Abstract: The present invention relates to a convolutional artificial neural network-based image and video recognition system, and a method therefor. The recognition system comprises: a mobile device for performing lower layer analysis, transmitting a user question to a server, receiving an answer thereto from the server, and managing the same; the server connected to the mobile device over a network so as to perform data processing in respective neural network layers corresponding to middle and upper layers of a convolutional artificial neural network, and register, analyze, search for, and classify a particular object (image) and video; and a retriever for comparing an FC layer value for artificial neural network processing of an image and video transmitted by the mobile device with an FC layer value for artificial neural network processing of an image and video registered by a content provider.

Abstract: A capsulorhexis device is inserted into an incision site of a cornea to make an incision in an anterior capsule surrounding a crystalline lens. The capsulorhexis device includes a loop having elasticity and conductivity; a moving member having one end fixed and coupled to the loop; an insertion guide configured so that, while the incision is being made in the crystalline lens capsule, a front end thereof is inserted into the incision site of the cornea; and a housing having one end coupled to a rear end of the insertion guide, wherein the loop is housed in the housing and, to make the incision in the crystalline lens capsule, slides in the housing together with the moving member to pass through the insertion guide and be deployed into an anterior chamber of the eye.

Abstract: An apparatus and a method for providing a streaming video by an application program are provided. A method for providing a streaming video by an application program includes a step of selecting two or more cloud game streaming videos according to a preset condition from amon11lity of cloud game streaming videos output by executing a plurality of cloud game application programs from among a plurality of cloud games, a step of generating one integrated streaming video by combining the selected two or more cloud game streaming videos such that the two or more cloud game streaming videos are output on one screen, and a step of extracting cloud game videos in respectively different progress situations corresponding to the cloud game requested to be executed by two or more user terminals from among the integrated streaming videos and providing the cloud game video to a corresponding user terminal.

Abstract: An image display apparatus and a color distortion compensation method are provided. The image display apparatus includes a detecting part which detects whether there is a color distortion in a block, using decoding information acquired from an incoming image and a pixel value of the incoming image, and a compensation part which compensates for the color distortion of the block when it is determined that the color distortion exists. Accordingly, it is possible to detect a color distortion in the block using the decoding information and pixel value acquired from the incoming image, and compensate for the color distortion of the block by performing pixel value adjustment and low pass filtering (LPF) when it is determined that the color distortion exists.

Abstract: Provided are an image processing method and apparatus for enhancing contrast. The image processing apparatus includes a determination unit that determines whether contrast enhancement processing should be performed on an input image; an intensity mapping unit that generates a plurality of images having different exposure times from the input image, using intensity mapping, if the contrast enhancement processing should be performed on the input image; and an image composition unit that composes the plurality of images into a composed image. Therefore, it is possible to reduce a color change or a false contour phenomenon that are generated in a related art contrast enhancement process.

Abstract: A method of improving picture quality in a composite video burst signal includes dividing the composite video burst signal into a plurality of frequency bands using a low pass filter and a high pass filter, performing wavelet packet filtering of frequency bands including a chrominance signal having energy higher than a specified threshold among the plurality of frequency bands, and performing Wiener filtering of frequency bands including a chrominance signal having energy lower than a specified threshold.

Abstract: A field sequential display apparatus and an image display method thereof are provided. A field sequential display apparatus includes: a color-coordinate conversion unit which analyses image state information of a plurality of input image signals of primary colors representing one image and converts the input image signals of primary colors into image signals of primary colors and at least one image signal of specific colors by using the image state information; a display panel displaying the converted image signals; and a light source driving unit which sequentially drives light sources corresponding to colors of the converted image signals. Accordingly, color breakup can be prevented, and image quality can be improved.

Abstract: A motion adaptive image processing apparatus includes a classifier to classify a current field of a sequence of in put fields into one or more class regions, a calculator to calculate a variance based on pixel values of pixels located in a predetermined region around a certain pixel of the current field, a threshold calculator to calculate a maximum variance and a minimum variance which are pre-set according to the one or more class regions, the threshold calculator calculating a threshold based on the calculated maximum variance and the minimum variance, a motion calculator to calculate a motion change of an image using a previous field and a next field of the current field in the sequence of the input fields, and a weight calculator to calculate the weight to be applied to the certain pixel based on the calculated threshold and the calculated motion change of the image.

Abstract: An image conversion apparatus to perform a motion compensation and a motion compensation method. The image conversion apparatus includes a first buffer, a second buffer, and a third buffer that store sequentially incoming video fields in sequence, a motion estimator that estimates a motion vector between a first field stored in the first buffer and a third field stored in the third buffer, a motion compensator that compensates for a motion between the first field and the third field using the motion vector output from the motion estimator and thereby outputs an interpolated field, an output unit that outputs at least one of the interpolated fields output from the motion compensator and a second field stored in the second buffer, and a field selector that selects a field to be output from the output unit according to a film mode of the sequentially incoming video fields.

Abstract: An image processing device and method to select a field to use in interpolation, if an input image is a film image. The image processing device includes a field selector to select a field to use to interpolate a current field based on difference values in field data among first, second, and third fields being consecutively input in an input image, the field to use to interpolate the current image being one of the first and the third fields and the current field being the second field, a film detector to generate a pattern based on the difference values in the field data between the first, second, and third fields and to determine whether the input image is a film image based on the generated pattern, and an interpolator to interpolate the second field using the one of the first and third field selected by the field selector if the film detector determines the input image to be a film image.

