an environment recognition device and an environment recognition method. the environment recognition device retains beforehand shape information that is information on a shape of a specific object; obtains a luminance of each of target portions, formed by dividing a detection area, and extracting a target portion including an edge; obtains a relative distance of the target portion including an edge; and determines a specific object indicated with the shape information by performing a Hough transform on the target portion having the edge based on the shape information according to the relative distance.

Abstract: A content-adaptive edge and detail enhancement method is described for image/video processing. Both 2D peaking and LTI/CTI are used in sharpening pictures. Image analysis is performed to generate a mask to control the use of the two peaking techniques. The strength or likelihood of edges or transitions is measured and such a strength or likelihood measurement will be transformed into a blending factor controlling the blending of the LTI/CTI and peaking outputs.

Abstract: An image processing device for identifying a characteristic of an eye from a face image comprising: a first differentiation unit configured to differentiate an eye region in at least a vertical direction of the eye to obtain a first luminance gradient; a first edge extraction unit configured to extract a first edge point according to the first luminance gradient; and a curve identification unit configured to identify a curve, which is a B-spline curve or a Bezier curve expressed by a control point and both end points and fits to the first edge point, as a curve expressing an upper-eyelid or lower-eyelid outline, the end points being an inner corner point of eye and a tail point of eye, by voting for the control point that is a voting target with respect to the first edge point using the Hough transform.

Abstract: A video processor is described, which is useful for implementing a forward transform process, in compliance with the H.264 standard. The video processor includes an input, for receiving a block of image data. The image data is loaded into an internal register. In response to receiving a SIMD instruction, a multiplier, which incorporates the H.264 forward transform matrix in its associated hardware, processes the block of image data, and writes the resulting partially transformed pixel data back to the internal register, transposing the data during the process.

Abstract: If a Hadamard matrix HN of order N=BF is a Kronecker product HF HB of an order F Hadamard matrix and an order B Hadamard matrix, then transformation by HN may be implemented by a fast Hadamard transform at coarse scale followed by fast Hadamard transforms at fine scale. Alternatively, transformation by HN may be achieved by performing order B transforms on columns of a two-dimensional array and order B transforms on rows of the array. As another alternative, transformation by HN may be achieved by computing intermediate values based on linear combinations of input elements and then computing linear combinations of the intermediate values. For compressive signal acquisition, any row of HN may be generated by concatenating selectively modified copies of a corresponding row of HB. Thus, modulation patterns may be generated on the fly.

Abstract: A video processor is described, which is useful for implementing a Hadamard transform process, in compliance with the H.264 standard. The video processor includes an input, for receiving a block of image data. The image data is loaded into an internal register. In response to receiving a SIMD instruction, a multiplier, which incorporates the H.264 Hadamard transform matrix in its associated hardware, processes the block of image data, and writes the resulting partially transformed pixel data back to the internal register, transposing the data during the process.

Abstract: An apparatus and method to detect a line segment or arc using Hough transform. A Hough transform unit performs contour extraction on brightness image data to generate contour image data, with pixels having a pixel value of 0 to 255, performs the Hough transform on points in the contour image data, and counts additional values represented by pixel values of points in the contour image data in a Hough table. The Hough transform unit performs contour extraction on first to third component data to generate first to third contour data with pixels having a pixel value of 0 to 255, performs the Hough transform on points in the first to third contour data, and counts additional values represented by pixel values of points in the first to third contour data in the Hough table. The detection unit comprehensively evaluates the counts to detect a line segment or arc.

Abstract: A method of retrieving information comprised in a barcode is disclosed. The method comprises detecting that the barcode is present in a first image having a first image quality and capturing a first region, acquiring, when it is detected that the barcode is present, a second image having a second image quality and capturing a second region, wherein the second image quality is higher than the first image quality, and wherein the second region at least partly overlaps the first region, and decoding the barcode based on the second image to retrieve the information. A corresponding program product and a corresponding arrangement are also disclosed along with a communication device comprising the arrangement.

