Abstract: An image processing apparatus includes a converter converting an interlace image including a first number of pixels into a first progressive image, an interpolator interpolating the first progressive image to generate a second progressive image including a second number of pixels, a classification unit classifying, in accordance with a feature of the second progressive image, into classes, subject pixels forming a third progressive image, which serves as a target image, including the second number of pixels and having a quality higher than the second progressive image, a storage unit storing a prediction coefficient for each of the classes obtained by conducting learning using a plurality of progressive images, each including the second number of pixels, and a computation unit performing computation using the second progressive image and the prediction coefficient for each of the classes to determine the third progressive image from the second progressive image.

Abstract: An image processing apparatus includes an interlace-progressive converter converting an interlace input image into a progressive intermediate image, a motion-vector detector detecting motion vectors of the input image by using a distance shorter than the pixel interval of the intermediate image, a cyclic-coefficient setting unit setting, on the basis of a vertical motion, a first cyclic coefficient for a first type of pixel and a second cyclic coefficient for a second type of pixel, a motion compensator motion-compensating, on the basis of the motion vectors, a past progressive output image to generate a motion-compensated image, and an output image generator generating a progressive output image by adding pixel values of the first type of pixels and the second type of pixels of the progressive intermediate image and the motion-compensated image by using the first cyclic coefficient and the second cyclic coefficient as weights.

Abstract: A class configuration generation unit generates (n?1) number of class configurations each of which is comprised of i number of the already selected features plus a feature selected from the remaining (n?i) number of the features (both of n and i are integers). A class configuration selection unit selects an optimal class configuration from the (n?i) number of the class configurations using an arbitrary evaluation value. The features used in the class configuration selected by the selection unit are used as the already selected features in the generation unit. The operations by the generation unit and the selection unit are repeated with values of i sequentially varying from 0 to r?1, thereby generating a class configuration comprised of r number of the features.

Abstract: An image processing apparatus and method, and a program-recorded recording medium perform detecting features of image information for classes which is input for learning, calculating the degrees of belongingness of the detected features to the classes, weighting pieces of the input image information based on the degrees of belongingness, and calculating prediction coefficients for the classes based on the weighted pieces of the image information.

Abstract: This invention relates to an apparatus for processing an information signal etc. that, when converting, for example, SD signal into HD signal, enables well to be obtained pixel data of HD signal no matter whether the dynamic range DR is large or small. DR in a class tap is detected. If DR?Th, items of pixel data y1-a?y4-a calculated by using item of coefficient data Wi-a corresponding to a class code Ca are estimated as items of pixel data of HD signal. If DR<Th, an addition mean value of items of pixel data y1-a?y4-a, y1-b?y4-b calculated by using items of coefficient data Wi-a, Wi-b corresponding to class codes Ca, Cb is estimated as item of the pixel data of HD signal. The items of coefficient data Wi-a, Wi-b are obtained by learning between a student signal corresponding to the SD signal and a teacher signal corresponding to the HD signal by using a portion of the DR having a value thereof that is not less than the threshold value Th.

Abstract: A coefficient-data generating apparatus generates coefficient data of an estimating equation used to convert a first information signal including pieces of information data into a second information signal including pieces of information data. The coefficient-data generating apparatus includes a determining unit determining whether a subject point in a second learning signal corresponding to the second information signal is a feature point having a predetermined feature; a data selector selecting pieces of information data at adjacent points of the subject point in the second learning signal, based on a first learning signal corresponding to the first information signal; and a calculator calculating the coefficient data using information data at the subject point in the second learning signal, the subject point being determined as the feature point by the determining unit, and the pieces of information data selected in association with the subject point by the data selector.

Abstract: Standard Definition signals are divided into blocks at a tap extracting unit, and pixel data contained in each block is extracted as a class tap. A class classification unit obtains class code based on the pixel data contained in the class tap. An auxiliary data generating unit generates auxiliary data regarding conversion into High Definition signals, based on the class tap extracted by the tap extracting unit. A data generation processing unit performs processing based on the class code and the auxiliary data, thereby yielding excellent High Definition signals.

Abstract: A signal processing apparatus includes a generator operable to generate a second image signal by converting a first image signal into the second image signal; a calculation unit operable to calculate a correction amount based on an evaluation of the second image signal relative to the first image signal; and a correction unit operable to correct the second image signal based on the correction amount.

Abstract: A pattern matching section compares reference taps supplied by one area extracting section with a reference tap component of a training pair pre-stored in a training-pair storing section, selects a training pair including a reference tap component having highest similarity to the reference taps, and supplies the selected training pair to a normal-equation generating section. Training pairs stored in the training-pair storing section include a SD signal and HD signal. A prediction-coefficient determining section obtains prediction coefficients by solving normal equations generated by the normal-equation generating section. A prediction calculating section produces a HD signal by applying the prediction coefficients to prediction taps extracted by another area extracting section.

Abstract: A device generates coefficient data of an estimating equation for converting a first information signal to a second information signal. The device performs decimation on a teacher signal to generate a student signal and acquires plural training data items from the teacher signal corresponding to the second information signal and the student signal corresponding to the first information signal. For each training data item, a similarity determination unit acquires the similarity of the student signal with respect to the first information signal corresponding to a second information signal at a position of interest. Coefficient seed data, which are coefficient data of a generation equation including the similarity as a parameter, are determined using the training data items and the similarity. A coefficient data computing unit determines, based on the generation equation, coefficient data of the estimating equation using the coefficient seed data and a similarity value indicating the highest similarity.