Abstract: An image conversion device and method of interpolating an input image by utilizing a motion estimation that is adaptive according to characteristics of the input image.

Abstract: An image processing apparatus using a judder-map and a method thereof. The image processing apparatus using a judder-map includes: a field storage block to receive an input interlaced video signal including an input image having a plurality of consecutive fields, a film detecting block to receive the plurality of the consecutive fields from the field storage block and to detect whether the input image is an image in a film mode, a judder-map generating block to detect judder created on a pixel or block of neighboring fields among the plurality of consecutive fields provided from the field storage block to generate a judder-map, and an image interpolating block to perform image interpolation using the judder-map if the input video signal is detected to be an image in the film mode and to generate and output a progressive video signal. The image processing apparatus prevents the creation of judder on CGI (Computer Graphic Imagery) or subtitles, to which pulldown technologies are not applied.

Abstract: An image display apparatus and a color distortion compensation method are provided. The image display apparatus includes a detecting part which detects whether there is a color distortion in a block, using decoding information acquired from an incoming image and a pixel value of the incoming image, and a compensation part which compensates for the color distortion of the block when it is determined that the color distortion exists. Accordingly, it is possible to detect a color distortion in the block using the decoding information and pixel value acquired from the incoming image, and compensate for the color distortion of the block by performing pixel value adjustment and low pass filtering (LPF) when it is determined that the color distortion exists.

Abstract: An apparatus to suppress artifacts in an image signal. The apparatus includes a differential value calculation unit to calculate a differential value between adjacent pixels with respect to an input image signal, a diffusion amount calculation unit to calculate an amount of diffusion between the adjacent pixels on the basis of the differential value calculated by the differential value calculation unit, and a pixel value conversion unit to convert the present pixel value of the image signal inputted on the basis of the diffusion amount between the pixels calculated by the diffusion amount calculation unit. The apparatus changes the pixel value in consideration of the differential value between the adjacent pixels, and thus it can provide a high quality image signal without losing or misrecognizing the artifacts of the image signal.

Abstract: A method and an apparatus for motion vector estimation using motion vectors of neighboring image blocks are provided. The method includes dividing a current image block into a plurality of sub-blocks; and estimating motion vectors of the sub-blocks using the motion vectors and pixel weights of the neighboring image blocks adjacent to the sub-blocks.

Abstract: An apparatus for and method of estimating a motion vector for a video image block are provided. The apparatus includes an error calculation unit calculating a motion estimation error for a predetermined prediction motion vector of the image block, an information extraction unit extracting geometric information regarding the image block, an update motion vector generation unit generating an update motion vector based on the motion estimation error calculated by the error calculation unit and the geometric information extracted by the information extraction unit, and an addition unit adding the prediction motion vector to the generated update motion vector.

Abstract: A contrast compensation apparatus and a method thereof utilize a simplified hardware structure and a low contrast distortion. The contrast compensation apparatus includes a pixel value detection unit to detect a distribution of pixel values of respective pixels of an input image signal, a pixel value limit unit having pre-set luminance limit values and re-configuring the distribution of the pixel values of the respective pixels based on the pre-set luminance limit values, and a mapping unit to set luminance for the respective pixels based on a cumulative distribution function with respect to the re-configured pixel values. When certain portions of an image are rendered very bright or dark, the phenomenon in which the entire image becomes abruptly bright or dark is reduced. Further, the contrast compensation apparatus uses the minimum number of multipliers, adders, and subtractors only, thus having an advantage of simplified structure and using less electric power.

Abstract: A de-interlacing apparatus with a noise reduction/removal device. The noise reduction/removal device can include a motion prediction unit that predicts motion vectors between an image one period ahead of a previous image and a current image with respect to individual images which are sequentially inputted; a motion checking unit that applies the motion vectors predicted by the motion prediction unit to the image one period ahead of the previous image and two different images ahead of the current image in time, and checks whether the motion vectors are precise motion vectors; a motion compensation unit that compensates for motions in use of the motion vectors checked for preciseness thereof by the motion checking unit; and a noise removal unit that removes noise on images using the motion-compensated images by the motion compensation unit and the inputted images. Accordingly, the noise reduction/removal device can reduce or remove noise through simple procedures on noise-bearing images.

Abstract: A deinterlacing apparatus and method thereof include a motion compensation unit implementing motion-compensated temporal interpolation for each of estimated motion vectors with reference to a previous field and a next field, which are respectively ahead of and behind a current field to be interpolated, producing interpolation values of a pixel to be interpolated, and outputting a selected value from the interpolation values as a first interpolation value. Further, a spatial interpolation unit producing a second interpolation value of the pixel to be interpolated using values of pixels around the pixel to be interpolated, and an output unit mixing the first and second interpolation.

Abstract: An image resolution conversion method and apparatus based on a projection onto convex sets (POCS) method are provided. The image resolution conversion method comprises detecting an edge region and a direction of the edge region in an input low-resolution image frame in order to generate an edge map and edge direction information, generating a directional point spread function based on the edge map and the edge direction information, interpolating the input low-resolution image frame into a high-resolution image frame, generating a residual term based on the input low-resolution image frame, the high-resolution image frame, and the directional point spread function, and renewing the high-resolution image frame according to a result of comparing the residual term with a threshold.