Abstract: A Hadamard transform-based image compression method includes performing a Hadamard transform on 2k pixel values according to a product of a 2k×2k stage Hadamard matrix and a 2k×2k adjustment matrix to generate 2k conversion values, where k is a positive integer and at least one of the 2k conversion values is zero. The adjustment matrix satisfies a condition that: when the 2k pixel values are divided into G pixel groups each comprising 2k/G adjacent pixels values and the adjustment matrix is multiplied with a first 2k×1 matrix formed by the 2k pixel values to transform the first 2k×1 matrix to a second 2k×1 matrix, each pixel value of the first matrix is transformed to an average of a pixel group comprising the pixel value to form the second 2k×1 matrix.

Abstract: In some embodiments, a method for performing and a system configured to perform a 2D transform (for example, an inverse discrete cosine transform) on each block of a sequence of data blocks, where the 2D transform includes a row transform and a column transform. To perform the row or column transform on a row or column of data, these embodiments determine whether each of different subsets of the data values comprising a partition of the row (column) includes at least one zero value, whether each of different subsets of a first subset of the partition includes at least one zero value, and whether each of different subsets of at least one other subset of the partition includes at least one zero value.

Abstract: A method of deriving a representation of an image by processing signals corresponding to the image, the method comprises deriving a function of the image, where a function of a translated, scaled or rotated version of the image is a translated or scaled version of the derived function of the image, and using a plurality of frequency components of a frequency representation of the function to derive a representation of the image.

Abstract: A method for estimating three-dimensional structure from a two-dimensional image (502) includes obtaining first and second vanishing points (120, 122); comparing image patches (202) along first and second virtual lines (204, 208) extending from the first and second vanishing points (120, 122), respectively, and through a pixel; generating values for each of the first and second virtual lines (204, 208) based on the comparison of the image patches (202); accumulating the values for each the pixel in the two-dimensional image (502); and determining a corner pixel (106) based on a highest of the accumulated values.

Abstract: An apparatus for recognizing gestures in a picture includes a Hough transformer configured to identify elements in the picture or in a pre-processed version of the picture as identified gesture elements and to obtain information about the identified gesture elements. The apparatus further includes a gesture description creator configured to obtain a gesture description while using the information about the identified gesture elements. Moreover, the apparatus includes a gesture classifier configured to compare the gesture description to a plurality of comparative gesture descriptions having gesture codes associated with them. The gesture classifier is configured to provide, as the result of the comparison, a gesture code of a recognized gesture.

Abstract: A video processing technique is performed on a video media service to identify video segments of interest. The video processing technique (200) is supplemented by a text extraction technique (245, 255, 270), as well. The resulting video segments of interest (289) can be stored as to produce a version of the media service which is shorter in time length than the original video media service.

Abstract: The present invention relates to an image processing apparatus for compressing image data used in an image generating apparatus for generating a free-viewpoint image. According to the invention, the apparatus has a selecting unit that selects one image as a first image, and defines other images as second images, a projective transformation unit that generates a projected depth map of a second image from a depth map of the first image, a subtracting unit that creates a difference map of the second image, and a storage unit that stores the depth map of the first image and the difference map of the second image. Here, the difference map is a difference between a depth map of the second image and the projected depth map of the second image, and the depth map indicates a depth value of each pixel of a corresponding image.

Abstract: A device for detecting characters in an image includes a Hough transformer implemented to identify, as identified elements of writing, circular arcs or elliptical arcs in the image or in a preprocessed version of the image. The device further includes a character description generator implemented to obtain, on the basis of the identified circular arcs or elliptical arcs, a character description which describes locations of the identified circular arcs or elliptical arcs. In addition, the device includes a database comparator implemented to compare the character description with a plurality of comparative character descriptions which have character codes associated with them, so as to provide, as a result of the comparison, a character code of a detected character.

