Abstract: An image sensor array of active pixel elements is arranged in rows and columns. Each column has an output circuit for reading out pixel image signals. The output circuit includes a pair of sample capacitors, a switching circuit operable based upon pixel switches for applying pixel voltages to the pair of sample capacitors. At least one optically masked pixel is provided so that output image signals obtained therefrom represent substantially only the column fixed pattern noise (FPN). An image processing circuit records the column FPN for each column from the optically masked pixels, records the image signal from the sensor array of active pixels, and subtracts the column FPN column-wise from the image signal.

Abstract: A color image sensor has a first, second, and third color component pixels. The third color component pixel includes more intensity information than the first and second color component pixels. A difference estimating section estimates difference in a signal level exerted on a pixel signal of the third color component pixel, using the pixel signal of a pixel peripheral to the third color component pixel. A correcting section corrects the pixel signal of the third color component pixel on the basis of the estimated difference. Accordingly, when the third color component is green in a Bayer array, a pixel signal of a green pixel in a row of red pixels becomes equal to the pixel signal of a green pixel in a row of blue pixels. Therefore, line crawling can be suppressed without smoothing an image.

Abstract: Image-processing device for ROP is disclosed. Upon receiving a ROP command, the image-processing device determines whether the size of the brush data is required to be modified through a size modifying operation, based on the printing resolution of the printing device. If the size modifying operation is necessary, the image-processing device performs a process to enlarge or reduce the brush data, depending on the printing resolution. The image-processing device converts the ROP code associated with the operational expression using brush data to a combination of ROP codes associated with operational expressions using the source bitmap, but not brush data. The image-processing device controls a ROP processor to perform the ROP processes corresponding to these ROP codes using the enlarged or reduced brush data as the source bitmap in order to implement a ROP process equivalent to the ROP code inputted with the ROP command.

Abstract: A method obtains media on a device, provides identification of an object in the media via image/video recognition and audio recognition, and displays on the device identification information based on the identified media object.

Abstract: A method and apparatus for translating color to grayscale images, wherein at least some pixels of the color image having the same color are mapped to different grays in the grayscale image depending on the spatial surround of each pixel. Further, the method and apparatus may include applying a high pass filter to at least one chrominance component of a color image to generate at least one high pass filtered chrominance component, and adjusting a luminance component of the color image based upon the at least one high pass filtered chrominance component.

Abstract: An image processing apparatus configured to obtain a resized image of a suitable size in a case where a plurality of image data are enlarged or reduced, while keeping a vertically long or horizontally long state of the image. Thus, a matched state between a resize object image and a resize rectangle as target values for a resized image as to whether being vertically long or horizontally long is determined. Responsive to a result of determination, the resize rectangle is rotated 90 degrees, and the resize object image is enlarged or reduced such that the enlarged or reduced resize object image is matched with the resize rectangle rotated.

Abstract: A method and system for processing image data to identify objects in an image. The method and system operate using various resolutions of the image to identify the objects. Information obtained while processing the image at one resolution is employed when processing the image at another resolution. The method and system identify objects in the image based on the information obtained at the various resolutions of the image.

Abstract: A server (1) analyzes request information from a client (11) by a request analyzer (4) and selects necessary image data from multiple viewpoint image data (2) to be output to an image generator (3). The image data is generated by interpolation and output to an image synthesizer (5), in which a plurality of images data is synthesized in a form suitable for encoding. An encoder (6) encodes the image data and transfers it to a network (7). The client (11) receives the encoded image data, decodes the data through a decoder (12) and outputs the decoded image data to an image processor (13), where it is converted into a stereoscopic display format and displayed on a display unit (14). The client (11) includes an input unit (16) for changing the viewpoint, and transmits the request information of viewpoint alternation to the network (7) by a request output unit (15).

Abstract: An apparatus, system, and method for generating an image are disclosed. A processor may generate a first output image based on a plurality of input images and remove an artefact, if any, from the first output image to generate a second output image. For example, in an embodiment, the processor may calculate a contribution of the artefact to image intensity values and subtract the calculated contribution from the image intensity values. In another embodiment, the processor may delete a predetermined portion of a transform image representing transform data obtained by applying an image transform to the first output image, thereby modifying the transform data, and may generate a non-transform image based on the modified transform data.

Abstract: A computer implemented method constructs a classifier for classifying test data. High-level features are generated from low-level features extracted from training data. The high level features are positive definite matrices in a form of an analytical manifold. A subset of the high-level features is selected. An intrinsic mean matrix is determined from the subset of the selected high-level features. Each high-level feature is mapped to a feature vector onto a tangent space of the analytical manifold using the intrinsic mean matrix. Then, an untrained classifier model can be trained with the feature vectors to obtain a trained classifier. Subsequently, the trained classifier can classify unknown test data.

Abstract: Methods, system apparatus and devices for classification and recognition that is based on principal component analysis and is implemented in the transform domain using the fast two-dimensional PCA to processes the signal in the transform domain. The signal is represented with a reduced number of coefficients, therefore reducing the storage requirements and computational complexity while yielding high recognition accuracy.