Abstract: The present invention relates to a signal processing apparatus and method, a recording medium, and a program, in which a more preferable class can be easily selected. A region extracting unit 151 extracts a class tap from an input composite video signal, and pattern detecting units 152 to 154 detect patterns for classes by using different predetermined methods. A class-code determining unit 155 determines a class based on information from an evaluation table 156, and a coefficient memory 157 supplies a predictive coefficient corresponding to the class to a predictive computing unit 159. The predictive computing unit 159 performs predictive computing based on a predictive tap supplied from a region extracting unit 158 and the predictive coefficient supplied from the coefficient memory 157, so that a component Y signal is generated and is output therefrom. The present invention can be applied to a television receiver.

Abstract: The present invention provides an apparatus for converting image data, including a block extraction unit extracts a class tap from a composite signal. A pixel-location-mode output unit determines a pixel location mode from the extracted class tap, and outputs it to a coefficient memory. A coefficient calculation unit acquires a seed coefficient from a seed coefficient memory to determine a predictive coefficient based on a transform method selection signal input from a designation unit, and stores the result into the coefficient memory. The coefficient memory supplies a predictive coefficient corresponding to the pixel location mode to a predictive calculation unit. A block extraction unit extracts a predictive tap from the composite signal, and outputs the result to the predictive calculation unit. The predictive calculation unit outputs a component signal or a transformed component signal based on the predictive tap and the predictive coefficient.

Abstract: An image processing apparatus includes a converter converting an interlace image including a first number of pixels into a first progressive image, an interpolator interpolating the first progressive image to generate a second progressive image including a second number of pixels, a classification unit classifying, in accordance with a feature of the second progressive image, into classes, subject pixels forming a third progressive image, which serves as a target image, including the second number of pixels and having a quality higher than the second progressive image, a storage unit storing a prediction coefficient for each of the classes obtained by conducting learning using a plurality of progressive images, each including the second number of pixels, and a computation unit performing computation using the second progressive image and the prediction coefficient for each of the classes to determine the third progressive image from the second progressive image.

Abstract: An image processing apparatus includes an interlace-progressive converter converting an interlace input image into a progressive intermediate image, a motion-vector detector detecting motion vectors of the input image by using a distance shorter than the pixel interval of the intermediate image, a cyclic-coefficient setting unit setting, on the basis of a vertical motion, a first cyclic coefficient for a first type of pixel and a second cyclic coefficient for a second type of pixel, a motion compensator motion-compensating, on the basis of the motion vectors, a past progressive output image to generate a motion-compensated image, and an output image generator generating a progressive output image by adding pixel values of the first type of pixels and the second type of pixels of the progressive intermediate image and the motion-compensated image by using the first cyclic coefficient and the second cyclic coefficient as weights.

Abstract: A class configuration generation unit generates (n?1) number of class configurations each of which is comprised of i number of the already selected features plus a feature selected from the remaining (n?i) number of the features (both of n and i are integers). A class configuration selection unit selects an optimal class configuration from the (n?i) number of the class configurations using an arbitrary evaluation value. The features used in the class configuration selected by the selection unit are used as the already selected features in the generation unit. The operations by the generation unit and the selection unit are repeated with values of i sequentially varying from 0 to r?1, thereby generating a class configuration comprised of r number of the features.

Abstract: A class configuration generation unit generates (n?1) number of class configurations each of which is comprised of i number of the already selected features plus a feature selected from the remaining (n?i) number of the features (both of n and i are integers). A class configuration selection unit selects an optimal class configuration from the (n?i) number of the class configurations using an arbitrary evaluation value. The features used in the class configuration selected by the selection unit are used as the already selected features in the generation unit. The operations by the generation unit and the selection unit are repeated with values of i sequentially varying from 0 to r?1, thereby generating a class configuration comprised of r number of the features.

Abstract: The absolute value of the difference between each of the pairs of pixel values of a target pixel and an upper pixel with respect to the target pixel, a pixel that is left-adjacent to the target pixel and an upper pixel with respect to the left-adjacent pixel, the left-adjacent pixel and a lower pixel with respect to the left-adjacent pixel, a pixel that is right-adjacent to the target pixel and an upper pixel with respect to the right-adjacent pixel, and the right-adjacent pixel and a lower pixel with respect to the right-adjacent pixel is obtained. A “1” is assigned if the absolute value of the difference is greater than or equal to a predetermined threshold value whereas a “0” is assigned if the absolute value of the difference is smaller than the predetermined threshold value, forming a pattern having six bits in total.

Abstract: An image processing apparatus and method, and a program-recorded recording medium perform detecting features of image information for classes which is input for learning, calculating the degrees of belongingness of the detected features to the classes, weighting pieces of the input image information based on the degrees of belongingness, and calculating prediction coefficients for the classes based on the weighted pieces of the image information.

Abstract: A semiconductor device and an image data processing apparatus are capable of easily realizing address control, simplifying a circuit for address control, and easily realizing access to image data while reducing the volumes of pieces of image data of different resolutions in different hierarchies. In the image data processing apparatus, a memory cell array stores pieces of composite data, each containing a piece of pixel data and additional information data, for example, a motion vector associated with the pixel data. A designated pixel data and a motion vector associated with the pixel data are read out and a search region is suitably predicted and set up on the basis of the motion vector, and motion estimation is performed in the search region, for example, by a block matching process.

Abstract: A device generates coefficient data of an estimating equation for converting a first information signal to a second information signal. The device performs decimation on a teacher signal to generate a student signal and acquires plural training data items from the teacher signal corresponding to the second information signal and the student signal corresponding to the first information signal. For each training data item, a similarity determination unit acquires the similarity of the student signal with respect to the first information signal corresponding to a second information signal at a position of interest. Coefficient seed data, which are coefficient data of a generation equation including the similarity as a parameter, are determined using the training data items and the similarity. A coefficient data computing unit determines, based on the generation equation, coefficient data of the estimating equation using the coefficient seed data and a similarity value indicating the highest similarity.