Abstract: In a method for virtual adaptation of an object to be medically inserted in a patient, to a body part of the patient, the object is shown in a desired view on a monitor, the body part is shown in 3D on the monitor, the body part in its unchanged 3D-imaging context, is automatically displaced relative to the object until the object is adapted to the body part with the orientation of the object relative to the monitor remaining unchanged.

Abstract: An electrical-device-implemented image coding method includes: a raw block is obtained from an image. There are a max pixel value and a min pixel value among the raw pixel values of the image. A first difference sum, which is sum of differences between the raw pixel values and the max pixel value respectively, are calculated. A second difference sum, which is sum of differences between the raw pixel values and the min pixel value respectively, are calculated. If the first difference sum is not smaller than the second difference sum, each of the residue values is calculated by subtracting the min pixel value from the corresponding raw pixel value. Otherwise, each of the residue values is calculated by subtracting the corresponding raw pixel value from the max pixel value. Each of the residue values is encoded into a compressed codeword. A bit stream is generated according to the compressed codewords.

Abstract: As set forth herein, a computer-based method is employed to align a sequences of images. Metadata associated with images from two or more sources is received and a time stamp is extracted from the metadata. The images are sorted into sequences based at least in part upon the image source. The similarity of images from disparate sequences is measured and image pairs from disparate sequences with a similarity greater than a predetermined threshold are identified. A sequence of images is aligned by minimizing the misalignment of pairs.

Abstract: In a biometrics authentication system inputting biometric information and outputting a user ID, speedup of 1:N matching on the side of server is achieved when cancelable biometrics authentication is realized in which matching is carried out. During registration, similarity of a feature to a dummy feature is searched and the feature is transformed by using a transform parameter corresponding to a nearest neighboring dummy feature. The transformed feature and a group ID are transmitted to the server and the server registers the transformed feature at the group in a DB. During authentication, like the registration process, similarity search of a feature is carried out and the feature is transformed by using transform parameter corresponding to a nearest neighboring dummy feature. The transformed feature and group ID are transmitted to the server. The server executes 1:1 matching between a template a the group in the DB and the transformed feature.

Abstract: A system and method for performing integral histogram convolution for filtering image data is disclosed. The method may include applying a filter window to a first portion of an image, wherein the filter window includes an interior region and a border region. The method may include generating a plurality of histograms for the pixels in the filter window. The method may include generating spatial weight coefficients for the pixels in the border of the filter window. The method may include generating a plurality of color weight coefficients for the pixels in the filter window. The method may include performing a filtering operation on the pixels in the filter window by applying a respective spatial weight coefficient and a respective color weight coefficient to the values in the plurality of histograms for each respective pixel in the filter window. The methods may be implemented by program instructions executing in parallel on CPU(s) or GPUs.

Abstract: An object detecting apparatus and method includes a pixel state determining unit that derives variance value for temporal properties of pixel characteristics of an input image, background model generating unit that adaptively generates a background model from characteristics in the characteristic storing unit and characteristic storing unit for background model generation using the characteristic distance and the pixel state determined as conditions, and an object judging unit that judges an object based on a characteristic distance indicative of a degree of similarity between a generated background model and pixel characteristics of an input image.

Abstract: A vector interpolator optimizes the conversion of an interlaced signal to a non-interlaced signal. The vector interpolator improves the visual clarity of slanted features in a displayed image by adjusting the luminance value of each pixel such that the appearance of “steps” or “jaggies” in the features is reduced. For each pixel, the vector interpolator determines a similarity measure for the pixels within a predetermined area around the pixel. From the similarity measure, an angle for interpolation is selected. The luminance value is then interpolated along the selected vector corresponding to the angle and applied to the pixel. One or more ambiguity measures such as a local edge count ambiguity measure may also be computed to indicate the reliability of the computed luminance value.

Abstract: The invention discloses a face recognition method that reconstructs a 3D face model from a single face image, synthesizes a set of face images under different conditions (such as pose, light . . . ) via the 3D face model, feeds the set of face images under different conditions to the face recognition classifier for training, and making intermediate decisions whether to-be identified individual from a series of video frames is a legal system user by the face recognition classifier. Moreover, the method not only recognizes legal system users, but also rejects imposters, a function inspired by the idea of LLE. Finally, better reliability can be achieved by fusing temporal intermediate decisions.

Abstract: Provided are a method and system for processing a low-illuminance image. The system includes an image acquisition unit to acquire two images having different levels of illuminance and sensitivity as first and second images; a motion vector estimation unit to extract a motion vector of the second image based on the first image; an image correction unit to correct the second image using the extracted motion vector; a synthesis coefficient calculation unit to calculate a synthesis coefficient used to synthesize the first image and the corrected second image; and an image synthesis unit to synthesize the first image and the corrected second image using the calculated synthesis coefficient.

Abstract: Disclosed are methods, devices, and computer program products for red-eye detection in a digital image. In one example embodiment, a method for detecting a red-eye effect in a digital image includes several acts. First, red pixels having a predetermined degree of redness are identified in the image. Next, redness contrast is detected with respect to each of the red pixels and redness is then enhanced for those red pixels having a predetermined level of redness contrast. The pixels identified as being red are then further refined by applying another redness threshold based on one or more color characteristics associated with the red pixels. The refined set of red pixels may then be partitioned into a set of one or more candidate red-eye objects. A candidate red-eye object may be removed as a false positive based on geometric constraints associated with red-eye objects and/or proximity of the object to pixels with human skin-like color tones.