Abstract: An exemplary method includes receiving an image data set that comprises a multidimensional property space and data classifiable into data classes, determining a projection vector for data of the data set wherein the projection vector maximizes a ratio of between-class scatter to within-class scatter, selecting a reference for the vector, projecting at least some of the data onto the vector, measuring distances from the reference to at least some of the data, classifying at least some of the data into data classes based on a nesting analysis of the distances, eliminating the classified data from the image data set to produce a modified image data set and deciding whether to determine another projection vector for the modified image data set. Various other exemplary methods, devices, systems, etc., are also disclosed.

Abstract: A method for constructing a database having digital video data information representing a plurality of video sequence is disclosed. The method includes the steps of: a) partitioning each image frame of each video sequence into L number of sub images; b) generating L number of edge histograms for each image frame; c) normalizing the edge histogram bins to thereby generate M number of normalized edge histogram bins; d) calculating M representative edge histogram bins in order to generate L number of representative edge histograms based on the normalized edge histogram bins; and e) non-linearly quantizing the representative edge histogram bins to generate M number of quantization index values for the each representative edge histogram, to be stored in the database.

Abstract: A system and method for imaging a three-dimensional scene having one or more objects. The system includes a light source, a detector array, a timing circuit, an inertial guidance system and a processor connected to the timing circuit and the inertial guidance system. The light source generates an optical pulse and projects the optical pulse on an object so that it is reflected as a reflected pulse. The detector array includes a plurality of detectors, wherein the detectors are oriented to receive the reflected pulse. The timing circuit determines when the reflected pulse reached detectors on the detector array. The inertial guidance system measures angular velocity and acceleration. The processor forms a composite image of the three-dimensional scene as a function of camera position and range to objects in the three-dimensional scene.

Abstract: Image data is efficiently encoded using a predict coding unit and a run-length coding unit. The predict coding unit encodes a target pixel X on the basis of difference between the value of the target pixel and a predict value calculated from pixels neighboring the target pixel. The run-length coding unit starts the measuring the run when the number of colors contained in four pixels “a”, “b”, “c”, and “d” near the target pixel X is 1, and outputs encoded data of the run when the target pixel is different from an immediately preceding pixel “a”. Then, the predict coding unit starts the encoding. At this time, since the target pixel is different from the preceding pixel, the preceding pixel is excluded from references for generating the predict value. Instead of the preceding pixel, an pixel, which has been encoded, satisfying a specific condition is referred to.

Abstract: This invention provides a dynamic image adjusting apparatus for dynamically adjusting the contrast of a first image signal which includes a plurality of pixels. The adjusting apparatus includes a first transformation module, an operation module, and a second transformation module. The first transformation module receives the first image signal for generating a plurality of adjusting signals, wherein each pixel of the first image signal corresponds to one of the plurality of adjusting signals. The operation module is coupled to the first transformation module and receives a first set of adjusting signals of the plurality of adjusting signals to generate a first gain curve. The second transformation module is coupled to the operation module and the first transformation module, and generates a second image signal according to the first gain curve and the plurality of adjusting signals.

Abstract: A method for presenting recognized handwritten symbols includes recognizing a detected handwritten pattern. Recognizing may include comparing the handwritten pattern to templates representing ways of writing symbols and returning a best interpretation of the handwritten pattern. The best interpretation may be based on the pattern of a best template. At least two templates may include different patterns which represent different ways of writing a single symbol, and the patterns of the at least two templates may return different best interpretations when being most similar to the handwritten pattern. The method further includes presenting the pattern of the best template on a screen. A device for implementing the method may also be provided.

Abstract: The present invention is methods for processing medical images so as to remove certain effects of the physical characteristics of the object being imaged and/or of the apparatus used to form the images. The invention further provides for the formation of a standardized image from the processed image and for the use of the standardized image or the processed image in the training of computer-aided detection/diagnosis algorithms. These algorithms may then be used to detect abnormalities in other standardized or processed images derived from any of a variety of image acquisition systems.

Abstract: A method includes (a) performing image deblur at the original size of underexposed and normally exposed images when their original image size is small. Otherwise, the method includes (b) downsizing the images and performing image deblur on the downsized images. The image deblur on downsized images includes (c) applying a global color transfer between the images if their difference of average gray values is small. Otherwise, the method includes four successive sub-procedures: (d) histogram customizing of the underexposed image based on the normally exposed image; (e) region segmentation for the histogram-customized underexposed image and local alignment of regions in the histogram-customized underexposed image with the normally exposed image; (f) color tuning for the histogram-customized underexposed image based on the local alignment and a global alignment between the underexposed and the normally exposed images; and (g) local color transfer with multi-size neighbor windows from between the images.

Abstract: A method and system for processing image data to identify objects in an image. The method and system operate using various resolutions of the image to identify the objects. Information obtained while processing the image at one resolution is employed when processing the image at another resolution.