Abstract: An information processing apparatus includes a self-position/self-orientation calculation unit calculating self-position and/or self-orientation in the predetermined coordinate system based on a marker in acquired imaged image data when it is determined that the marker exists within a predetermined area and the marker is imaged in the imaged image data and based on received position information and physical amounts measured by sensors for measuring the physical amounts to be used for autonomous navigation when the marker does not exist within the predetermined area or the marker is not imaged in the imaged image data.

Abstract: A method for atlas-based segmentation is provided where a patient image is rigidly registered with each of a plurality of atlas images. Based on the rigid registration results, which indicate a degree of similarity for each atlas, an atlas image is selected for deformable registration where the rigid registration result is used as initialization. Regions of interest are segmented in the patient image using the results of the deformable registration.

Abstract: A method that receives a visual media item and determines an identity of at least one intended viewer of the visual media item is disclosed. The method may further identify a visual representation of an object that is comprised by the visual media item and determine a relevance of the object based, at least in part, on the identity. The method may further generate a modified visual media item such that the modified visual media item differs from the visual media item, at least, by visual emphasis of the visual representation of the object.

Abstract: The distance to a target vehicle is calculated. To achieve this, a target vehicle traveling ahead is imaged and it is determined to what vehicle group, such as a light-duty vehicle group, standard passenger car group or heavy-duty vehicle group, the image of the target vehicle belongs. Representative vehicle widths are stored in a vehicle group table on a per-vehicle-group basis. The distance from one's own vehicle to a target vehicle is calculated using the vehicle width that corresponds to the vehicle group decided.

Abstract: An image recognition apparatus determines whether an image of a pedestrian is captured in a frame of video data captured by a vehicle mounted camera. A pre-processing unit determines a detection block from within a frame, and cuts out block image data corresponding to the detection block from the frame. Block data with a predetermined size that is smaller than the size of the detection block is created from the block image data. A neuro calculation unit executes neuro calculation on the block data, and calculates an output synapse. A post-processing unit determines whether a pedestrian exists within the detection block on the basis of the output synapse. When a pedestrian is detected, the post-processing unit creates result data, which is obtained by superimposing the detection block within which the pedestrian was detected onto the frame.

Abstract: In an example embodiment, a computer-implemented method is disclosed that determines a depth image; detects an object blob in the depth image; segments the object blob into a set of layers; and compares the set of layers associated with the object blob with a set of object models to determine a match. Comparing the set of layers with the set of object models can include determining a likelihood of the object blob as belonging to each of the object models and determining the object blob to match a particular object model based on the likelihood. Further, determining the likelihood of the object blob as belonging to the one of the object models can include computing a recognition score for each layer of the set of layers; and aggregating the recognition score of each layer of the set of layers.

Abstract: A red-eye detection and correction method includes computing a first difference image based on a difference between red pixel intensity and green pixel intensity of a set of pixels. The set of pixels are associated with a first eye region of an image. The method further includes processing the first difference image for computing at least one gradient and at least one projection associated with the at least one gradient. Furthermore, the method includes determining at least one central point based on the at least one projection; and thereafter computing the first eye color defect region based on the at least one central point and a plurality of red pixels of the set of pixels. The method also includes mapping an eye color defect location information computed from a low resolution image to be applicable on the original high resolution image, thereby avoiding recomputation.

Abstract: An information processing system acquires an image captured by an image pickup unit and one or more templates where each template includes one or more fields. The acquired image is then compared to the templates and a result based on the comparison is generated such that the result indicates whether recognition of the fields of the templates was successful.

Abstract: A method of measuring road markings. The steps of which include first projecting a red laser line onto the road including a portion on the road surface and a portion on the road markings. Second, filter out all colors except red and convert to a grayscale image. Thirdly, separate image into one pixel wide columns and apply a filtering so only white pixels having a predetermined minimum brightness remain. Next, calculate the road-paint boundary to determine which columns belong to the road surface and which belong to road marking. Using linear regression, a line equation for points corresponding to the road surface is generated and extrapolated to convert the road marking measurements to a height.

Abstract: An image processing device includes processing circuitry configured to acquire a plurality of basis patterns for converting a first region and a second region in a target image, calculate a contribution ratio of each basis pattern, and select one or more basis patterns from higher contribution ratio to lower contribution ratio until a sum of contribution ratios of the selected one or more basis patterns reaches a specific first threshold. The processing circuitry is further configured to generate a control signal of an intensity that increases with a number of the one or more selected basis patterns, project a third region in the target image to the basis patterns to obtain a projection coefficient, reduce noise of the projection coefficient based on the intensity of the control signal, and reconstruct the third region using the projection coefficient after the noise is reduced.