Abstract: Disclosed are methods, devices, and computer program products for red-eye detection in an image. In one example embodiment, a method for detecting red-eye objects in an image includes several acts. First, a set of candidate red-eye objects identified in the image is received. Then, features are extracted from the candidate red-eye objects and, with a plurality of classifiers, a false red-eye object is eliminated from the set of candidate red-eye objects based on the extracted features. First and second ones of the plurality of classifiers are optimized for classifying objects in a first range of sizes using first and second ones of the extracted features, respectively. Furthermore, third and fourth ones of the plurality of classifiers are also optimized for classifying objects using the first and second ones of the extracted features, respectively, but for objects in a second range of sizes.

Abstract: An image processing device and a method thereof are provided. In the image processing method, an adjusting value is computed according to a pixel of a current image and a plurality of reference pixels nearby. In addition, a weighted coefficient is computed according to a gray-scale difference between the pixel of the current image and a pixel of a previous image. Further, a weighted adjusting value is computed according to the adjusting value and the weighted coefficient. Besides, the pixel of the current image is adjusted according to the weighted adjusting value to generate a pixel of an output image. Thereby, the current image is adjusted adaptively to increase the quality of the output image.

Abstract: To improve modification result where areas targeted for modification overlap, an image processing apparatus is provided. The image processing apparatus has a modification area establishing unit capable to establish a plurality of modification areas on a target image by analyzing the target image; an overlap detecting unit that detects overlapping of the plurality of modification areas; and a modification processing unit that modifies an image within a partial area including an overlapping portion of the plurality of modification areas in a reduced modification level in the case that the overlapping is detected by the overlap detecting unit, the reduced modification level being smaller than a modification level used in the case that the overlapping is not detected, the partial area belonging to at least one of the plurality of modification areas.

Abstract: When determining an area in which color blur occurs from an image, there may be an area where color blur suppression is performed in an area even though color blur does not occur, and the area becomes more noticeable or less noticeable depending on a degree of gamma correction processing. To solve this problem, an image processing apparatus obtains a color blur suppression coefficient based on a gradient value obtained from the image, performs color blur suppression on the image data using the color blur suppression coefficient, and changes the color blur suppression coefficient corresponding to the gradient value according to brightness.

Abstract: A configuration for the optical detection of a specimen, wherein the specimen or at least part of the specimen is scanned by means of linear illumination by scanning means, means for linear beam shaping of the illuminating light are provided, and the illuminating light has a preferably periodic structure in at least one spatial direction in that means for generating the structure are disposed in the illuminating beam path, light coming from the specimen is detected and images of the specimen are generated therefrom, at least one optical sectional image through the specimen and/or one image with increased resolution is/are calculated from the images, and means for generating the structure are disposed downstream of the scanning means in the direction of the illumination.

Abstract: When N kinds of foreground identifiers for identifying color information of a foreground of a color image is to be reduced to M smaller than N, foreground identifiers corresponding to similar chromatic colors are merged and foreground identifiers corresponding to similar achromatic colors are merged but a foreground identifier corresponding to a chromatic color and a foreground identifier corresponding to an achromatic color are not merged. Thus, image quality degradation derived from color change from an achromatic color to a chromatic color or vice versa may be suppressed. On the basis of a foreground layer including the M kinds of foreground identifiers, M binary images respectively corresponding to the M kinds of foreground identifiers are generated, each of the binary images is subjected to lossless compression and a background layer is subjected to lossy compression.

Abstract: In one embodiment, a method comprises determining an initial set of feature correspondences for a pair of images of a three dimensional (3D) space; identifying one or more first regions in one of the pair of images and one or more second regions in the other one of the pair of images. The method comprises identifying additional feature correspondences by matching first regions and second regions in the pair of images and generating a motion of a camera in the 3D space responsive to the initial set of feature correspondences and the additional feature correspondences, wherein the camera captured the images. Each of the first regions and the second regions potentially maps to a plane in the 3D space corresponding to the images. Additionally, each of the first regions and the second regions includes at least one feature in the initial set of feature correspondences.

Abstract: An image processing apparatus, includes an input unit that inputs a plurality of input images. The image processing apparatus also includes an acquisition unit that acquires photographic information added to each of the plurality of input images. The image processing apparatus further includes a first setting unit that sets control information used to control a layout of each of the plurality of input images in accordance with whether or not the photographic information indicates that the respective input image is photographed with emphasis on an object. The image processing apparatus further includes a generator that generates an output image on which each of the plurality of input images is configured to have an image size according to the control information of the respective input image.

Abstract: A plurality of images are input, specific regions are set for each of the inputted plurality of images, a typical size of the specific regions of each of the plurality of images is determined, and the ratio of the typical size to the size of the entire image is calculated as a typical size ratio. A typical size ratio used as a reference is selected from the typical size ratios calculated for the respective images as a reference ratio, and a candidate region upon trimming that image is determined based on the specific regions of each of the plurality of images. A trimming region for each of the plurality of images is determined based on the reference ratio, and the typical size ratio and candidate region of each image, and each of the plurality of images is trimmed according to the trimming region.

Abstract: An image processing device which is capable of efficiently performing face-detection, red-eye detection, and/or image data correction including red-eye correction without waste when there are a plurality of chances of performing the image data correction including the red-eye correction on a single image. A printing apparatus acquires image data and information concerning a face detection process performed on the image data from an image pickup apparatus. A controller of the printing apparatus determines to cause the printing apparatus to perform a face detection process on the acquired image data, based on the acquired information. The information contains an information item concerning detection accuracy of the performed face detection process, and a face detection process with higher detection accuracy than the detection accuracy described in the information is determined as the face detection process to be performed by the printing apparatus.