Abstract: A method of navigating a space vehicle. An image of a planet surface is received. The received image is processed to identify edge pixels and angle data. The edge pixels and angle data are used to identify planetary features by shape, size, and spacing relative to other planetary features. At least some of the planetary features are compared with a predefined planet surface description including sizes and locations of planet landmarks. One or more matches are determined between the planetary feature(s) and the planet surface description. Based on the match(es), a location of the space vehicle relative to the planet is determined.

Abstract: Methods and apparatus, including computer program products, implementing and using techniques for document authentication. An electronic document is presented to a user. The electronic document has data representing a signed state and a current state. A disallowed difference between the signed state and the current state is detected, based on one or more rules that are associated with the electronic document. A digital signature associated with the electronic document is invalidated in response to the detecting.

Abstract: A method for converting a portion of an image from a first domain to a second domain. The method may apply a Hough transform on the converted portion of the image, including calculating a range of angles for each tested pixel q relative to a center pixel p, quantizing the range of angles into a plurality of bins, voting each tested pixel q using a range of bins using a weighted voting schema; and detecting one or more features in the portion of the image. The methods may be implemented by program instructions executing in parallel on CPU(s) or GPUs.

Abstract: There is provided an information processing apparatus according to the present invention including a Hough transform unit executing the Hough transformation on a biometrics image which is image information unique to a living body, and a block detection unit detecting whether a block having a predetermined shape is included in a Hough space image which is a Hough-transformed biometrics image, outputting present position information representing a position at which the block is located in the Hough space image when the block is determined to be included in the Hough space image.

Abstract: In a method and apparatus for difference measurement of an image, the image is detected by an image sensor array having (N×N) sensor elements, wherein N=2n, n?2, and n being an integer number, each detected sample value is read as a positive or as a negative sample value according to a sign control pattern from a pattern control unit, and measurement values are generated for measurement template blocks of the image sensor array by a computing unit which computes for each measurement template block a weighted sum of the read sample values of the respective measurement template sensor elements depending on a predetermined measurement template. A measurement template block is, provided for at least four adjacent sensor elements of the image sensor array. The method/apparatus can be used in any kind of digital camera and allows for a fast generation of image data with a hardware implementation of low complexity.

Abstract: Embodiments of the invention compress ultrasound RF data after the receiver beamformer. An efficient compression algorithm is disclosed that incorporates the use of the Discrete Cosine Transform (DCT) and the Discrete Wavelet Packet (DWP) transform, followed by quantization of the wavelet coefficients. The algorithm first processes ultrasound scanlines from the receiver beamformer using a DCT transform. In a low-power configuration, a Hadamard transform may be used instead of the DCT. The output of the DCT processing is DCT coefficients at different frequencies. An optimized wavelet packet transform is then used for each coefficient line. The wavelet filter kernels and the wavelet packet trees can be jointly optimization using a two-step optimization algorithm.

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: There is provided a matching method, capable of positioning with higher accuracy at higher speed. The method includes the steps of: constructing a pattern model of a registered image in which a plurality of reference points are set on an extracted contour and a corresponding point search line having a predetermined length and passing through each reference point; acquiring an image and acquiring an initial position corresponding to the registered image to arrange the corresponding point search line of the pattern model; finding a corresponding point on the image corresponding to each reference point with regard to each corresponding point search line at a position along the corresponding point search line on the image; and regarding a relation between each reference point and the corresponding point as an evaluation value and performing fine positioning such that an accumulated value of the evaluation values becomes minimal or maximal.

Abstract: To provide a registration apparatus, a collation apparatus, an extraction method and an extraction program capable of improving authentication accuracy. An image signal obtained by imaging a biometric subject in a predetermined biological position is subjected to predetermined processing so as to extract a feature portion of the biometric subject in the image signal. The feature portion is subjected to Hough transform. Parameter extraction is performed by changing the extraction condition such that the number of parameters obtained by the Hough transform becomes a predetermined value. In the case where the value of the extraction condition at the time point when the number of parameters is the predetermined value falls within a range from the upper limit set value to the lower limit set value, the predetermined number of parameters are set as registration data or data to be collated with the registration data.