Abstract: An apparatus includes: an input/output interface configured to have a reference surface model and a floating surface model inputted thereto; a memory having instructions for registration of the reference surface model and the floating surface model stored therein; and a processor configured for registration of the reference surface model and the floating surface model according to the instructions.

Abstract: An image encoding device includes an intra predictor for, when a coding mode corresponding to one of coding blocks into which an inputted image is divided is an intra coding mode, carrying out an intra-frame prediction process on each block which is a unit for prediction of the coding block to generate a prediction image; and an encoder for encoding coding mode information and an intra prediction parameter indicating an average prediction. When the intra predictor carries out the average prediction, the intra predictor carries out a filtering process only on a target pixel of the intra prediction located at an upper end or a left end of the bock, the filtering process using an intermediate prediction value, which is an average value of adjacent pixels of the block, and at least one adjacent pixel of the target pixel.

Abstract: A device for automatically determining the contours of heights of a relief includes an image receiving unit configured to receive an image of a predetermined geographical area. The image includes altimetric data. A background surface elimination unit eliminates the background surface of the relief. The background surface illustrates gradual variations in one or more altitudes of the geographical area. A top-hat algorithm having an adjustable structuring element can be used. A thresholding unit generates a binary image which only contains those reliefs having heights above a threshold value. An extracting unit extracts rough contours of heights of the relief from the binary image. A simplifying unit obtains a set of polygons by simplifying the rough contours. The polygons illustrate contours of the heights of the relief of the geographical area.

Abstract: The invention addresses the problem that traditional Total Variation only enforces the horizontal and vertical intensity continuity and thus fails to reconstruct oblique edges well. In an embodiment, Directional Total Variation is defined which supports multiple gradient directions. It first pre-processes the image and determines the direction of edges/texture; Directional TV is calculated based on the gradients along the determined direction and its orthogonal direction. By utilizing adaptive weights to different directions in the regularization, Directional TV is capable of preserving edges independent from their orientation nicely. Thus, Directional TV regularization based image denoising, compression or super resolution tend to obtain better quality.

Abstract: A computer-implemented method for calculating a multi-dimensional wavelet transform in an image processing system comprising a plurality of computation units includes receiving multi-dimensional image data. An overlap value corresponding to a number of non-zero filter coefficients associated with the multi-dimensional wavelet transform is identified. Then the multi-dimensional image data is divided into a plurality of multi-dimensional arrays, wherein the multi-dimensional arrays overlap in each dimension by a number of pixels equal to the overlap value. A multi-dimensional wavelet transform is calculated for each multi-dimensional array, in parallel, across the plurality of computation units.

Abstract: A method of large-radius edge-preserving low-pass filtering is performed on a digital signal having data points. The method includes computing a weighted average of a signal layer at vertices spaced a distance apart such that an amount of the vertices is less than an amount of the data points and producing, for each of the data points, a large-radius edge-preserving low-pass filtered signal based on the weighted average of the signal layer at vertices neighboring the data point.

Abstract: A system and method are described for enhancing readability of scanned document images by operating on each document individually. Via principle component analysis, edge detection and local color contrast computation, an automated method removes image background noise and improves sharpness of the scripts and characters.

Abstract: Optical system for reading structurally distinct microbeads. The microbeads are encoded and provide output light signals onto a Fourier plane when illuminated by an incident light. The system includes an input light source configured to illuminate the structurally distinct microbeads thereby providing the output light signals. The output light signals are configured to be projected onto the Fourier plane in a readable manner. The system also includes a reading device that is positioned to detect the output light signals from the Fourier plane. The input light source and the reading device are configured to automatically illuminate the microbeads and detect the output light signals in a predetermined manner. The system also includes a processor configured to perform Fourier plane analysis of the output light signals to determine corresponding codes of the microbeads.

Abstract: Disclosed herein are methods and apparatus for obtaining at least one non-birefringence image and at least one birefringence image of a stained sample, and classifying regions of the stained sample into a plurality of classes based on the at least one non-birefringence image and the at least one birefringence image.

Abstract: An image processing device, method, computer readable storage medium, and image forming device that calculate spatial frequency of input image data and determine a frequency component in which a distribution amount of the spatial frequency maximizes, and perform an enhancement process by applying, to the input image data, an enhancement amount that varies according to the distribution amount of the spatial frequency, based on the determined frequency component.