Abstract: A method for the extraction of Lunar data and/or planetary features is provided. The feature extraction method can include one or more image processing techniques, including, but not limited to, a watershed segmentation and/or the generalized Hough Transform. According to some embodiments, the feature extraction method can include extracting features, such as, small rocks. According to some embodiments, small rocks can be extracted by applying a watershed segmentation algorithm to the Canny gradient. According to some embodiments, applying a watershed segmentation algorithm to the Canny gradient can allow regions that appear as close contours in the gradient to be segmented.

Abstract: The invention relates to a device and method for detecting changes in background of successive images. More particularly, the invention relates to a device and method for detecting changes in background of successive images by obtaining vertically accumulated values and horizontally accumulated values from multiple binary images of previous image and current image, obtaining pairs of rows and pairs of columns based on the vertically accumulated values and horizontally accumulated values, and transforming pairs of rows and pairs of columns using Hough transformation. Also, the invention relates to an interface system including background changes detecting device and display device connected to the background changes detecting device.

Abstract: A registration device, verification device, authentication method and authentication program that can improve the accuracy of authentication are proposed. A predetermined process is performed for an image signal obtained as a result of taking a picture of a image-capturing target which is given as an object for biometrics authentication and which is a predetermined part of a living body; a characteristic part of the image-capturing target is extracted from the image signal; the Hough transform is carried out by characteristic extraction means for the extracted characteristic part; a plurality of characteristic parameter points are extracted from a parameter point obtained as a result of the Hough transform under a predetermined extraction condition; and a determination is made as to whether the plurality of characteristic parameter points are those to be registered or to be compared with the registered one according to an angle component of the plurality of the characteristic parameter points.

Abstract: A reading machine that operates in various modes includes image correction processing is described. The reading device pre-processes an image prior to optical character recognition processing by detect distortion in an image of a page by measuring an extent to which page boundaries in the image deviate from a simple rectangular shape and correcting for the optical distortion by transforming the image to restore the page to a rectangular shape.

Abstract: A method for converting a portion of an image from a first domain to a second domain. The method may apply a Hough transform on the converted portion of the image, including calculating a range of angles for each tested pixel q relative to a center pixel p, quantizing the range of angles into a plurality of bins, voting each tested pixel q using a range of bins using a weighted voting schema; and detecting one or more features in the portion of the image. The methods may be implemented by program instructions executing in parallel on CPU(s) or GPUs.

Abstract: A vehicle travel support device determines presence of a recognition inhibiting factor of a lane mark on a road on which a vehicle is traveling with high accuracy irrespective of an imaging history by a vehicular camera from the same position. The vehicle travel support system generates an edge image by extracting an edge or actualizing an edge in an image obtained through the vehicular camera. When Hough transform of the edge image is performed, votes for a specified vote value of a linear component is evaluated in a ?-? space (Hough space). Presence of a recognition inhibiting factor of a lane mark on a road is determined by determining whether or not the votes of a specified vote value in a specified region denoting a standard travel lane of a vehicle in the real space is ?a threshold in the ?-? space.

Abstract: Disclosed is a method and system for simultaneously acquiring and producing results for multiple image modes using a common sensor without optical filtering, scanning, or other moving parts. The system and method utilize the Walsh-Hadamard correlation detection process (e.g., functions/matrix) to provide an all-binary structure that permits seamless bridging between analog and digital domains.

Abstract: Objects having a flat surface such as a table are detected by processing a depth image and a color image. A mask indicating an area likely to include an object having the flat surface is generated by processing a depth image including the depth information. A color image corresponding to the depth image is then cropped using the mask to detect a portion of the color image that likely include the object having the flat surface. Geometric features of the cropped color image such as lines are then detected to determine the location and orientation of the object having the flat surface. A subset of the detected geometric features is selected as outlines of the flat surface.